DETAILED ACTION
Notice of Pre-AIA or AIA Status
The present application, filed on or after March 16, 2013, is being examined under the first inventor to file provisions of the AIA .
Status of the Claims
The office action is in response to the claims filed on July 12, 2024, for the application filed on April 15, 2024, which claims the benefit of U.S. Provisional Application No. 63/495,975 filed on April 13, 2023. Claims 1 – 9, 20 – 21, 24, 26 – 28, and 30 – 34 are currently pending and have been examined as discussed below.
Claim Objections
Claims 20 and 34 are objected to because of the following informalities:
the limitation of “The method of claim 1” in line 1 of claim 20 should be replaced with “The method of claim 9”;
the limitation of “a respective subject in a plurality of subjects of the first species” in lines 20 – 21 of claim 34 should be replaced with “a respective test subject in a plurality of test subjects of the first species”;
the limitation of “generating, for each respective training sample fin the first plurality of training samples” in lines 38 – 39 of claim 34 should be replaced with “generating, for each respective training sample in the first plurality of training samples”;
the limitation of “the information about the training subject” in line 60 of claim 34 should be replaced with “the information about the training sample”;
the limitation of “the second corresponding representation of the corresponding plurality of abundance values in the second latent feature space” in lines 67 – 69 of claim 34 should be replaced with “the second corresponding representation of the corresponding plurality of abundance values in the second latent feature space according to the third mapping function”; and
the limitation of “the corresponding representation of the corresponding plurality of abundance values in the second latent feature space” in lines 71 – 73 of claim 34 should be replaced with “the corresponding representation of the corresponding plurality of abundance values in the second latent feature space according to the second mapping function”.
Appropriate correction is required.
Claim Rejections - 35 USC § 112
The following is a quotation of 35 U.S.C. 112(b):
(b) CONCLUSION.—The specification shall conclude with one or more claims particularly pointing out and distinctly claiming the subject matter which the inventor or a joint inventor regards as the invention.
The following is a quotation of 35 U.S.C. 112 (pre-AIA ), second paragraph:
The specification shall conclude with one or more claims particularly pointing out and distinctly claiming the subject matter which the applicant regards as his invention.
Claim 20 is rejected under 35 U.S.C. 112(b) or 35 U.S.C. 112 (pre-AIA ), second paragraph, as being indefinite for failing to particularly point out and distinctly claim the subject matter which the inventor or a joint inventor (or for applications subject to pre-AIA 35 U.S.C. 112, the applicant), regards as the invention.
Claim 20 recites the limitation "the plurality of tissue organoids" in lines 1 – 2. There is insufficient antecedent basis for this limitation in the claim.
Claim Rejections - 35 USC § 101
35 U.S.C. 101 reads as follows:
Whoever invents or discovers any new and useful process, machine, manufacture, or composition of matter, or any new and useful improvement thereof, may obtain a patent therefor, subject to the conditions and requirements of this title.
Claims 1 – 9, 20 – 21, 24, 26 – 28, and 30 – 34 are rejected under 35 U.S.C. 101 because the claimed invention is directed to a judicial exception (i.e., a law of nature, a natural phenomenon, or an abstract idea) without significantly more.
Examiners should determine whether a claim satisfies the criteria for subject matter eligibility by evaluating the claim in accordance with the flowchart in MPEP 2016(III).
Eligibility Step 1:
Under Step 1 of the 2019 Revised Patent Subject Matter Eligibility Guidance, it must be determined whether each claim as a whole falls within one of the statutory categories of invention (i.e., a process, machine, manufacture, or composition of matter). See MPEP 2106.03. In the instant application, claims 1 – 8, 20 – 21, 24, 26 – 28, and 30 – 33 are directed to a method (i.e., a process); claim 9 is directed to a method (i.e., a process); and claim 34 is directed to a method (i.e., a process).
While each one of claims 1 – 9, 20 – 21, 24, 26 – 28, and 30 – 34 appears to fall within one or more statutory categories of invention, the Office has determined that the full eligibility analysis is required because there is doubt as to whether the applicant is effectively seeking coverage for a judicial exception itself. The eligibility of each claim is not self-evident at least because each claim as a whole did not appear to clearly improve a technology or computer functionality. To the contrary, each claim as a whole appeared to merely apply one or more judicial exceptions on a computer.
Accordingly, it has been determined that each one of claims 1 – 9, 20 – 21, 24, 26 – 28, and 30 – 34 as a whole falls within one or more statutory categories under Step 1, and the Office proceeds with the full eligibility analysis (the Alice/Mayo test described in MPEP 2106(III)) as discussed below.
Eligibility Step 2A, Prong One:
Under Step 2A, Prong One of the 2019 Revised Patent Subject Matter Eligibility Guidance, it must be determined whether each claim is directed to one or more of the judicial exceptions (i.e., an abstract idea, law of nature, or natural phenomenon). See MPEP 2106.04(II)(A)(1). After evaluation, it has been determined that claims 1 – 9, 20 – 21, 24, 26 – 28, and 30 – 34 are directed to judicial exceptions because claims 1 – 9, 20 – 21, 24, 26 – 28, and 30 – 34 recite abstract ideas.
Independent claims 1 and 9 are determined to be directed to a judicial exception (i.e., an abstract idea in the “mental process” grouping). Representative claim 9 recites the abstract idea identified in bold as:
A method for identifying one or more tissue organoids in a plurality of tissue organoids matching a biological property of a tissue in a subject (mental process), the method comprising:
at a computer system comprising at least one processor and a memory storing at least one program for execution by the at least one processor:
inputting information about the test subject into a multi-task model comprising a plurality of parameters and one or more hidden layers, wherein the multi-task model is trained to apply the plurality of parameters to the information about the test subject through a plurality of instructions to generate, as output from the multi-task model, a plurality of outputs comprising (i) a predicted effect of the pharmaceutical agent in the test subject and (ii) for each respective cell type variable in a set of one or more cell type variables, a corresponding cell type classification, wherein the information about the test subject comprises, for each respective cellular constituent in a plurality of cellular constituents, a corresponding abundance value for the abundance of the respective cellular constituent in a biological sample of the test subject (mental process);
obtaining a latent representation of the information about the test subject (mental process) from a respective hidden layer in the one or more hidden layers; [[and]] (This limitation is not in claim 1.)
comparing the latent representation of the information about the test subject to a plurality of latent representations, wherein each respective latent representation in the plurality of latent representations is of information about a respective tissue organoid, in a plurality of tissue organoids (mental process), obtained from the multi-task model; and (This limitation is not in claim 1.)
identifying one or more respective tissue organoids, in the plurality of tissue organoids, that satisfy a set of one or more similarity criterion based on the comparing, thereby identifying the one or more tissue organoids matching the biological property of the tissue in the subject (mental process). (This limitation is not in claim 1.)
With further respect to claim 1, the claim recites the abstract idea identified in bold as “a method for predicting an effect of a pharmaceutical agent in a test subject of a first species.”
Regarding claims 1 and 9, the limitations in bold identified as “a method for predicting an effect of a pharmaceutical agent in a test subject of a first species” in claim 1, “a method for identifying one or more tissue organoids in a plurality of tissue organoids matching a biological property of a tissue in a subject” in claim 9, “inputting information about the test subject … to generate, as output from the multi-task model, a plurality of outputs comprising (i) a predicted effect of the pharmaceutical agent in the test subject and (ii) for each respective cell type variable in a set of one or more cell type variables, a corresponding cell type classification, wherein the information about the test subject comprises, for each respective cellular constituent in a plurality of cellular constituents, a corresponding abundance value for the abundance of the respective cellular constituent in a biological sample of the test subject” in claims 1 and 9, “obtaining a latent representation of the information about the test subject” in claim 9, “comparing the latent representation of the information about the test subject to a plurality of latent representations, wherein each respective latent representation in the plurality of latent representations is of information about a respective tissue organoid, in a plurality of tissue organoids” in claim 9, and “identifying one or more respective tissue organoids, in the plurality of tissue organoids, that satisfy a set of one or more similarity criterion based on the comparing, thereby identifying the one or more tissue organoids matching the biological property of the tissue in the subject.” These limitations fall into the “mental process” grouping of abstract ideas because they require an assessment of information about tissue organoids and a biological sample from a test subject to identify the tissue organoids used in evaluating a predicted effect of a pharmaceutical agent on the test subject. See MPEP 2106.04(a)(2)(III). Thus, the limitations may be practically performed in the human mind using observation, evaluation, judgment, and/or opinions, and claims 1 and 9 recite an abstract idea in the "mental process" grouping under Step 2A, Prong One.
Independent claim 34 is determined to be directed to a judicial exception including abstract ideas (i.e., an abstract idea in the “mental process” grouping). Claim 34 recites the abstract idea identified in bold as:
A method for training a model to predict an effect of a candidate pharmaceutical agent in a test subject of a first species, the method comprising:
at a computer system comprising at least one processor and a memory storing at least one program for execution by the at least one processor:
obtaining, for each respective training sample in a first plurality of training samples, wherein each respective training sample in the first plurality of training samples comprises a tissue organoid or tissue organoid culture, formed from cells of the first species, that has been exposed to a perturbation:
a corresponding plurality of abundance values comprising, for each respective cellular constituent in a plurality of cellular constituents, a corresponding abundance value for the abundance of the respective cellular constituent in the respective training sample after exposure to the candidate pharmaceutical agent (mental process),
a corresponding experimentally measured effect of the candidate pharmaceutical agent on the respective training sample (mental process), and
a corresponding set of one or more cell type classifications comprising, for each respective cell type variable in a set of one or more cell type variables, a corresponding cell type classification for the respective cell type variable (mental process);
obtaining, for each respective training sample in a second plurality of training samples, wherein each respective training sample in the second plurality of training samples comprises a biological sample from a respective subject in a plurality of subjects of the first species:
a corresponding plurality of abundance values comprising, for each respective cellular constituent in the plurality of cellular constituents, a corresponding abundance value for the abundance of the respective cellular constituent in the respective training sample (mental process), and
a corresponding set of one or more cell type classifications comprising, for each respective cell type variable in the set of one or more cell type variables, a corresponding cell type classification for the respective cell type variable (mental process);
performing a first dimensionality reduction analysis across the corresponding plurality of abundance values for each respective training sample in the first plurality of training samples (mental process), thereby:
learning a first mapping function that maps a corresponding plurality of abundance values comprising, for each respective cellular constituent in the plurality of cellular constituents, a corresponding abundance value for the abundance of the respective cellular constituent in a tissue organoid or tissue organoid culture, formed from cells of the first species, that has been exposed to a perturbation, into a first latent feature space comprising a first plurality of dimensions that is less than the number of cellular constituents in the plurality of constituents (mental process), and
generating, for each respective training sample [in] the first plurality of training samples, a first corresponding representation of the corresponding plurality of abundance values in the first latent feature space according to the first mapping function (mental process);
performing a second dimensionality reduction analysis across the corresponding plurality of abundance values for each respective training sample in the second plurality of training samples (mental process), thereby:
learning a second mapping function that maps a corresponding plurality of abundance values comprising, for each respective cellular constituent in the plurality of cellular constituents, a corresponding abundance value for the abundance of the respective cellular constituent in a biological sample of the first species, into a second latent feature space comprising the first plurality of dimensions (mental process), and
generating, for each respective training sample in the second plurality of training samples, a corresponding representation of the corresponding plurality of abundance values in the second latent feature space according to the second mapping function (mental process);
learning a third mapping function that maps a representation of a corresponding plurality of abundance values in the first latent feature space to the second latent feature space (mental process);
generating, for each respective training sample in the first plurality of training samples, a second corresponding representation of the corresponding plurality of abundance values in the second latent feature space according to the third mapping function (mental process);
inputting, for each respective training sample in the first plurality of training samples, corresponding information about the respective training sample into a multi-task model comprising a plurality of parameters, wherein the multi-task model applies the plurality of parameters to the information about the training [subject] through a plurality of instructions to generate, as output from the multi-task model, a corresponding plurality of outputs (mental process), wherein:
the corresponding plurality of outputs comprises (i) a predicted effect of the candidate pharmaceutical agent on the respective training sample and (ii) for each respective cell type variable in the set of one or more cell type variables, a corresponding cell type classification for the respective training sample (mental process), and
the information about the respective training sample comprises the second corresponding representation of the corresponding plurality of abundance values in the second latent feature space (mental process);
inputting, for each respective training sample in the second plurality of training samples, corresponding information about the respective training sample into the multi-task model, wherein the information about the respective training sample comprises the corresponding representation of the corresponding plurality of abundance values in the second latent feature space (mental process); and
adjusting the plurality of parameters based on:
for each respective training sample in the first plurality of training samples, one or more differences between (i) the corresponding plurality of outputs and (ii) a set of labels comprising (a) the corresponding experimentally measured effect of the candidate pharmaceutical agent on the respective training sample and (b) the corresponding cell type classification for each respective cell type variable in the set of one or more cell type variables (mental process), and
for each respective training sample in the second plurality of training samples, one or more differences between (i) the corresponding plurality of outputs and (ii) a set of labels comprising the corresponding cell type classification for each respective cell type variable in the set of one or more cell type variables (mental process).
Regarding claim 34, the limitations in bold identified as “predict an effect of a candidate pharmaceutical agent in a test subject of a first species,” “obtaining, for each respective training sample in a first plurality of training samples, wherein each respective training sample in the first plurality of training samples comprises a tissue organoid or tissue organoid culture, formed from cells of the first species, that has been exposed to a perturbation: a corresponding plurality of abundance values comprising, for each respective cellular constituent in a plurality of cellular constituents, a corresponding abundance value for the abundance of the respective cellular constituent in the respective training sample after exposure to the candidate pharmaceutical agent; a corresponding experimentally measured effect of the candidate pharmaceutical agent on the respective training sample; and a corresponding set of one or more cell type classifications comprising, for each respective cell type variable in a set of one or more cell type variables, a corresponding cell type classification for the respective cell type variable,” “obtaining, for each respective training sample in a second plurality of training samples, wherein each respective training sample in the second plurality of training samples comprises a biological sample from a respective subject in a plurality of subjects of the first species: a corresponding plurality of abundance values comprising, for each respective cellular constituent in the plurality of cellular constituents, a corresponding abundance value for the abundance of the respective cellular constituent in the respective training sample; and a corresponding set of one or more cell type classifications comprising, for each respective cell type variable in the set of one or more cell type variables, a corresponding cell type classification for the respective cell type variable,” “performing a first dimensionality reduction analysis across the corresponding plurality of abundance values for each respective training sample in the first plurality of training samples,” “learning a first mapping function that maps a corresponding plurality of abundance values comprising, for each respective cellular constituent in the plurality of cellular constituents, a corresponding abundance value for the abundance of the respective cellular constituent in a tissue organoid or tissue organoid culture, formed from cells of the first species, that has been exposed to a perturbation, into a first latent feature space comprising a first plurality of dimensions that is less than the number of cellular constituents in the plurality of constituents (mental process),” “generating, for each respective training sample [in] the first plurality of training samples, a first corresponding representation of the corresponding plurality of abundance values in the first latent feature space according to the first mapping function,” “performing a second dimensionality reduction analysis across the corresponding plurality of abundance values for each respective training sample in the second plurality of training samples,” “learning a second mapping function that maps a corresponding plurality of abundance values comprising, for each respective cellular constituent in the plurality of cellular constituents, a corresponding abundance value for the abundance of the respective cellular constituent in a biological sample of the first species, into a second latent feature space comprising the first plurality of dimensions,” “generating, for each respective training sample in the second plurality of training samples, a corresponding representation of the corresponding plurality of abundance values in the second latent feature space according to the second mapping function,” “learning a third mapping function that maps a representation of a corresponding plurality of abundance values in the first latent feature space to the second latent feature space,” “generating, for each respective training sample in the first plurality of training samples, a second corresponding representation of the corresponding plurality of abundance values in the second latent feature space according to the third mapping function,” “inputting, for each respective training sample in the first plurality of training samples, corresponding information about the respective training sample … to generate, as output from the multi-task model, a corresponding plurality of outputs, wherein: the corresponding plurality of outputs comprises (i) a predicted effect of the candidate pharmaceutical agent on the respective training sample and (ii) for each respective cell type variable in the set of one or more cell type variables, a corresponding cell type classification for the respective training sample, and the information about the respective training sample comprises the second corresponding representation of the corresponding plurality of abundance values in the second latent feature space,” “inputting, for each respective training sample in the second plurality of training samples, corresponding information about the respective training sample … wherein the information about the respective training sample comprises the corresponding representation of the corresponding plurality of abundance values in the second latent feature space,” and “adjusting the plurality of parameters based on: for each respective training sample in the first plurality of training samples, one or more differences between (i) the corresponding plurality of outputs and (ii) a set of labels comprising (a) the corresponding experimentally measured effect of the candidate pharmaceutical agent on the respective training sample and (b) the corresponding cell type classification for each respective cell type variable in the set of one or more cell type variables (mental process), and for each respective training sample in the second plurality of training samples, one or more differences between (i) the corresponding plurality of outputs and (ii) a set of labels comprising the corresponding cell type classification for each respective cell type variable in the set of one or more cell type variables.” These limitations fall into the “mental process” grouping of abstract ideas because they require a dynamic assessment of information about tissue organoids and a biological sample from a test subject to identify the tissue organoids used in evaluating a predicted effect of a pharmaceutical agent on the test subject. See MPEP 2106.04(a)(2)(III). Thus, the limitations may be practically performed in the human mind using observation, evaluation, judgment, and/or opinions, and claim 34 recites an abstract idea in the "mental process" grouping under Step 2A, Prong One.
Dependent claims 2 – 8, 20 – 21, 24, 26 – 28, and 30 – 33 are directed to one or more judicial exceptions (i.e., abstract idea exceptions) under Step 2A, Prong One of the full eligibility analysis as follows:
Regarding claims 2 – 8, 20 – 21, 24, 26 – 28, and 30 – 33, the claims recite the abstract idea identified in bold as “the predicted effect comprises a prediction for cell death of a cancer cell in the subject in response to administration of the pharmaceutical agent to the subject” in claim 2, “the pharmaceutical agent is a chemotherapeutic agent” in claim 3, “the pharmaceutical agent is selected from the group consisting of lenalidomid, pembrolizumab, trastuzumab, bevacizumab, rituximab, ibrutinib, human papillomavirus quadrivalent (types 6, 11, 16, and 18) vaccine, pertuzumab, pemetrexed, nilotinib, denosumab, abiraterone acetate, promacta, imatinib, everolimus, palbociclib, erlotinib, bortezomib, bortezomib, nivolumab, atezolizumab,daratumumab, enzalutamide, obinutuzumab, ruxolitinib, venetoclax, osimertinib, and pomalidomide” in claim 4, “a respective output in the plurality of outputs” in claim 5, “when the predicted effect of the pharmaceutical agent in the test subject satisfies a first set of one or more criterion, recommending a first therapy that comprises administration of the pharmaceutical agent to the subject; and when the predicted effect of the pharmaceutical agent in the test subject does not satisfy the first set of one or more criterion, recommending a second therapy that is different from the first therapy” in claim 6, “when the predicted effect of the pharmaceutical agent in the test subject satisfies a first set of one or more criterion, administering a first therapy to the subject, wherein the first therapy that comprises administration of the pharmaceutical agent; and when the predicted effect of the pharmaceutical agent in the test subject does not satisfy the first set of one or more criterion, administering a second therapy to the subject, wherein the second therapy is different from the first therapy” in claim 7, “when the predicted effect of the pharmaceutical agent in the test subject satisfies a first set of one or more criterion, recommending the test subject for a clinical trial of the pharmaceutical agent to the subject; and when the predicted effect of the pharmaceutical agent in the test subject does not satisfy the first set of one or more criterion, not recommending the test subject for the clinical trial” in claim 8, “the plurality of tissue organoids comprises a plurality of tumor organoids” in claim 20, “a linear mapping function that transforms the plurality of abundance values into a first latent feature space comprising fewer dimensions than the number of respective cellular constituents in the plurality of cellular constituents” in claim 21, “the set of one or more cell type variables comprises a variable selected from the group consisting of cell histology, disease type, disease stage, disease grade, tissue type, and tissue site” in claim 24, “each respective cellular constituent in the plurality of cellular constituents is a different mRNA species” in claim 26, “obtaining, in electronic form, a plurality of nucleic acid sequences for mRNA from the biological sample of the test subject; and determining, for each respective cellular constituent in the plurality of cellular constituents, the corresponding abundance value from the plurality of nucleic acid sequences” in claim 27, “the obtaining (i) comprises sequencing the mRNA from the biological sample of the test subject, thereby obtaining the plurality of nucleic acid sequences” in claim 28, “the biological sample of the subject comprises a diseased tissue of the subject” in claim 30, “the diseased tissue of the subject is a cancerous tissue” in claim 31, “the biological sample of the subject comprises a biological fluid from the subject” in claim 32, and “the subject has a cancer selected from the group consisting of a carcinoma, lymphoma, blastoma, glioblastoma, sarcoma, leukemia, breast cancer, squamous cell cancer, lung cancer, small-cell lung cancer, non-small cell lung cancer (NSCLC), adenocarcinoma of the lung, squamous carcinoma of the lung, head and neck cancer, cancer of the peritoneum, hepatocellular cancer, gastric or stomach cancer, pancreatic cancer, ovarian cancer, cervical cancer, liver cancer, bladder cancer, hepatoma, colon cancer, colorectal cancer, endometrial or uterine carcinoma, salivary gland carcinoma, kidney or renal cancer, liver cancer, prostate cancer, vulval cancer, thyroid cancer, hepatic carcinoma, B-cell lymphoma, low grade/follicular non-Hodgkin's lymphoma (NHL), small lymphocytic (SL) NHL, intermediate grade/follicular NHL, intermediate grade diffuse NHL, high grade immunoblastic NHL, high grade lymphoblastic NHL, high grade small non-cleaved cell NHL, bulky disease NHL, mantle cell lymphoma, AIDS-related lymphoma, Waldenstrom's Macroglobulinemia, chronic lymphocytic leukemia (CLL), acute lymphoblastic leukemia (ALL), hairy cell leukemia, and chronic myeloblastic leukemia” in claim 33. These limitations fall into the “mental process” grouping of abstract ideas because they require a dynamic assessment of information on tissue organoids and a biological sample from a test subject to identify the tissue organoids used in evaluating a predicted effect of a pharmaceutical agent on the test subject. See MPEP 2106.04(a)(2)(III). Thus, the limitations may be practically performed in the human mind using observation, evaluation, judgment, and/or opinions, and claims 2 – 8, 20 – 21, 24, 26 – 28, and 30 – 33 recite an abstract idea in the "mental process" grouping under Step 2A, Prong One.
Eligibility Step 2A, Prong Two:
Independent claims 1 and 9 recite additional limitations beyond the judicial exceptions. Representative claim 9 recites the additional limitations identified in bold as:
A method for identifying one or more tissue organoids in a plurality of tissue organoids matching a biological property of a tissue in a subject, the method comprising:
at a computer system comprising at least one processor and a memory storing at least one program for execution by the at least one processor:
inputting information about the test subject into a multi-task model comprising a plurality of parameters and one or more hidden layers, wherein the multi-task model is trained to apply the plurality of parameters to the information about the test subject through a plurality of instructions to generate, as output from the multi-task model, a plurality of outputs comprising (i) a predicted effect of the pharmaceutical agent in the test subject and (ii) for each respective cell type variable in a set of one or more cell type variables, a corresponding cell type classification, wherein the information about the test subject comprises, for each respective cellular constituent in a plurality of cellular constituents, a corresponding abundance value for the abundance of the respective cellular constituent in a biological sample of the test subject;
obtaining a latent representation of the information about the test subject from a respective hidden layer in the one or more hidden layers; [[and]] (This limitation is not in claim 1.)
comparing the latent representation of the information about the test subject to a plurality of latent representations, wherein each respective latent representation in the plurality of latent representations is of information about a respective tissue organoid, in a plurality of tissue organoids, obtained from the multi-task model; and (This limitation is not in claim 1.)
identifying one or more respective tissue organoids, in the plurality of tissue organoids, that satisfy a set of one or more similarity criterion based on the comparing, thereby identifying the one or more tissue organoids matching the biological property of the tissue in the subject. (This limitation is not in claim 1.)
Claims 1 and 9 recite the additional limitations identified in bold as “a computer system comprising at least one processor and a memory storing at least one program for execution by the at least one processor,” “a multi-task model comprising a plurality of parameters and one or more hidden layers, wherein the multi-task model is trained to apply the plurality of parameters to the information about the test subject through a plurality of instructions,” “hidden layer in the one or more hidden layers,” and “the multi-task model.”
Regarding the consideration under MPEP 2106.04(d)(2), each one of the claims as a whole does not amount to a particular treatment or prophylaxis, but an assessment of information about tissue organoids and a biological sample from a test subject to identify the tissue organoids used in evaluating a predicted effect of a pharmaceutical agent on the test subject.
Regarding the consideration under MPEP 2106.05(a), each one of claims 1 and 9 as a whole does not purport to improve the functioning of a computer itself or improve any other technology or technical field. Claims 1 and 9 recite a combination of limitations (including the abstract idea in italicized font and the additional limitation in bold font) “inputting information about the test subject into a multi-task model comprising a plurality of parameters and one or more hidden layers, wherein the multi-task model is trained to apply the plurality of parameters to the information about the test subject through a plurality of instructions to generate, as output from the multi-task model, a plurality of outputs comprising (i) a predicted effect of the pharmaceutical agent in the test subject and (ii) for each respective cell type variable in a set of one or more cell type variables, a corresponding cell type classification, wherein the information about the test subject comprises, for each respective cellular constituent in a plurality of cellular constituents, a corresponding abundance value for the abundance of the respective cellular constituent in a biological sample of the test subject.” The claims do not improve the functionality of the computer, the processor, the memory, the computer program, or the multi-task model (i.e., comprising a plurality of parameters and one or more hidden layers), but rather invoke these generic computer components as tools to automate the abstract idea (i.e., the mental process of assessing information on tissue organoids and a biological sample from a test subject to identify the tissue organoids used in evaluating the predicted effect of the pharmaceutical agent on the test subject). Furthermore, the computer, the processor, the memory, the computer program, and the multi-task model are claimed at a high level of generality using inputs to generate outputs without explaining how the model processes those inputs and generates the outputs. The claims fail to cover a technical solution to a technological problem (i.e., with a computer functionality). At best, the combination of all claim elements amounts to an improvement to the abstract idea itself rather than improving any technology. Thus, even when considering any combination of the elements, each one of claims 1 and 9 as a whole still does not integrate the recited exception into a practical application.
Regarding the consideration under MPEP 2106.05(f), each one of the additional limitations in bold above is determined to be mere instructions to apply an abstract idea. For instance, claims 1 and 9 recite a combination of limitations (including the abstract idea in italicized font and the additional limitation in bold font) “inputting information about the test subject into a multi-task model comprising a plurality of parameters and one or more hidden layers, wherein the multi-task model is trained to apply the plurality of parameters to the information about the test subject through a plurality of instructions to generate, as output from the multi-task model, a plurality of outputs comprising (i) a predicted effect of the pharmaceutical agent in the test subject and (ii) for each respective cell type variable in a set of one or more cell type variables, a corresponding cell type classification, wherein the information about the test subject comprises, for each respective cellular constituent in a plurality of cellular constituents, a corresponding abundance value for the abundance of the respective cellular constituent in a biological sample of the test subject.” The multi-task model, parameters, and hidden layers are claimed at a high level of generality to implement the abstract idea, and thus the claims merely amount to instructions to implement the abstract idea on generic computer technology. Accordingly, even when considering any combination of the elements, each one of claims 1 and 9 as a whole fails to integrate the abstract idea into a practical application.
Regarding the consideration under MPEP 2106.05(g), even if the steps of inputting, obtaining, and comparing were additional elements (which the Office does not concede), these steps would not add more than insignificant extra-solution activity to the judicial exception (i.e., the abstract idea of predicting anti-cancer drug response in the test subject). These steps represent the well-known pre-solution activity of necessary input data gathering, which is incidental to the primary process of using that input data to generate the outputs, including the predicted drug response. Thus, the steps of inputting, obtaining, and comparing would merely be nominal or tangential additions to the claim; however, the Office maintains that the steps of inputting, obtaining, and comparing are not actually additional elements, but rather the abstract idea as described above. Accordingly, even when considering any combination of the elements, each one of claims 1 and 9 as a whole does not integrate the abstract idea into a practical application under Step 2A, Prong Two.
Regarding the consideration under MPEP 2106.05(h), the additional limitations, individually or in combination, also amount to merely indicating a field of use or technological environment in which to apply the judicial exception. In the instant application, the additional limitations of the computer, the processor, the memory, the computer program, or the multi-task model do no more than link the abstract idea (i.e., the mental process of assessing information to identify tissue organoids used in evaluating the predicted drug response) to a particular technological environment, i.e., the multi-task model. Thus, the additional limitations fail to add an inventive concept to the claims.
Accordingly, in view of these considerations, the Office has determined that, even when considering any combination of the elements, each one of claims 1 and 9 as a whole does not integrate the abstract idea exception into a practical application under Step 2A, Prong Two.
Independent claim 34 recite additional limitations beyond the judicial exceptions. Claim 34 recites the additional limitations identified in bold as:
A method for training a model to predict an effect of a candidate pharmaceutical agent in a test subject of a first species, the method comprising:
at a computer system comprising at least one processor and a memory storing at least one program for execution by the at least one processor:
obtaining, for each respective training sample in a first plurality of training samples, wherein each respective training sample in the first plurality of training samples comprises a tissue organoid or tissue organoid culture, formed from cells of the first species, that has been exposed to a perturbation:
a corresponding plurality of abundance values comprising, for each respective cellular constituent in a plurality of cellular constituents, a corresponding abundance value for the abundance of the respective cellular constituent in the respective training sample after exposure to the candidate pharmaceutical agent,
a corresponding experimentally measured effect of the candidate pharmaceutical agent on the respective training sample, and
a corresponding set of one or more cell type classifications comprising, for each respective cell type variable in a set of one or more cell type variables, a corresponding cell type classification for the respective cell type variable;
obtaining, for each respective training sample in a second plurality of training samples, wherein each respective training sample in the second plurality of training samples comprises a biological sample from a respective subject in a plurality of subjects of the first species:
a corresponding plurality of abundance values comprising, for each respective cellular constituent in the plurality of cellular constituents, a corresponding abundance value for the abundance of the respective cellular constituent in the respective training sample, and
a corresponding set of one or more cell type classifications comprising, for each respective cell type variable in the set of one or more cell type variables, a corresponding cell type classification for the respective cell type variable;
performing a first dimensionality reduction analysis across the corresponding plurality of abundance values for each respective training sample in the first plurality of training samples, thereby:
learning a first mapping function that maps a corresponding plurality of abundance values comprising, for each respective cellular constituent in the plurality of cellular constituents, a corresponding abundance value for the abundance of the respective cellular constituent in a tissue organoid or tissue organoid culture, formed from cells of the first species, that has been exposed to a perturbation, into a first latent feature space comprising a first plurality of dimensions that is less than the number of cellular constituents in the plurality of constituents, and
generating, for each respective training sample f in the first plurality of training samples, a first corresponding representation of the corresponding plurality of abundance values in the first latent feature space according to the first mapping function;
performing a second dimensionality reduction analysis across the corresponding plurality of abundance values for each respective training sample in the second plurality of training samples, thereby:
learning a second mapping function that maps a corresponding plurality of abundance values comprising, for each respective cellular constituent in the plurality of cellular constituents, a corresponding abundance value for the abundance of the respective cellular constituent in a biological sample of the first species, into a second latent feature space comprising the first plurality of dimensions, and
generating, for each respective training sample in the second plurality of training samples, a corresponding representation of the corresponding plurality of abundance values in the second latent feature space according to the second mapping function;
learning a third mapping function that maps a representation of a corresponding plurality of abundance values in the first latent feature space to the second latent feature space;
generating, for each respective training sample in the first plurality of training samples, a second corresponding representation of the corresponding plurality of abundance values in the second latent feature space according to the third mapping function;
inputting, for each respective training sample in the first plurality of training samples, corresponding information about the respective training sample into a multi-task model comprising a plurality of parameters, wherein the multi-task model applies the plurality of parameters to the information about the training [subject] through a plurality of instructions to generate, as output from the multi-task model, a corresponding plurality of outputs, wherein:
the corresponding plurality of outputs comprises (i) a predicted effect of the candidate pharmaceutical agent on the respective training sample and (ii) for each respective cell type variable in the set of one or more cell type variables, a corresponding cell type classification for the respective training sample, and
the information about the respective training sample comprises the second corresponding representation of the corresponding plurality of abundance values in the second latent feature space;
inputting, for each respective training sample in the second plurality of training samples, corresponding information about the respective training sample into the multi-task model, wherein the information about the respective training sample comprises the corresponding representation of the corresponding plurality of abundance values in the second latent feature space; and
adjusting the plurality of parameters based on:
for each respective training sample in the first plurality of training samples, one or more differences between (i) the corresponding plurality of outputs and (ii) a set of labels comprising (a) the corresponding experimentally measured effect of the candidate pharmaceutical agent on the respective training sample and (b) the corresponding cell type classification for each respective cell type variable in the set of one or more cell type variables, and
for each respective training sample in the second plurality of training samples, one or more differences between (i) the corresponding plurality of outputs and (ii) a set of labels comprising the corresponding cell type classification for each respective cell type variable in the set of one or more cell type variables.
Claim 34 recites the additional limitations identified in bold as “a method for training a model,” “a computer system comprising at least one processor and a memory storing at least one program for execution by the at least one processor,” and “a multi-task model comprising a plurality of parameters, wherein the multi-task model applies the plurality of parameters to the information about the training [subject] through a plurality of instructions.”
Regarding the consideration under MPEP 2106.04(d)(2), claim 34 as a whole does not amount to a particular treatment or prophylaxis, but an assessment of information about tissue organoids and a biological sample from a test subject to identify the tissue organoids used in evaluating a predicted effect of a pharmaceutical agent on the test subject.
Regarding the consideration under MPEP 2106.05(a), claim 34 as a whole does not purport to improve the functioning of a computer itself or improve any other technology or technical field. Claim 34 recites a combination of limitations (including the abstract idea in italicized font and the additional limitation in bold font) “inputting, for each respective training sample in the first plurality of training samples, corresponding information about the respective training sample into a multi-task model comprising a plurality of parameters, wherein the multi-task model applies the plurality of parameters to the information about the training [subject] through a plurality of instructions to generate, as output from the multi-task model, a corresponding plurality of outputs.” The claims do not improve the functionality of the computer, the processor, the memory, the computer program, or the multi-task model (i.e., comprising a plurality of parameters and one or more hidden layers), but rather invoke these generic computer components as tools to automate the abstract idea (i.e., the mental process of assessing information on tissue organoids and a biological sample from a test subject to identify the tissue organoids used in evaluating the predicted effect of the pharmaceutical agent on the test subject). Furthermore, the computer, the processor, the memory, the computer program, and the multi-task model are claimed at a high level of generality using inputs to generate outputs without explaining how the model processes those inputs and generates the outputs. The claims fail to cover a technical solution to a technological problem (i.e., with a computer functionality). At best, the combination of all claim elements amounts to an improvement to the abstract idea itself rather than improving any technology. Thus, even when considering any combination of the elements, claim 34 as a whole still does not integrate the recited exception into a practical application.
Regarding the consideration under MPEP 2106.05(f), each one of the additional limitations in bold above is determined to be mere instructions to apply an abstract idea. For instance, claim 34 recites a combination of limitations (including the abstract idea in italicized font and the additional limitation in bold font) “inputting, for each respective training sample in the first plurality of training samples, corresponding information about the respective training sample into a multi-task model comprising a plurality of parameters, wherein the multi-task model applies the plurality of parameters to the information about the training [subject] through a plurality of instructions to generate, as output from the multi-task model, a corresponding plurality of outputs.” The multi-task model, parameters, and hidden layers are claimed at a high level of generality, and thus the claims merely amount to instructions to implement the abstract idea on generic computer technology. Accordingly, even when considering any combination of the elements, claim 34 as a whole fails to integrate the abstract idea into a practical application.
Regarding the consideration under MPEP 2106.05(g), claim 34 recites the step of obtaining, for each respective training sample in an associated on of the first and second pluralities of training samples, the abundance values and the cell type classifications. Claim 34 further recites the step of obtaining, for each respective training sample in the first plurality of training samples, the experimentally measured effect. Claim 34 further recites the steps of performing the first dimensionality reduction analysis, learning the first mapping function, generating the first corresponding representation, performing the second dimensionality reduction analysis, learning the second mapping function, generating the corresponding representation, learning the third mapping function, generating the second corresponding representation, and inputting the second corresponding representation. These steps represent the abstract idea as explained under Step 2A, Prong One above. Even if one or more of these steps were an additional element (which the Office does not concede), those steps would not add more than insignificant extra-solution activity to the judicial exception (i.e., the abstract idea of dynamically predicting anti-cancer drug response in the test subject). These steps represent the well-known pre-solution activity of necessary input data gathering, which is incidental to the primary process of using that input data to generate the outputs, including the predicted drug response. Thus, the step of preparing and gathering necessary input data would merely be nominal or tangential additions to the claim; however, the Office maintains that these steps are not actually additional elements, but rather the abstract idea as described above. Accordingly, even when considering any combination of the elements, claim 34 as a whole does not integrate the abstract idea into a practical application.
Regarding the consideration under MPEP 2106.05(h), the additional limitations, individually or in combination, also amount to merely indicating a field of use or technological environment in which to apply the judicial exception. In the instant application, the additional limitations of the computer, the processor, the memory, the computer program, or the multi-task model do no more than link the abstract idea (i.e., the mental process of assessing information to identify tissue organoids used in evaluating the predicted drug response) to a particular technological environment, i.e., the multi-task model. Thus, the additional limitations fail to add an inventive concept to the claims.
Accordingly, in view of these considerations, the Office has determined that, even when considering any combination of the elements, claim 34 as a whole does not integrate the abstract idea exception into a practical application under Step 2A, Prong Two.
Dependent claims 2 – 8, 20 – 21, 24, 26 – 28, and 30 – 33 present additional information in tandem with further details regarding elements from independent claim 1 and are therefore directed to one or more abstract ideas for similar reasons as given Under Step 2A, Prong One above. With further regard to claims 5, 7, and 21, these claims further recite additional limitations, and these additional limitations fail to integrate the abstract idea into a practical application under Step 2A, Prong Two of the full eligibility analysis as follows:
Regarding claims 5, 7, and 21, the claims recite the additional limitations identified in bold as “the multi- task model comprises a partially connected neural network defining a plurality of tasks, each respective task in the plurality of tasks corresponding to a respective output in the plurality of outputs, wherein the partially-connected neural network comprises (a) a first set of layers shared between the plurality of tasks and (b) for each respective task in the plurality of tasks a corresponding second set of layers unique to the respective task” in claim 5, “when the predicted effect of the pharmaceutical agent in the test subject satisfies a first set of one or more criterion, administering a first therapy to the subject, wherein the first therapy that comprises administration of the pharmaceutical agent; and when the predicted effect of the pharmaceutical agent in the test subject does not satisfy the first set of one or more criterion, administering a second therapy to the subject, wherein the second therapy is different from the first therapy” in claim 7, and “the multi- task model comprises a linear mapping function that transforms the plurality of abundance values into a first latent feature space comprising fewer dimensions than the number of respective cellular constituents in the plurality of cellular constituents” in claim 21.
Regarding the consideration under MPEP 2106.04(d)(2), each one of claims 5, 7, and 21 as a whole does not amount to a particular treatment or prophylaxis. At best, claim 7 recites “when the predicted effect of the pharmaceutical agent in the test subject satisfies a first set of one or more criterion, administering a first therapy to the subject, wherein the first therapy that comprises administration of the pharmaceutical agent; and when the predicted effect of the pharmaceutical agent in the test subject does not satisfy the first set of one or more criterion, administering a second therapy to the subject, wherein the second therapy is different from the first therapy.” However, these steps are not particular at least because the steps do not cover any particular criterion, any particular therapy, way of satisfying the criterion, or any particular pharmaceutical agent. The steps are instead merely instructions to "apply" the judicial exception in a generic way. Furthermore, the administration steps are extra-solution activities (i.e., post-solution activity for the main process of identifying a tissue organoid matching the test subject). Thus, the administration step does not integrate the mental analysis step into a practical application.
Regarding the consideration under MPEP 2106.05(a), each one of claims 5, 7, and 21 as a whole does not purport to improve the functioning of a computer itself or improve any other technology or technical field. At best, claim 5 recites a combination of limitations (including the abstract idea in italicized font and the additional limitation in bold font) ““the multi- task model comprises a partially connected neural network defining a plurality of tasks, each respective task in the plurality of tasks corresponding to a respective output in the plurality of outputs, wherein the partially-connected neural network comprises (a) a first set of layers shared between the plurality of tasks and (b) for each respective task in the plurality of tasks a corresponding second set of layers unique to the respective task.” Claim 5 does not improve the functionality of the multi-task model (i.e., being a generic multi-task model comprising the partially connected neural network defining a plurality of tasks, wherein the partially-connected neural network comprises (a) a first set of layers shared between the plurality of tasks and (b) for each respective task in the plurality of tasks a corresponding second set of layers unique to the respective task). To the contrary, claim 5 merely defines the generic multi-task model as a tool to automate the abstract idea (i.e., the mental process of assessing information on tissue organoids and a biological sample from a test subject to identify the tissue organoids used in evaluating the predicted effect of the pharmaceutical agent on the test subject). Furthermore, the partially connected neural network, the plurality of tasks, the first set of layers shared between the plurality of tasks, and, for each respective task in the plurality of tasks, a corresponding second set of layers unique to the respective task are claimed at a high level of generality using inputs to generate outputs without explaining how any of this computer technology processes those inputs and generates the outputs. Claim 5 fails to cover a technical solution to a technological problem (i.e., with a computer functionality). At best, the combination of all claim elements amounts to an improvement to the abstract idea itself rather than improving any technology. Thus, even when considering any combination of the elements, each one of claims 5, 7, and 21 as a whole still does not integrate the recited exception into a practical application.
Regarding the consideration under MPEP 2106.05(f), each one of the additional limitations in claims 5, 7, and 21 is determined to be mere instructions to apply an abstract idea. . At best, claim 5 recites the combination of limitations (including the abstract idea in italicized font and the additional limitation in bold font) ““the multi- task model comprises a partially connected neural network defining a plurality of tasks, each respective task in the plurality of tasks corresponding to a respective output in the plurality of outputs, wherein the partially-connected neural network comprises (a) a first set of layers shared between the plurality of tasks and (b) for each respective task in the plurality of tasks a corresponding second set of layers unique to the respective task.” The partially connected neural network, the plurality of tasks, the first set of layers, and the corresponding second set of layers are claimed at a high level of generality, and thus, the claims merely amount to instructions to implement the abstract idea on generic computer technology. Accordingly, even when considering any combination of the elements, claim 5 as a whole fails to integrate the abstract idea into a practical application.
Regarding the consideration under MPEP 2106.05(g), claim 7 recites the combination of limitations (including the abstract idea in italicized font and the additional limitation in bold font) “when the predicted effect of the pharmaceutical agent in the test subject satisfies a first set of one or more criterion, administering a first therapy to the subject, wherein the first therapy that comprises administration of the pharmaceutical agent; and when the predicted effect of the pharmaceutical agent in the test subject does not satisfy the first set of one or more criterion, administering a second therapy to the subject, wherein the second therapy is different from the first therapy.” This combination represents the insignificant post-solution activity of administering a first and second therapies based on the identified tissue organoid matching with the test subject and the predicted effect of the pharmaceutical agent on the test subject. This administration step is well-known, and also nominal and incidental to the main process of using that input data to generate the outputs, including the predicted drug response. Claim 21 recites the combination of limitations (including the abstract idea in italicized font and the additional limitation in bold font) “the multi- task model comprises a linear mapping function that transforms the plurality of abundance values into a first latent feature space comprising fewer dimensions than the number of respective cellular constituents in the plurality of cellular constituents.” This combination amounts to the insignificant pre-solution activity of necessary data gathering, i.e., obtaining the first latent feature space, which is one of multiple steps used to obtain the inputs for the multi-task model. This transforming step is well-known, and also nominal and incidental to the main process of using that input data to generate the outputs, including the predicted drug response. Accordingly, even when considering any combination of the elements, each one of claims 7 and 21 as a whole does not integrate the abstract idea into a practical application.
Regarding the consideration under MPEP 2106.05(h), the additional limitations, individually or in combination, also amount to merely indicating a field of use or technological environment in which to apply the judicial exception. In the instant application, the additional limitations of the computer, the processor, the memory, the computer program, the multi-task model, the partially connected neural network, the plurality of tasks, the first set of layers, and the corresponding second set of layers do no more than link the abstract idea (i.e., the mental process of assessing information to identify tissue organoids used in evaluating the predicted drug response) to a particular technological environment, i.e., the multi-task model. Thus, the additional limitations fail to add an inventive concept to the claims.
Accordingly, in view of these considerations, the Office has determined that each one of claims 2 – 8, 20 – 21, 24, 26 – 28, and 30 – 33 as a whole do not have one or more additional limitations, individually or in combination, that integrate the abstract idea exception into a practical application under Step 2A, Prong Two.
Eligibility Step 2B:
Regarding independent claims 1, 9, and 34, the Office carries over its identification of the additional elements from Step 2A, Prong Two so as to apply the same additional elements in Step 2B. See MPEP 2106.05(II). The Office further carries over its conclusions from the considerations discussed in MPEP 2106.05(a) through (c), (e) through (h) in Step 2A, Prong Two so as to apply the same considerations in Step 2B.
Under Step 2B of the 2019 Revised Patent Subject Matter Eligibility Guidance, it must be determined whether provide an inventive concept by determining if the claims include additional elements or a combination of elements that are sufficient to amount to significantly more than the judicial exception. After evaluation, there is no indication that an additional element or combination of elements are sufficient to amount to significantly more than the judicial exception. As discussed above with respect to integration of the abstract idea into a practical application, the additional limitations amount to mere instructions to apply an abstract idea under MPEP 2106.05(f). Evidence that using multi-task models (i.e., having latent components and mapping functions to predict drug responses) is well-understood, routine and conventional is provided by NPL Ammad-ud-din. Furthermore, the claimed inputting steps amount to the insignificant pre-solution activity of mere necessary data gathering under MPEP 2106.05(g). Evidence that retrieving information and performing repetitive calculations are well-understood, routine and conventional is provided by MPEP 2106.05(d), subsection II.
Furthermore, looking at the limitations as an ordered combination adds nothing that is not already present when looking at the elements taken individually. There is no indication that the combination of elements amounts to an inventive concept.
Therefore, whether taken individually or as an ordered combination, claims 1, 9, and 34 are rejected under 35 U.S.C. 101 as being directed to non-statutory subject matter.
Regarding dependent claims 2 – 8, 20 – 21, 24, 26 – 28, and 30 – 33, the Office carries over its identification of the additional elements from Step 2A, Prong Two so as to apply the same additional elements in Step 2B. See MPEP 2106.05(II). The Office further carries over its conclusions from the considerations discussed in MPEP 2106.05(a) through (c), (e) through (h) in Step 2A, Prong Two so as to apply the same considerations in Step 2B.
The dependent claims present additional abstract information in tandem with further details regarding the elements from the independent claim 1 and are, therefore, directed to an abstract idea under Step 2A, Prong Two for similar reasons as given above. With further regard to claims 5, 7, and 21, these claims further recite additional limitations, and these additional limitations fail to integrate the abstract idea into a practical application under Step 2A, Prong Two as described above. Evidence that using multi-task models (i.e., having latent components and mapping functions to predict drug responses) is well-understood, routine and conventional is provided by NPL Ammad-ud-din. Furthermore, the claimed inputting steps amount to the insignificant pre-solution activity of mere necessary data gathering under MPEP 2106.05(g). Evidence that retrieving information and performing repetitive calculations are well-understood, routine and conventional is provided by MPEP 2106.05(d), subsection II. Claims 2 – 8, 20 – 21, 24, 26 – 28, and 30 – 33 are all encompassed by the abstract idea grouping of mental processes.
Therefore, whether taken individually or as an ordered combination, claims 2 – 8, 20 – 21, 24, 26 – 28, and 30 – 33 are rejected under 35 U.S.C. 101 as being directed to non-statutory subject matter.
Claim Rejections - 35 USC § 103
The following is a quotation of 35 U.S.C. 103 which forms the basis for all obviousness rejections set forth in this Office action:
A patent for a claimed invention may not be obtained, notwithstanding that the claimed invention is not identically disclosed as set forth in section 102, if the differences between the claimed invention and the prior art are such that the claimed invention as a whole would have been obvious before the effective filing date of the claimed invention to a person having ordinary skill in the art to which the claimed invention pertains. Patentability shall not be negated by the manner in which the invention was made.
The factual inquiries set forth in Graham v. John Deere Co., 383 U.S. 1, 148 USPQ 459 (1966), that are applied for establishing a background for determining obviousness under 35 U.S.C. 103 are summarized as follows:
Determining the scope and contents of the prior art.
Ascertaining the differences between the prior art and the claims at issue.
Resolving the level of ordinary skill in the pertinent art.
Considering objective evidence present in the application indicating obviousness or nonobviousness.
This application currently names joint inventors. In considering patentability of the claims the examiner presumes that the subject matter of the various claims was commonly owned as of the effective filing date of the claimed invention(s) absent any evidence to the contrary. Applicant is advised of the obligation under 37 CFR 1.56 to point out the inventor and effective filing dates of each claim that was not commonly owned as of the effective filing date of the later invention in order for the examiner to consider the applicability of 35 U.S.C. 102(b)(2)(C) for any potential 35 U.S.C. 102(a)(2) prior art against the later invention.
Claims 1, 4, 21, 24, and 32 – 33 are rejected under 35 U.S.C. 103(a) as being unpatentable over NPL Ammad-ud-din in view of Colley (U.S. Pub. No. 2021/0090694 A1) and NPL Yang.
Regarding independent claim 1, NPL Ammad-ud-din teaches the limitations identified in bold as:
A method for predicting an effect of a pharmaceutical agent in a test subject of a first species (First Paragraph on Page 11 of NPL Ammad-ud-din. In the instant application, the broadest reasonable interpretation of “a method for predicting an effect of a pharmaceutical agent in a test subject of a first species” reads on the method in NPL Ammad-ud-din (First Paragraph on Page 11) for predicting the response of a new patient’s sample to a drug.), the method comprising:
at a computer system comprising at least one processor and a memory storing at least one program for execution by the at least one processor:
inputting information about the test subject into (Abstract and Second Paragraph on Page 12 of NPL Ammad-ud-din. In the instant application, the broadest reasonable interpretation of “inputting information about the test subject” reads on the activity in NPL Ammad-ud-din (Abstract and Second Paragraph on Page 12 of NPL Ammad-ud-din) of using genome-wide features of cell lines, i.e., including new/unseen patient cell lines, as input for a model.) a multi-task model comprising a plurality of parameters (Second Paragraph on Page 34 of NPL Ammad-ud-din. In the instant application, the broadest reasonable interpretation of “a multi-task model comprising a plurality of parameters” reads on the multi-task learning (MTL) approach in NPL Ammad-ud-din (Second Paragraph on Page 34 of NPL Ammad-ud-din) that applies shared parameters (e.g., representations) common among related tasks and task-specific parameters unique to the associated task as well-known in the art of machine learning at the time of filing.), wherein the multi-task model applies the plurality of parameters to the information about the test subject through a plurality of instructions to generate, as output from the multi-task model, a plurality of outputs comprising (i) a predicted effect of the pharmaceutical agent in the test subject (Second Paragraph on Page 12 and Third Paragraph to Fourth Paragraph on Page 34 of NPL Ammad-ud-din. In the instant application, the broadest reasonable interpretation of “the multi-task model applies the plurality of parameters to the information about the test subject through a plurality of instructions to generate, as output from the multi-task model, a plurality of outputs comprising (i) a predicted effect of the pharmaceutical agent in the test subject” reads on the multi-task learning (MTL) approach in NPL Ammad-ud-din (Second Paragraph on Page 12 and Third Paragraph to Fourth Paragraph on Page 34) that applies the shared parameters (e.g., representations) to the genomic-wide features of the new patient’s unseen cell line to generate the predicted drug responses in the new patients.) and (ii) for each respective cell type variable in a set of one or more cell type variables, a corresponding cell type classification, wherein the information about the test subject comprises a plurality of abundance values, the plurality of abundance values comprising, for each respective cellular constituent in a plurality of cellular constituents, a corresponding abundance value for the abundance of the respective cellular constituent in a biological sample of the test subject (First Paragraph on Page 24 of NPL Ammad-ud-din. In the instant application, the broadest reasonable interpretation of “the information about the test subject comprises a plurality of abundance values, the plurality of abundance values comprising, for each respective cellular constituent in a plurality of cellular constituents, a corresponding abundance value for the abundance of the respective cellular constituent in a biological sample of the test subject” reads on the genomic and transcriptomic profiles in NPL Ammad-ud-din (First Paragraph on Page 24) including the number of RNA copies present in a cell correlated with the abundance of the corresponding proteins produced in the cell and nucleic acid sequences determined via RNA sequencing.).
NPL Ammad-ud-din does not appear to explicitly disclose, but Colley teaches the limitation identified in bold as “a computer system comprising at least one processor and a memory storing at least one program for execution by the at least one processor” (Paragraph [0379] of Colley. In the instant application, the broadest reasonable interpretation of “a computer system comprising at least one processor and a memory storing at least one program for execution by the at least one processor” reads on the computer in Colley (Paragraph [0379]) including the computer system having one or more processors and memory storing one or more programs for execution by the one or more processors.).
NPL Ammad-ud-din does not appear to explicitly disclose, but NPL Yang teaches the limitation identified in bold as “for each respective cell type variable in a set of one or more cell type variables, a corresponding cell type classification” (Eighth Paragraph on Page 7 of NPL Yang. In the instant application, the broadest reasonable interpretation of “for each respective cell type variable in a set of one or more cell type variables, a corresponding cell type classification” reads on the cancer types, tissue types, and tissue sites in NPL Yang (Eighth Paragraph on Page 7).).
Therefore, it would have been obvious to one of ordinary skill in the art of computer-aided diagnosis in oncology at the time of filing to modify the method of NPL Ammad-ud-din to: include the computer system comprising at least one processor and a memory storing at least one program for execution by the at least one processor, as taught by Colley (Paragraph [0379]) in order to allow for precision-level analysis of patient health needs, in order to provide the right resources, at the right time, to the right patients (Paragraph [0120] of Colley); and implement, for each respective cell type variable in a set of one or more cell type variables, a corresponding cell type classification, as taught by NPL Yang (Eighth Paragraph on Page 7) in order to leverage the power of multitask learning to provide novel insights into the molecular underpinnings of drug response (Last Paragraph on Page 1 of NPL Yang).
Regarding claim 4, NPL Ammad-ud-din as modified by Colley and NPL Yang and applied to claim 1 teaches the limitation identified in bold as “the pharmaceutical agent is selected from the group consisting of lenalidomid, pembrolizumab, trastuzumab, bevacizumab, rituximab, ibrutinib, human papillomavirus quadrivalent (types 6, 11, 16, and 18) vaccine, pertuzumab, pemetrexed, nilotinib, denosumab, abiraterone acetate, promacta, imatinib, everolimus, palbociclib, erlotinib, bortezomib, bortezomib, nivolumab, atezolizumab,daratumumab, enzalutamide, obinutuzumab, ruxolitinib, venetoclax, osimertinib, and pomalidomide” (Paragraph [2820], [3431], [3442] , and [3482] of Colley. In the instant application, the broadest reasonable interpretation of “the pharmaceutical agent is selected from the group consisting of lenalidomid, pembrolizumab, trastuzumab, bevacizumab, rituximab, ibrutinib, human papillomavirus quadrivalent (types 6, 11, 16, and 18) vaccine, pertuzumab, pemetrexed, nilotinib, denosumab, abiraterone acetate, promacta, imatinib, everolimus, palbociclib, erlotinib, bortezomib, bortezomib, nivolumab, atezolizumab,daratumumab, enzalutamide, obinutuzumab, ruxolitinib, venetoclax, osimertinib, and pomalidomide” reads on the pharmaceutical agents in Colley (Paragraph [2820], [3431] ], [3442], and [3482]) including nivolumab, pembrolizumab, atezolizumab, avelumab, durvalumab, the triplet chemotherapy (i.e., of pemetrexed, bevacizumab, and carboplatin), pemetrexed, bevacizumab, carboplatin, capecitabine, trastuzumab, and tamoxifen).
Regarding claim 21, NPL Ammad-ud-din as modified by Colley and NPL Yang and applied to claim 1 teaches the limitation identified in bold as “the multi- task model comprises a linear mapping function that transforms the plurality of abundance values into a first latent feature space comprising fewer dimensions than the number of respective cellular constituents in the plurality of cellular constituents” (First Paragraph on Page 24; First Paragraph on Page 32 of NPL Ammad-ud-din. In the instant application, the broadest reasonable interpretation of “the multi- task model comprises a linear mapping function that transforms the plurality of abundance values into a first latent feature space comprising fewer dimensions than the number of respective cellular constituents in the plurality of cellular constituents” reads on the activity of latent variable modeling in NPL Ammad-ud-din (First Paragraph on Page 24; First Paragraph to Second Paragraph on Page 32) dependencies (i.e., the linear mapping function) between the high-dimensional data variables and the low-dimensional variables (i.e., the latent components presenting a concise and denoised summary). These dependencies are used in combination with matrix factorization for dimensionality reduction analysis, where the high-dimensional observed data is decomposed into multiple low-dimensional latent factors (typically much smaller than the observed data dimensionality that includes: the number of RNA copies present in a cell, the correlated abundance of the corresponding proteins produced in the cell, the nucleic acid sequences produced via RNA sequencing).).
Regarding claim 24, NPL Ammad-ud-din as modified by Colley and NPL Yang and applied to claim 1 does not appear to explicitly disclose, but NPL Yang teaches the limitation identified in bold as “the set of one or more cell type variables comprises a variable selected from the group consisting of cell histology, disease type, disease stage, disease grade, tissue type, and tissue site” (Eighth Paragraph on Page 7 of NPL Yang. In the instant application, the broadest reasonable interpretation of “the set of one or more cell type variables comprises a variable selected from the group consisting of cell histology, disease type, disease stage, disease grade, tissue type, and tissue site” reads on the cancer types, tissue types, and tissue sites in NPL Yang (Eighth Paragraph on Page 7).).
Regarding claim 32, NPL Ammad-ud-din as modified by Colley and NPL Yang and applied to claim 1 teaches the limitation identified in bold as “the biological sample of the subject comprises a biological fluid from the subject” (Paragraph [1197] of Colley. In the instant application, the broadest reasonable interpretation of “the biological sample of the subject comprises a biological fluid from the subject” reads on the sample type in Colley (Paragraph [1197]) including human fluid.).
Regarding claim 33, NPL Ammad-ud-din as modified by Colley and NPL Yang and applied to claim 1 teaches the limitation identified in bold as “the subject has a cancer selected from the group consisting of a carcinoma, lymphoma, blastoma, glioblastoma, sarcoma, leukemia, breast cancer, squamous cell cancer, lung cancer, small-cell lung cancer, non-small cell lung cancer (NSCLC), adenocarcinoma of the lung, squamous carcinoma of the lung, head and neck cancer, cancer of the peritoneum, hepatocellular cancer, gastric or stomach cancer, pancreatic cancer, ovarian cancer, cervical cancer, liver cancer, bladder cancer, hepatoma, colon cancer, colorectal cancer, endometrial or uterine carcinoma, salivary gland carcinoma, kidney or renal cancer, liver cancer, prostate cancer, vulval cancer, thyroid cancer, hepatic carcinoma, B-cell lymphoma, low grade/follicular non-Hodgkin's lymphoma (NHL), small lymphocytic (SL) NHL, intermediate grade/follicular NHL, intermediate grade diffuse NHL, high grade immunoblastic NHL, high grade lymphoblastic NHL, high grade small non-cleaved cell NHL, bulky disease NHL, mantle cell lymphoma, AIDS-related lymphoma, Waldenstrom's Macroglobulinemia, chronic lymphocytic leukemia (CLL), acute lymphoblastic leukemia (ALL), hairy cell leukemia, and chronic myeloblastic leukemia” (Paragraph [0362] of Colley. In the instant application, the broadest reasonable interpretation of “the subject has a cancer selected from the group consisting of a carcinoma, lymphoma, blastoma, glioblastoma, sarcoma, leukemia, breast cancer, squamous cell cancer, lung cancer, small-cell lung cancer, non-small cell lung cancer (NSCLC), adenocarcinoma of the lung, squamous carcinoma of the lung, head and neck cancer, cancer of the peritoneum, hepatocellular cancer, gastric or stomach cancer, pancreatic cancer, ovarian cancer, cervical cancer, liver cancer, bladder cancer, hepatoma, colon cancer, colorectal cancer, endometrial or uterine carcinoma, salivary gland carcinoma, kidney or renal cancer, liver cancer, prostate cancer, vulval cancer, thyroid cancer, hepatic carcinoma, B-cell lymphoma, low grade/follicular non-Hodgkin's lymphoma (NHL), small lymphocytic (SL) NHL, intermediate grade/follicular NHL, intermediate grade diffuse NHL, high grade immunoblastic NHL, high grade lymphoblastic NHL, high grade small non-cleaved cell NHL, bulky disease NHL, mantle cell lymphoma, AIDS-related lymphoma, Waldenstrom's Macroglobulinemia, chronic lymphocytic leukemia (CLL), acute lymphoblastic leukemia (ALL), hairy cell leukemia, and chronic myeloblastic leukemia” reads on the cancer in Colley (Paragraph [0362]) being one selected from the group consisting of acute lymphocytic cancer, acute myeloid leukemia, alveolar rhabdomyosarcoma, bone cancer, brain cancer, breast cancer (e.g., triple negative breast cancer), cancer of the anus, anal canal, or anorectum, cancer of the eye, cancer of the intrahepatic bile duct, cancer of the joints, cancer of the head or neck, gallbladder, or pleura, cancer of the nose, nasal cavity, or middle ear, cancer of the oral cavity, cancer of the vulva, chronic lymphocytic leukemia, chronic myeloid cancer, colon cancer, esophageal cancer, cervical cancer, gastrointestinal cancer (e.g., gastrointestinal carcinoid tumor), glioblastoma, Hodgkin lymphoma, hypopharynx cancer, hematological malignancy, kidney cancer, larynx cancer, liver cancer, lung cancer (e.g., non-small cell lung cancer (NSCLC), small cell lung cancer (SCLC), bronchioloalveolar carcinoma), malignant mesothelioma, melanoma, multiple myeloma, nasopharynx cancer, non-Hodgkin lymphoma, ovarian cancer, pancreatic cancer, peritoneum, omentum, and mesentery cancer, pharynx cancer, prostate cancer, rectal cancer, renal cancer (e.g., renal cell carcinoma (RCC)), small intestine cancer, soft tissue cancer, stomach cancer, testicular cancer, thyroid cancer, ureter cancer, and urinary bladder cancer.).
Claims 2 – 3 and 30 – 31 are rejected under 35 U.S.C. 103(a) as being unpatentable over NPL Ammad-ud-din as modified by Colley and NPL Yang and applied to claim 1, and further in view of NPL Yuan.
Regarding claim 2, NPL Ammad-ud-din as modified by Colley and NPL Yang and applied to claim 1 teaches the limitation identified in bold as “the predicted effect comprises a prediction for cell death of a cancer cell in the subject in response to administration of the pharmaceutical agent to the subject” (Abstract of NPL Yuan. In the instant application, the broadest reasonable interpretation of “the predicted effect comprises a prediction for cell death of a cancer cell in the subject in response to administration of the pharmaceutical agent to the subject” reads on the measured drug responses in NPL Yuan (Abstract) for cytotoxic and targeted therapies across large collections of genomically and transcriptomically characterized cancer cell lines.)
Regarding claim 3, NPL Ammad-ud-din as modified by Colley and NPL Yang and applied to claim 1 teaches the limitation identified in bold as “the pharmaceutical agent is a chemotherapeutic agent” (Last Paragraph on Page 2 of NPL Yuan. In the instant application, the broadest reasonable interpretation of “the pharmaceutical agent is a chemotherapeutic agent” reads on the CCLE, profiling 24 drugs (mostly targeted agents) and NCI60, with 255 drug (mostly cytotoxic agents) in NPL Yuan (Last Paragraph on Page 2).).
Regarding claim 30, NPL Ammad-ud-din as modified by Colley and NPL Yang and applied to claim 1 teaches the limitation identified in bold as “the biological sample of the subject comprises a diseased tissue of the subject” (Abstract of NPL Yuan. In the instant application, the broadest reasonable interpretation of “the biological sample of the subject comprises a diseased tissue of the subject” reads on the cancer cell lines in NPL Yuan (Abstract) for the individual patient.).
Regarding claim 31, NPL Ammad-ud-din as modified by Colley and NPL Yang and applied to claim 30 teaches the limitation identified in bold as “the diseased tissue of the subject is a cancerous tissue” (Abstract of NPL Yuan. In the instant application, the broadest reasonable interpretation of “the diseased tissue of the subject is a cancerous tissue” reads on the cancer cell lines in NPL Yuan (Abstract) for the individual patient.).
Therefore, it would have been obvious to one of ordinary skill in the art of computer-aided diagnosis in oncology at the time of filing to modify the method of NPL Ammad-ud-din to implement the predicted effect comprises a prediction for cell death of a cancer cell in the subject in response to administration of the pharmaceutical agent to the subject, as taught by NPL Yuan (Abstract), implement the pharmaceutical agent being a chemotherapeutic agent, as taught by NPL Yuan (Last Paragraph on Page 2), implement the biological sample of the subject comprising a diseased tissue of the subject, as taught by NPL Yuan (Abstract), and implement the diseased tissue of the subject being a cancerous tissue, as taught by NPL Yuan (Abstract) in order to individualize treatment by selecting therapeutics that are most likely to be effective given the molecular profile of a patient’s tumor (First Paragraph on Page 1 of NPL Yuan).
Claim 5 is rejected under 35 U.S.C. 103(a) as being unpatentable over NPL Ammad-ud-din as modified by Colley and NPL Yang and applied to claim 1, and further in view of Rapaka (U.S. Pub. No. 2018/0315182 A1).
Regarding claim 5, NPL Ammad-ud-din as modified by Colley and NPL Yang and applied to claim 1 teaches the limitation identified in bold as “the multi-task model comprises a partially connected neural network defining a plurality of tasks, each respective task in the plurality of tasks corresponding to a respective output in the plurality of outputs, wherein the partially-connected neural network comprises (a) a first set of layers shared between the plurality of tasks and (b) for each respective task in the plurality of tasks a corresponding second set of layers unique to the respective task” (In the instant application, the broadest reasonable interpretation of “a respective output in the plurality of outputs” reads on the new patients’ predicted drug responses in NPL Ammad-ud-din (Second Paragraph on Page 12 and Third Paragraph to Fourth Paragraph on Page 34) and the drug response matrix in NPL Ammad-ud-din (Second Paragraph on Page 42) for modelling responses to multiple drugs over multiple cells having diverse cancer types, as applied to claim 1.).
NPL Ammad-ud-din as modified by Colley and NPL Yang and applied to claim 1 does not appear to explicitly disclose, but Rapaka teaches the limitation identified in bold as “the multi-task model comprises a partially connected neural network defining a plurality of tasks, each respective task in the plurality of tasks corresponding to a respective output in the plurality of outputs, wherein the partially-connected neural network comprises (a) a first set of layers shared between the plurality of tasks and (b) for each respective task in the plurality of tasks a corresponding second set of layers unique to the respective task” (Paragraph [0044] of Rapaka. In the instant application, the broadest reasonable interpretation of “the multi- task model comprises a partially connected neural network defining a plurality of tasks, wherein the partially-connected neural network comprises (a) a first set of layers shared between the plurality of tasks and (b) for each respective task in the plurality of tasks a corresponding second set of layers unique to the respective task” reads on the multi-task neural network in Rapaka (Paragraph [0044]) having a pool of shared layers to determine common features to the task at hand and additional layers that are trained for specific tasks, i.e., each one of the additional layers having a unique associated task.).
Therefore, it would have been obvious to one of ordinary skill in the art of computer-aided diagnosis in oncology at the time of filing to modify the method of NPL Ammad-ud-din to include the multi-task model comprising a partially connected neural network defining a plurality of tasks, each respective task in the plurality of tasks corresponding to a respective output in the plurality of outputs, wherein the partially-connected neural network comprises (a) a first set of layers shared between the plurality of tasks and (b) for each respective task in the plurality of tasks a corresponding second set of layers unique to the respective task, as taught by Rapaka (Paragraph [0044]) in order to provide fast patient assessment and outcome analysis, manage large quantities of heterogeneous data, provide consistent predictions in an automated manner, provide a machine learning algorithm having superior predictive capabilities in complex tasks, show expert-level performance, provide and a comprehensive patient assessment model combining all available information from the patient to present an integrated understanding of the patient state and enable the clinician to guide therapy (Paragraph [0014] of Rapaka).
Claims 6 – 8 and 20 are rejected under 35 U.S.C. 103(a) as being unpatentable over NPL Ammad-ud-din as modified by Colley and NPL Yang and applied to claim 1, and further in view of Larsen (U.S. Pub. No. 2021/0172931 A1).
Regarding claim 6, NPL Ammad-ud-din as modified by Colley and NPL Yang and applied to claim 1 does not appear to explicitly disclose, but Larsen teaches the limitation identified in bold as:
when the predicted effect of the pharmaceutical agent in the test subject satisfies a first set of one or more criterion, recommending a first therapy that comprises administration of the pharmaceutical agent to the subject (Paragraph [0158] of Larsen. In the instant application, the broadest reasonable interpretation of “when the predicted effect of the pharmaceutical agent in the test subject satisfies a first set of one or more criterion, recommending a first therapy that comprises administration of the pharmaceutical agent to the subject” reads on the activity in Larsen (Paragraph [0158]) of communicating a recommended therapy comprising the therapeutic agent to the test subject when the probability or likelihood that the cancer will be sensitive to the therapeutic agent satisfies a threshold likelihood.); and
when the predicted effect of the pharmaceutical agent in the test subject does not satisfy the first set of one or more criterion, recommending a second therapy that is different from the first therapy (Paragraph [0158] of Larsen. In the instant application, the broadest reasonable interpretation of “when the predicted effect of the pharmaceutical agent in the test subject does not satisfy the first set of one or more criterion, recommending a second therapy that is different from the first therapy” reads on the activity in Larsen (Paragraph [0158]) of communicating a recommended therapy that does not include the therapeutic agent to the test subject when the probability or likelihood that the cancer will be sensitive to the therapeutic agent does not satisfy a threshold likelihood.).
Regarding claim 7, NPL Ammad-ud-din as modified by Colley and NPL Yang and applied to claim 1 does not appear to explicitly disclose, but Larsen teaches the limitation identified in bold as:
when the predicted effect of the pharmaceutical agent in the test subject satisfies a first set of one or more criterion, administering a first therapy to the subject, wherein the first therapy that comprises administration of the pharmaceutical agent (Paragraph [0158] of Larsen. In the instant application, the broadest reasonable interpretation of “when the predicted effect of the pharmaceutical agent in the test subject satisfies a first set of one or more criterion, administering a first therapy to the subject, wherein the first therapy that comprises administration of the pharmaceutical agent” reads on the activity in Larsen (Paragraph [0158]) of administering a recommended therapy comprising the therapeutic agent to the test subject when the probability or likelihood that the cancer will be sensitive to the therapeutic agent satisfies a threshold likelihood.); and
when the predicted effect of the pharmaceutical agent in the test subject does not satisfy the first set of one or more criterion, administering a second therapy to the subject, wherein the second therapy is different from the first therapy (Paragraph [0158] of Larsen. In the instant application, the broadest reasonable interpretation of “when the predicted effect of the pharmaceutical agent in the test subject does not satisfy the first set of one or more criterion, administering a second therapy to the subject, wherein the second therapy is different from the first therapy” reads on the activity in Larsen Paragraph [0158]) of administering a recommended therapy that does not include the therapeutic agent to the test subject when the probability or likelihood that the cancer will be sensitive to the therapeutic agent does not satisfy a threshold likelihood.).
Regarding claim 8, NPL Ammad-ud-din as modified by Colley and NPL Yang and applied to claim 1 does not appear to explicitly disclose, but Larsen teaches the limitation identified in bold as:
when the predicted effect of the pharmaceutical agent in the test subject satisfies a first set of one or more criterion, recommending the test subject for a clinical trial of the pharmaceutical agent to the subject (Paragraphs [0158] and [0175] of Larsen. In the instant application, the broadest reasonable interpretation of “when the predicted effect of the pharmaceutical agent in the test subject satisfies a first set of one or more criterion, recommending the test subject for a clinical trial of the pharmaceutical agent to the subject” reads on the activity in Larsen (Paragraphs [0158] and [0175]) of determining whether the cancer patient is eligible for the clinical trial based on at least the measured fitness of the cells in the one or more tumor organoids. The Office has determined that it was well known in the art of machine learning in oncology at the time of filing that a cancer patient is recommended for a clinical trial after: (i) that cancer patient has undergone a tumor biopsy; (ii) a tumor organoid has been cultured from the tumor biopsy; (iii) a reduced fitness of the tumor organoid following exposure to the candidate cancer pharmaceutical agent has been measured; and (iv) the cancer patient was determined to be eligible for a clinical test.); and
when the predicted effect of the pharmaceutical agent in the test subject does not satisfy the first set of one or more criterion, not recommending the test subject for the clinical trial (Paragraphs [0158] and [0175] of Larsen. In the instant application, the broadest reasonable interpretation of “when the predicted effect of the pharmaceutical agent in the test subject does not satisfy the first set of one or more criterion, not recommending the test subject for the clinical trial” reads on the activity in Larsen (Paragraphs [0158] and [0175]) of determining whether the cancer patient is eligible for the clinical trial based on at least the measured fitness of the cells in the one or more tumor organoids. The Office has determined that it was well known in the art of machine learning in oncology at the time of filing that a cancer patient is not recommended for a clinical trial after: (i) that cancer patient has undergone a tumor biopsy; (ii) a tumor organoid has been cultured from the tumor biopsy; (iii) a reduced fitness of the tumor organoid following exposure to the candidate cancer pharmaceutical agent has not been measured; and (iv) the cancer patient was not determined to be eligible for a clinical test.).
Regarding claim 20, NPL Ammad-ud-din as modified by Colley and NPL Yang and applied to claim 1 does not appear to explicitly disclose, but Larsen teaches the limitation identified in bold as “the plurality of tissue organoids comprises a plurality of tumor organoids” (Paragraphs [0115] and [0156] of Larsen. In the instant application, the broadest reasonable interpretation of “the plurality of tissue organoids comprises a plurality of tumor organoids” reads on the tissue samples in Larsen (Paragraphs [0115] and [0156]) obtained from the patients (e.g., normal tissue) and cultured (e.g., in organoid culture).).
Therefore, it would have been obvious to one of ordinary skill in the art of computer-aided diagnosis in oncology at the time of filing to modify the method of NPL Ammad-ud-din to: include the steps of (i) when the predicted effect of the pharmaceutical agent in the test subject satisfies a first set of one or more criterion, recommending a first therapy that comprises administration of the pharmaceutical agent to the subject, and (ii) when the predicted effect of the pharmaceutical agent in the test subject does not satisfy the first set of one or more criterion, recommending a second therapy that is different from the first therapy, as taught by Larsen (Paragraph [0158]), include the steps of (i) when the predicted effect of the pharmaceutical agent in the test subject satisfies a first set of one or more criterion, administering a first therapy to the subject, wherein the first therapy that comprises administration of the pharmaceutical agent, and (ii) when the predicted effect of the pharmaceutical agent in the test subject does not satisfy the first set of one or more criterion, administering a second therapy to the subject, wherein the second therapy is different from the first therapy, as taught by Larsen (Paragraph [0158]), and include the activities of (i) when the predicted effect of the pharmaceutical agent in the test subject satisfies a first set of one or more criterion, recommending the test subject for a clinical trial of the pharmaceutical agent to the subject, and (ii) when the predicted effect of the pharmaceutical agent in the test subject does not satisfy the first set of one or more criterion, not recommending the test subject for the clinical trial, as taught by Larsen (Paragraphs [0158] and [0175]), and implement the plurality of tissue organoids comprises a plurality of tumor organoids, as taught by Larsen (Paragraphs [0115] and [0156]) in order to provide systems and methods for using organoid cultures to improve treatment predictions and outcomes, assessing the effectiveness of various drugs on one or more tumor organoid lines, and determine if a specific drug may be useful in killing cancer cells with specific genetic mutations or phenotypes (Paragraph [0007] of Larsen).
Claim 9 is rejected under 35 U.S.C. 103(a) as being unpatentable over NPL Ding in view of Colley, NPL Yuan, NPL Yang, and NPL Ammad-ud-din.
Regarding claim 9, NPL Ding teaches the limitations of representative claim 9 identified in bold as:
A method for identifying one or more tissue organoids in a plurality of tissue organoids matching a biological property of a tissue in a subject (Fourth Paragraph in First Column on Page 905 of NPL Ding. In the instant application, the broadest reasonable interpretation of “method for identifying one or more tissue organoids in a plurality of tissue organoids matching a biological property of a tissue in a subject” reads on the platform in NPL Ding (Fourth Paragraph in First Column on Page 905) for generating patient-derived micro-organospheres (MOS) to facilitate personalized medicine treatment. The Office has determined that MOSs are a specific technological application of the organoid but designed to be smaller and quicker to create.), the method comprising:
at a computer system comprising at least one processor and a memory storing at least one program for execution by the at least one processor:
inputting information about the test subject into a multi-task model comprising a plurality of parameters and one or more hidden layers, wherein the multi-task model is trained to apply the plurality of parameters to the information about the test subject through a plurality of instructions to generate, as output from the multi-task model, a plurality of outputs comprising (i) a predicted effect of the pharmaceutical agent in the test subject and (ii) for each respective cell type variable in a set of one or more cell type variables, a corresponding cell type classification, wherein the information about the test subject comprises, for each respective cellular constituent in a plurality of cellular constituents, a corresponding abundance value for the abundance of the respective cellular constituent in a biological sample of the test subject;
obtaining a latent representation of the information about the test subject from a respective hidden layer in the one or more hidden layers;
comparing the latent representation of the information about the test subject to a plurality of latent representations, wherein each respective latent representation in the plurality of latent representations is of information about a respective tissue organoid, in a plurality of tissue organoids, obtained from the multi-task model (First Paragraph in Second Column on Page 909 of NPL Ding. In the instant application, the broadest reasonable interpretation of “comparing … the information about the test subject to … information about a respective tissue organoid, in a plurality of tissue organoids” reads on the activities in NPL Ding (First Paragraph in Second Column on Page 909) comparing the primary tissue and the tissue (i.e., the tumor and stromal cell types for each of the primary tissue and the tissue).); and
identifying one or more respective tissue organoids, in the plurality of tissue organoids, that satisfy a set of one or more similarity criterion based on the comparing, thereby identifying the one or more tissue organoids matching the biological property of the tissue in the subject (First Paragraph in Second Column on Page 909 of NPL Ding. In the instant application, the broadest reasonable interpretation of “identifying one or more respective tissue organoids, in the plurality of tissue organoids, that satisfy a set of one or more similarity criterion based on the comparing, thereby identifying the one or more tissue organoids matching the biological property of the tissue in the subject” reads on the pseudobulk analysis in NPL Ding (First Paragraph in Second Column on Page 909) determining that the overall gene expression for the cell population of the primary tissue is comparable to the overall gene expression for the cell population of the MOS.).
NPL Ding does not appear to explicitly disclose, but Colley teaches the limitation identified in bold as “a computer system comprising at least one processor and a memory storing at least one program for execution by the at least one processor” (Paragraph [0379] of Colley. In the instant application, the broadest reasonable interpretation of “a computer system comprising at least one processor and a memory storing at least one program for execution by the at least one processor” reads on the computer in Colley (Paragraph [0379]) including the computer system having one or more processors and memory storing one or more programs for execution by the one or more processors.).
NPL Ding does not appear to explicitly disclose, but NPL Yuan teaches the limitation identified in bold as “inputting information about the test subject into a multi-task model comprising a plurality of parameters and one or more hidden layers, wherein the multi-task model is trained to apply the plurality of parameters to the information about the test subject through a plurality of instructions to generate, as output from the multi-task model, a plurality of outputs comprising (i) a predicted effect of the pharmaceutical agent in the test subject and (ii) for each respective cell type variable in a set of one or more cell type variables, a corresponding cell type classification, wherein the information about the test subject comprises, for each respective cellular constituent in a plurality of cellular constituents, a corresponding abundance value for the abundance of the respective cellular constituent in a biological sample of the test subject” (First Paragraph and Third Paragraph on Page 1, Third Paragraph on Page 6, and Fourth Paragraph to Fifth Paragraph on Page 8 of NPL Yuan. In the instant application, the broadest reasonable interpretation of “wherein the information about the test subject comprises a plurality of abundance values, the plurality of abundance values comprising, for each respective cellular constituent in a plurality of cellular constituents, a corresponding abundance value for the abundance of the respective cellular constituent in a biological sample of the test subject” reads on the molecular profile in NPL Yuan (First Paragraph and Third Paragraph on Page 1) of a patient’s tumor having the molecular feature space including mutation calls, gene copy number, and gene expression levels. The broadest reasonable interpretation of “inputting information about the test subject into a multi-task model comprising a plurality of parameters, wherein the multi-task model applies the plurality of parameters to the information about the test subject through a plurality of instructions to generate, as output from the multi-task model, a plurality of outputs comprising (i) a predicted effect of the pharmaceutical agent in the test subject” reads on a lower dimensional subspace in NPL Yuan (Third Paragraph on Page 6 and Fourth Paragraph to Fifth Paragraph on Page 8) of the molecular feature space for the patient’s tumor, being spanned or inputted into the multitask learning model for all drugs and the prediction of each drug response.).
NPL Ding does not appear to explicitly disclose, but Pan teaches the limitation identified in bold as “for each respective cell type variable in a set of one or more cell type variables, a corresponding cell type classification” (Eighth Paragraph on Page 7 of NPL Yang. In the instant application, the broadest reasonable interpretation of “for each respective cell type variable in a set of one or more cell type variables, a corresponding cell type classification” reads on the cancer types, tissue types, and tissue sites in NPL Yang (Eighth Paragraph on Page 7).).
NPL Ding does not appear to explicitly disclose, but NPL Ammad-ud-din teaches the limitation identified in bold as “obtaining a latent representation of the information about the test subject from a respective hidden layer in the one or more hidden layers” (Abstract, Second Paragraph on Page 12; First Paragraph on Page 24; First Paragraph to Second Paragraph on Page 32 of NPL Ammad-ud-din. In the instant application, the broadest reasonable interpretation of “obtaining a latent representation of the information about the test subject from a respective hidden layer in the one or more hidden layers” reads on the activity in NPL Ammad-ud-din (Abstract, Second Paragraph on Page 12; First Paragraph on Page 24; First Paragraph to Second Paragraph on Page 32) of latent variable modeling dependencies (i.e., the linear mapping function) between the high-dimensional data variables and the low-dimensional variables (i.e., the latent components presenting a concise and denoised summary). The high-dimensional observed data includes the new patient’s unseen cell line including the number of RNA copies present in a cell, the correlated abundance of the corresponding proteins produced in the cell, and the nucleic acid sequences produced via RNA sequencing).).
NPL Ding does not appear to explicitly disclose, but NPL Ammad-ud-din teaches the limitation identified in bold as “comparing the latent representation of the information about the test subject to a plurality of latent representations, wherein each respective latent representation in the plurality of latent representations is of information about a respective tissue organoid, in a plurality of tissue organoids, obtained from the multi-task model” (Second Paragraph on Page 12, First Paragraph to Second Paragraph on Page 32, and Third Paragraph to Fourth Paragraph on Page 34 of NPL Ammad-ud-din. In the instant application, the broadest reasonable interpretation of “the latent representation … a plurality of latent representations, wherein each respective latent representation in the plurality of latent representations … the multi-task model” reads on the multi-task learning (MTL) in NPL Ammad-ud-din (Second Paragraph on Page 12, First Paragraph to Second Paragraph on Page 32, and Third Paragraph to Fourth Paragraph on Page 34) leading to a better model for all the tasks including latent variable modeling and matrix factorization to decompose high-dimensional observed data into multiple low-dimensional latent factors.).
Therefore, it would have been obvious to one of ordinary skill in the art of computer-aided diagnosis in oncology at the time of filing to modify the method of NPL Ammad-ud-din to: include the computer system comprising at least one processor and a memory storing at least one program for execution by the at least one processor, as taught by Colley (Paragraph [0379]) in order to allow for precision-level analysis of patient health needs, in order to provide the right resources, at the right time, to the right patients (Paragraph [0120] of Colley); include the step of inputting information about the test subject into a multi-task model comprising a plurality of parameters and one or more hidden layers, wherein the multi-task model is trained to apply the plurality of parameters to the information about the test subject through a plurality of instructions to generate, as output from the multi-task model, a plurality of outputs comprising (i) a predicted effect of the pharmaceutical agent in the test subject … wherein the information about the test subject comprises, for each respective cellular constituent in a plurality of cellular constituents, a corresponding abundance value for the abundance of the respective cellular constituent in a biological sample of the test subject, as taught by NPL Yuan (First Paragraph and Third Paragraph on Page 1, Third Paragraph on Page 6, and Fourth Paragraph to Fifth Paragraph on Page 8 of NPL Yuan) in order to individualize treatment by selecting therapeutics that are most likely to be effective given the molecular profile of a patient’s tumor (First Paragraph on Page 1 of NPL Yuan); implement, for each respective cell type variable in a set of one or more cell type variables, a corresponding cell type classification, as taught by NPL Yang (Eighth Paragraph on Page 7) in order to leverage the power of multitask learning to provide novel insights into the molecular underpinnings of drug response (Last Paragraph on Page 1 of NPL Yang); include the step of obtaining a latent representation of the information about the test subject from a respective hidden layer in the one or more hidden layers, and include the latent representation in the plurality of latent representations and the multi-task model, as taught by NPL Ammad-ud-din (Abstract, Second Paragraph on Page 12; First Paragraph on Page 24; First Paragraph to Second Paragraph on Page 32) in order to improve the accuracy of predicting drug responses in cancer cells (Second Paragraph on Page 13 of NPL Ammad-ud-din)
Claims 26 – 28 are rejected under 35 U.S.C. 103(a) as being unpatentable over NPL Yuan as modified by Colley and NPL Yang and applied to claim 1, and further in view of Garraway (U.S. Pub. No. 2020/0390786 A1).
Regarding claim 26, NPL Ammad-ud-din as modified by Colley and NPL Yang and applied to claim 1 teaches the limitation identified in bold as “each respective cellular constituent in the plurality of cellular constituents is a different mRNA species” (Paragraph [0087] of Garraway. In the instant application, the broadest reasonable interpretation of “each respective cellular constituent in the plurality of cellular constituents is a different mRNA species” reads on the patient tumor sample in Garraway (Paragraph [0087]) being assayed to generate a genomic expression profile for determining a level of a nucleic acid sequence, e.g., quantitative hybridization of microRNA, labeled microRNA, amplified microRNA, complementary/synthetic DNA (cDNA), etc., quantitative polymerase chain reaction (PCR), and ELISA for quantitation, and allow the analysis of differential gene expression between two samples.).
Therefore, it would have been obvious to one of ordinary skill in the art of computer-aided diagnosis in oncology at the time of filing to modify the method of NPL Ammad-ud-din to implement the feature of each respective cellular constituent in the plurality of cellular constituents is a different mRNA species, as taught by Garraway (Paragraph [0087]) in order to develop methods to identify genetic profiles of treatment-resistant ovarian cancer to identify therapeutic targets for recurrent disease (Paragraph [0003] of Garraway).
Regarding claim 27, NPL Ammad-ud-din as modified by Colley and Garraway and applied to claim 26 teaches the limitation identified in bold as:
obtaining, in electronic form, a plurality of nucleic acid sequences for mRNA from the biological sample of the test subject (Paragraph [0087] of Garraway. In the instant application, the broadest reasonable interpretation of “obtaining, in electronic form, a plurality of nucleic acid sequences for mRNA from the biological sample of the test subject” reads on the patient tumor sample in Garraway (Paragraph [0087]) being assayed to generate a genomic expression profile for determining a level of a nucleic acid sequence.); and
determining, for each respective cellular constituent in the plurality of cellular constituents, the corresponding abundance value from the plurality of nucleic acid sequences (Paragraph [0087] of Garraway. In the instant application, the broadest reasonable interpretation of “determining, for each respective cellular constituent in the plurality of cellular constituents, the corresponding abundance value from the plurality of nucleic acid sequences” reads on the activity in Garraway (Paragraph [0087]) of generating an expression profile, i.e., means measuring the relative abundance of the nucleic acid sequences in the measured samples .).
Regarding claim 28, NPL Ammad-ud-din as modified by Colley and Garraway and applied to claim 27 teaches the limitation identified in bold as “the obtaining (i) comprises sequencing the mRNA from the biological sample of the test subject, thereby obtaining the plurality of nucleic acid sequences” (Paragraph [0087] of Garraway. In the instant application, the broadest reasonable interpretation of “the obtaining (i) comprises sequencing the mRNA from the biological sample of the test subject, thereby obtaining the plurality of nucleic acid sequences” reads on the patient tumor sample in Garraway (Paragraph [0087]) being assayed to generate a genomic expression profile for measuring the relative abundance of the nucleic acid sequences in the measured samples.).
Claim 34 is rejected under 35 U.S.C. 103(a) as being unpatentable over NPL Ammad-ud-din in view of Colley, NPL Yang, NPL Ding, and Larsen.
Regarding claim 34, NPL Ammad-ud-din teaches the limitations identified in bold as:
A method for training a model to predict an effect of a candidate pharmaceutical agent in a test subject of a first species (First Paragraph on Page 11 to First Paragraph on Page 12 of NPL Ammad-ud-din. In the instant application, the broadest reasonable interpretation of “a method for training a model to predict an effect of a candidate pharmaceutical agent in a test subject of a first species” reads on the method in NPL Ammad-ud-din (First Paragraph on Page 11 to First Paragraph on Page 12) for predicting the response of a new patient’s sample to a drug and identifying features predictive of drug responses to develop computational models that provide better predictions for the individual patient.), the method comprising:
at a computer system comprising at least one processor and a memory storing at least one program for execution by the at least one processor:
obtaining, for each respective training sample in a first plurality of training samples, wherein each respective training sample in the first plurality of training samples comprises a tissue organoid or tissue organoid culture, formed from cells of the first species, that has been exposed to a perturbation (First Paragraph on Page 23, and Figure 3.1 of NPL Ammad-ud-din. In the instant application, the broadest reasonable interpretation of “obtaining, for each respective training sample in a first plurality of training samples, wherein each respective training sample in the first plurality of training samples comprises a tissue organoid or tissue organoid culture, formed from cells of the first species, that has been exposed to a perturbation” reads on the activity in NPL Ammad-ud-din (First Paragraph on Page 23 and Figure 3.1.) of obtaining cells from cancer patients, cultured in a laboratory, and exposed to an anti-cancer drug.):
a corresponding plurality of abundance values comprising, for each respective cellular constituent in a plurality of cellular constituents, a corresponding abundance value for the abundance of the respective cellular constituent in the respective training sample after exposure to the candidate pharmaceutical agent (First Paragraph on Page 24 of NPL Ammad-ud-din. In the instant application, the broadest reasonable interpretation of “a corresponding plurality of abundance values comprising, for each respective cellular constituent in a plurality of cellular constituents, a corresponding abundance value for the abundance of the respective cellular constituent in the respective training sample after exposure to the candidate pharmaceutical agent” reads on the genomic and transcriptomic profiles in NPL Ammad-ud-din (First Paragraph on Page 24) for each respective tissue organoid culture exposed to the anti-cancer drug, including the number of RNA copies present in a cell correlated with the abundance of the corresponding proteins produced in the cell and nucleic acid sequences determined via RNA sequencing.),
a corresponding experimentally measured effect of the candidate pharmaceutical agent on the respective training sample (First Paragraph on Page 23 and Figure 3.1. In the instant application, the broadest reasonable interpretation of “a corresponding experimentally measured effect of the candidate pharmaceutical agent on the respective training sample” reads on the corresponding growth in NPL Ammad-ud-din (First Paragraph on Page 23 and Figure 3.1) of the cells, for each respective tissue organoid culture, after being exposed to the anti-cancer drug.), and
a corresponding set of one or more cell type classifications comprising, for each respective cell type variable in a set of one or more cell type variables, a corresponding cell type classification for the respective cell type variable;
obtaining, for each respective training sample in a second plurality of training samples, wherein each respective training sample in the second plurality of training samples comprises a biological sample from a respective subject in a plurality of subjects of the first species (First Paragraph on Page 11, First Paragraph on Page 23, and Figure 3.1 of NPL Ammad-ud-din. In the instant application, the broadest reasonable interpretation of “obtaining, for each respective training sample in a second plurality of training samples, wherein each respective training sample in the second plurality of training samples comprises a biological sample from a respective subject in a plurality of subjects of the first species” reads on the activity in NPL Ammad-ud-din (First Paragraph on Page 23 and Figure 3.1.) of obtaining new patient samples from cancer patients.):
a corresponding plurality of abundance values comprising, for each respective cellular constituent in the plurality of cellular constituents, a corresponding abundance value for the abundance of the respective cellular constituent in the respective training sample (First Paragraph on Page 24 of NPL Ammad-ud-din. In the instant application, the broadest reasonable interpretation of “a corresponding plurality of abundance values comprising, for each respective cellular constituent in the plurality of cellular constituents, a corresponding abundance value for the abundance of the respective cellular constituent in the respective training sample” reads on the genomic and transcriptomic profiles in NPL Ammad-ud-din (First Paragraph on Page 24) for each respective new patient sample, including the number of RNA copies present in a cell correlated with the abundance of the corresponding proteins produced in the cell and nucleic acid sequences determined via RNA sequencing.), and
a corresponding set of one or more cell type classifications comprising, for each respective cell type variable in the set of one or more cell type variables, a corresponding cell type classification for the respective cell type variable;
performing a first dimensionality reduction analysis across the corresponding plurality of abundance values for each respective training sample in the first plurality of training samples (First Paragraph on Page 24; First Paragraph on Page 32 of NPL Ammad-ud-din. In the instant application, the broadest reasonable interpretation of “the multi- task model comprises a linear mapping function that transforms the plurality of abundance values into a first latent feature space comprising fewer dimensions than the number of respective cellular constituents in the plurality of cellular constituents” reads on the activity in NPL Ammad-ud-din (First Paragraph on Page 24; First Paragraph to Second Paragraph on Page 32) of performing, for each respective tissue organoid culture, dimensionality reduction to decompose the high-dimensional observed data into multiple low-dimensional latent factors, e.g., performing latent variable modeling for the high-dimensional data variables, such as the number of RNA copies present in a cell, the correlated abundance of the corresponding proteins produced in the cell, the nucleic acid sequences produced via RNA sequencing).), thereby:
learning a first mapping function that maps a corresponding plurality of abundance values comprising, for each respective cellular constituent in the plurality of cellular constituents, a corresponding abundance value for the abundance of the respective cellular constituent in a tissue organoid or tissue organoid culture, formed from cells of the first species, that has been exposed to a perturbation, into a first latent feature space comprising a first plurality of dimensions that is less than the number of cellular constituents in the plurality of constituents (First Paragraph on Page 24; First Paragraph on Page 32 of NPL Ammad-ud-din. In the instant application, the broadest reasonable interpretation of “learning a first mapping function that maps a corresponding plurality of abundance values comprising, for each respective cellular constituent in the plurality of cellular constituents, a corresponding abundance value for the abundance of the respective cellular constituent in a tissue organoid or tissue organoid culture, formed from cells of the first species, that has been exposed to a perturbation, into a first latent feature space comprising a first plurality of dimensions that is less than the number of cellular constituents in the plurality of constituents” reads on the activity in NPL Ammad-ud-din (First Paragraph on Page 24; First Paragraph to Second Paragraph on Page 32) of latent variable modeling, for each respective tissue organoid culture, dependencies (i.e., the linear mapping function) between the high-dimensional data components and the low-dimensional components (i.e., the low-dimensional components being latent components presenting a concise and denoised summary of the high-dimensional components). These dependencies are used for dimensionality reduction analysis in combination with matrix factorization, where the high-dimensional observed data is decomposed into multiple low-dimensional latent components (typically much smaller than the observed data dimensionality), which includes: the number of RNA copies present in a cell, the correlated abundance of the corresponding proteins produced in the cell, the nucleic acid sequences produced via RNA sequencing).), and
generating, for each respective training sample f in the first plurality of training samples, a first corresponding representation of the corresponding plurality of abundance values in the first latent feature space according to the first mapping function (Third Paragraph on Page 32 of NPL Ammad-ud-din. In the instant application, the broadest reasonable interpretation of “generating, for each respective training sample f in the first plurality of training samples, a first corresponding representation of the corresponding plurality of abundance values in the first latent feature space according to the first mapping function” reads on the activity in NPL Ammad-ud-din (Third Paragraph on Page 32 of NPL Ammad-ud-din) of modeling, for each respective tissue organoid culture, a single matrix of low-dimensional latent factors (including: the number of RNA copies present in a cell, the correlated abundance of the corresponding proteins produced in the cell, the nucleic acid sequences produced via RNA sequencing) in the low-dimensional components according to the dependency for dimensionality reduction.);
performing a second dimensionality reduction analysis across the corresponding plurality of abundance values for each respective training sample in the second plurality of training samples (First Paragraph on Page 24; First Paragraph on Page 32 of NPL Ammad-ud-din. In the instant application, the broadest reasonable interpretation of “the multi- task model comprises a linear mapping function that transforms the plurality of abundance values into a first latent feature space comprising fewer dimensions than the number of respective cellular constituents in the plurality of cellular constituents” reads on the activity in NPL Ammad-ud-din (First Paragraph on Page 24; First Paragraph to Second Paragraph on Page 32) of performing, for each respective new patient sample, dimensionality reduction to decompose the high-dimensional observed data into multiple low-dimensional latent factors, e.g., performing latent variable modeling for the high-dimensional data variables, such as the number of RNA copies present in a cell, the correlated abundance of the corresponding proteins produced in the cell, the nucleic acid sequences produced via RNA sequencing).), thereby:
learning a second mapping function that maps a corresponding plurality of abundance values comprising, for each respective cellular constituent in the plurality of cellular constituents, a corresponding abundance value for the abundance of the respective cellular constituent in a biological sample of the first species, into a second latent feature space comprising the first plurality of dimensions (First Paragraph on Page 24; First Paragraph to Second Paragraph on Page 32 of NPL Ammad-ud-din. In the instant application, the broadest reasonable interpretation of “learning a second mapping function that maps a corresponding plurality of abundance values comprising, for each respective cellular constituent in the plurality of cellular constituents, a corresponding abundance value for the abundance of the respective cellular constituent in a biological sample of the first species, into a second latent feature space comprising the first plurality of dimensions” reads on the activity in NPL Ammad-ud-din (First Paragraph on Page 24; First Paragraph to Second Paragraph on Page 32) of latent variable modeling the dependency (i.e., the linear mapping function) that maps the number of RNA copies present in a cell, the correlated abundance of the corresponding proteins produced in the cell, and the nucleic acid sequences for the new patient sample, into the low-dimensional latent component.).), and
generating, for each respective training sample in the second plurality of training samples, a corresponding representation of the corresponding plurality of abundance values in the second latent feature space according to the second mapping function (Third Paragraph on Page 32 of NPL Ammad-ud-din. In the instant application, the broadest reasonable interpretation of “generating, for each respective training sample [in] the first plurality of training samples, a first corresponding representation of the corresponding plurality of abundance values in the first latent feature space according to the first mapping function” reads on the activity in NPL Ammad-ud-din (Third Paragraph on Page 32 of NPL Ammad-ud-din) of modeling, for the respective new patient sample, a matrix of low-dimensional latent factors (including: the number of RNA copies present in a cell, the correlated abundance of the corresponding proteins produced in the cell, the nucleic acid sequences produced via RNA sequencing) in the low-dimensional components according to the dependency for dimensionality reduction.);
learning a third mapping function that maps a representation of a corresponding plurality of abundance values in the first latent feature space to the second latent feature space;
generating, for each respective training sample in the first plurality of training samples, a second corresponding representation of the corresponding plurality of abundance values in the second latent feature space according to the third mapping function (Third Paragraph on Page 32 of NPL Ammad-ud-din. In the instant application, the broadest reasonable interpretation of “a second corresponding representation … in the second latent feature space” reads on the matrix in NPL Ammad-ud-din (Third Paragraph on Page 32 of NPL Ammad-ud-din) of low-dimensional latent components.);
inputting, for each respective training sample in the first plurality of training samples, corresponding information about the respective training sample into a multi-task model comprising a plurality of parameters, wherein the multi-task model applies the plurality of parameters to the information about the training [subject] through a plurality of instructions to generate, as output from the multi-task model, a corresponding plurality of outputs (Second Paragraph on Page 12 and Third Paragraph to Fourth Paragraph on Page 34 of NPL Ammad-ud-din. In the instant application, the broadest reasonable interpretation of “inputting, for each respective training sample in the first plurality of training samples, corresponding information about the respective training sample into a multi-task model comprising a plurality of parameters, wherein the multi-task model applies the plurality of parameters to the information about the training [subject] through a plurality of instructions to generate, as output from the multi-task model, a corresponding plurality of outputs” reads on the activity in NPL Ammad-ud-din (Second Paragraph on Page 12 and Third Paragraph to Fourth Paragraph on Page 34) of inputting, for each respective tissue organoid culture, the genomic-wide features of the tumor organoid culture into the multi-task learning (MTL) model, wherein the MTL model applies the shared parameters (e.g., representations) to the genomic-wide features of the tumor organoid culture to generate a corresponding predicted drug response in a new patient.), wherein:
the corresponding plurality of outputs comprises (i) a predicted effect of the candidate pharmaceutical agent on the respective training sample (Second Paragraph on Page 12 and Third Paragraph to Fourth Paragraph on Page 34 of NPL Ammad-ud-din. In the instant application, the broadest reasonable interpretation of “the corresponding plurality of outputs comprises (i) a predicted effect of the candidate pharmaceutical agent on the respective training sample” reads on the predicted drug response in NPL Ammad-ud-din (Second Paragraph on Page 12 and Third Paragraph to Fourth Paragraph on Page 34) in the respective new patient sample.) and (ii) for each respective cell type variable in the set of one or more cell type variables, a corresponding cell type classification for the respective training sample, and
the information about the respective training sample comprises the second corresponding representation of the corresponding plurality of abundance values in the second latent feature space [according to the third mapping function] (Paragraphs [0440] and [0445] of Larsen. In the instant application, the broadest reasonable interpretation of “the information about the respective training sample comprises the second corresponding representation of the corresponding plurality of abundance values in the second latent feature space [according to the third mapping function]” reads on the genomic-wide features in NPL Ammad-ud-din (Third Paragraph on Page 32) about the respective tumor organoid culture comprising the cohort level pan-cancer molecular characterization in Larsen (Paragraphs [0440] and [0445]) of the respective tumor organoid’s number of RNA copies present in a cell, the correlated abundance of the corresponding proteins produced in the cell, and the nucleic acid sequences in the low-dimensional latent component for the new patient sample.);
inputting, for each respective training sample in the second plurality of training samples, corresponding information about the respective training sample into the multi-task model, wherein the information about the respective training sample comprises the corresponding representation of the corresponding plurality of abundance values in the second latent feature space [according to the second mapping function] (Second Paragraph on Page 12, Third Paragraph on Page 32, and Third Paragraph to Fourth Paragraph on Page 34 of NPL Ammad-ud-din. In the instant application, the broadest reasonable interpretation of “inputting, for each respective training sample in the second plurality of training samples, corresponding information about the respective training sample into the multi-task model, wherein the information about the respective training sample comprises the corresponding representation of the corresponding plurality of abundance values in the second latent feature space [according to the second mapping function]” reads on the activity in NPL Ammad-ud-din (Second Paragraph on Page 12, Third Paragraph on Page 32, and Third Paragraph to Fourth Paragraph on Page 34) of inputting, for each respective new patient sample, corresponding information about the respective new patient sample into the multi-task learning (MTL) model, wherein the information about the respective new patient sample comprises the matrix of the number of RNA copies present in a cell, the correlated abundance of the corresponding proteins produced in the cell, and the nucleic acid sequences for the respective new patient sample in the low-dimensional latent component.); and
adjusting the plurality of parameters based on:
for each respective training sample in the first plurality of training samples, one or more differences between (i) the corresponding plurality of outputs and (ii) a set of labels comprising (a) the corresponding experimentally measured effect of the candidate pharmaceutical agent on the respective training sample and (b) the corresponding cell type classification for each respective cell type variable in the set of one or more cell type variables, and
for each respective training sample in the second plurality of training samples, one or more differences between (i) the corresponding plurality of outputs and (ii) a set of labels comprising the corresponding cell type classification for each respective cell type variable in the set of one or more cell type variables.
NPL Ammad-ud-din does not appear to explicitly disclose, but Colley teaches the limitation identified in bold as “at a computer system comprising at least one processor and a memory storing at least one program for execution by the at least one processor” (Paragraph [0379] of Colley. In the instant application, the broadest reasonable interpretation of “a computer system comprising at least one processor and a memory storing at least one program for execution by the at least one processor” reads on the computer in Colley (Paragraph [0379]) including the computer system having one or more processors and memory storing one or more programs for execution by the one or more processors.).
NPL Ammad-ud-din does not appear to explicitly disclose, but NPL Yang teaches the limitation identified in bold as “obtaining, for each respective training sample in a first plurality of training samples, … a corresponding set of one or more cell type classifications comprising, for each respective cell type variable in a set of one or more cell type variables, a corresponding cell type classification for the respective cell type variable” (Eighth Paragraph on Page 7 of NPL Yang. In the instant application, the broadest reasonable interpretation of “a corresponding set of one or more cell type classifications comprising, for each respective cell type variable in a set of one or more cell type variables, a corresponding cell type classification for the respective cell type variable” reads on the cancer types, tissue types, and tissue sites in NPL Yang (Eighth Paragraph on Page 7) for the respective tissue organoid culture.).
NPL Ammad-ud-din does not appear to explicitly disclose, but NPL Yang teaches the limitation identified in bold as “obtaining, for each respective training sample in a second plurality of training samples, … a corresponding set of one or more cell type classifications comprising, for each respective cell type variable in the set of one or more cell type variables, a corresponding cell type classification for the respective cell type variable” (Eighth Paragraph on Page 7 of NPL Yang. In the instant application, the broadest reasonable interpretation of “a corresponding set of one or more cell type classifications comprising, for each respective cell type variable in the set of one or more cell type variables, a corresponding cell type classification for the respective cell type variable” reads on the cancer types, tissue types, and tissue sites in NPL Yang (Eighth Paragraph on Page 7) for the respective new patient sample.).
NPL Ammad-ud-din does not appear to explicitly disclose, but NPL Ding teaches the limitation identified in bold as “learning a third mapping function that maps a representation of a corresponding plurality of abundance values in the first latent feature space to the second latent feature space” (First Paragraph in Second Column on Page 909 of NPL Ding. In the instant application, the broadest reasonable interpretation of “learning a third mapping function that maps a representation of a corresponding plurality of abundance values in the first latent feature space to the second latent feature space” reads on the activities in NPL Ding (First Paragraph in Second Column on Page 909) of performing, for primary tissues (i.e., the respective new patient sample) and MOS (i.e., the respective tissue organoid culture), RNA sequencing and using Uniform Manifold Approximation and Projection (UMAP) reductions (i.e., the third mapping function) to cluster concordant groups with comparable relative abundance levels.).
NPL Ammad-ud-din does not appear to explicitly disclose, but Larsen teaches the limitation identified in bold as “generating, for each respective training sample in the first plurality of training samples, a second corresponding representation of the corresponding plurality of abundance values in the second latent feature space according to the third mapping function” (Paragraphs [0440] and [0445] of Larsen. In the instant application, the broadest reasonable interpretation of “generating, for each respective training sample in the first plurality of training samples, a second corresponding representation of the corresponding plurality of abundance values in the second latent feature space according to the third mapping function” reads on the activity in Larsen (Paragraphs [0440] and [0445]) of determining molecular concordance between TOs (the tumor organoid cultures) and source tumors (the new patient samples) and carrying out a cohort level pan-cancer molecular characterization of TOs (tumor organoids) to compare to an independent cohort of representative cancer patients i.e., providing their new patient samples. The matrix in NPL Ammad-ud-din (Third Paragraph on Page 32 of NPL Ammad-ud-din) can be modified to represent the cohort level pan-cancer molecular characterization of TOs (tumor organoids) in the low-dimensional latent components for new patient samples.).
NPL Ammad-ud-din does not appear to explicitly disclose, but NPL Yang teaches the associated limitation identified in bold as “inputting … into a multi-task model … to generate… (ii) for each respective cell type variable in the set of one or more cell type variables, a corresponding cell type classification for the respective training sample” (Eighth Paragraph on Page 7 of NPL Yang. In the instant application, the broadest reasonable interpretation of “for each respective cell type variable in the set of one or more cell type variables, a corresponding cell type classification for the respective training sample” reads on the cancer types, tissue types, and tissue sites in NPL Yang (Eighth Paragraph on Page 7).).
NPL Ammad-ud-din does not appear to explicitly disclose, but Larsen teaches the limitation identified in bold as “adjusting the plurality of parameters based on: for each respective training sample in the first plurality of training samples, one or more differences between (i) the corresponding plurality of outputs and (ii) a set of labels comprising (a) the corresponding experimentally measured effect of the candidate pharmaceutical agent on the respective training sample, and (b) the corresponding cell type classification for each respective cell type variable in the set of one or more cell type variables” (Paragraph [0204] of Larsen. In the instant application, the broadest reasonable interpretation of “adjusting the plurality of parameters based on for each respective training sample in the first plurality of training samples, one or more differences between (i) the corresponding plurality of outputs and (ii) a set of labels comprising (a) the corresponding experimentally measured effect of the candidate pharmaceutical agent on the respective training sample, and (b) the corresponding cell type classification for each respective cell type variable in the set of one or more cell type variables” reads on the activity in Larsen (Paragraph [0204]) of adding covariates to the linear model to adjust for confounding technical effects including initial TO viability, differences in growth rate between TOs derived from different patients, etc.), and for each respective training sample in the second plurality of training samples, one or more differences between (i) the corresponding plurality of outputs and (ii) a set of labels comprising the corresponding cell type classification for each respective cell type variable in the set of one or more cell type variables” (Paragraph [0204] of Larsen. In the instant application, the broadest reasonable interpretation of “adjusting the plurality of parameters based on … for each respective training sample in the second plurality of training samples, one or more differences between (i) the corresponding plurality of outputs and (ii) a set of labels comprising the corresponding cell type classification for each respective cell type variable in the set of one or more cell type variables” reads on the activity in Larsen (Paragraph [0204]) of adding covariates to the linear model to adjust for confounding technical effects including differences in growth rate between TOs derived from different patients, different cancer types, etc..).
Therefore, it would have been obvious to one of ordinary skill in the art of machine learning in oncology at the time of filing to modify the method of NPL Ammad-ud-din to: include the computer system comprising at least one processor and a memory storing at least one program for execution by the at least one processor, as taught by Colley (Paragraph [0379]) in order to allow for precision-level analysis of patient health needs, in order to provide the right resources, at the right time, to the right patients (Paragraph [0120] of Colley); include the steps of obtaining, for each respective training sample in a first plurality of training samples, … a corresponding set of one or more cell type classifications comprising, for each respective cell type variable in a set of one or more cell type variables, a corresponding cell type classification for the respective cell type variable, and obtaining, for each respective training sample in a second plurality of training samples, … a corresponding set of one or more cell type classifications comprising, for each respective cell type variable in the set of one or more cell type variables, a corresponding cell type classification for the respective cell type variable, as taught by NPL Yang (Eighth Paragraph on Page 7) in order to leverage the power of multitask learning to provide novel insights into the molecular underpinnings of drug response (Last Paragraph on Page 1 of NPL Yang); include the step of learning a third mapping function that maps a representation of a corresponding plurality of abundance values in the first latent feature space to the second latent feature space, and inputting … into a multi-task model … to generate… (ii) for each respective cell type variable in the set of one or more cell type variables, a corresponding cell type classification for the respective training sample, as taught by NPL Ding (First Paragraph in Second Column on Page 909) in order to facilitate drug screening and personalized medicine treatment (Fourth Paragraph in First Column on Page 907 of NPL Ding); include the step of generating, for each respective training sample in the first plurality of training samples, a second corresponding representation of the corresponding plurality of abundance values in the second latent feature space according to the third mapping function, as taught by NPL Yang (Paragraphs [0440] and [0445 in order to leverage the power of multitask learning to provide novel insights into the molecular underpinnings of drug response (Last Paragraph on Page 1 of NPL Yang); and include the step of adjusting the plurality of parameters based on for each respective training sample in the first plurality of training samples, one or more differences between (i) the corresponding plurality of outputs and (ii) a set of labels comprising (a) the corresponding experimentally measured effect of the candidate pharmaceutical agent on the respective training sample, and (b) the corresponding cell type classification for each respective cell type variable in the set of one or more cell type variables, and the step of adjusting the plurality of parameters based on … for each respective training sample in the second plurality of training samples, one or more differences between (i) the corresponding plurality of outputs and (ii) a set of labels comprising the corresponding cell type classification for each respective cell type variable in the set of one or more cell type variables, as taught by Larsen (Paragraph [0204]) in order to provide systems and methods for using organoid cultures to improve treatment predictions and outcomes, assessing the effectiveness of various drugs on one or more tumor organoid lines, and determine if a specific drug may be useful in killing cancer cells with specific genetic mutations or phenotypes (Paragraph [0007] of Larsen).
Conclusion
Any inquiry concerning this communication or earlier communications from the examiner should be directed to VINCENT CAESAR ILAGAN whose telephone number is (703) 756-1639. The examiner can normally be reached Monday - Friday 8:30 am - 6:00pm.
Examiner interviews are available via telephone, in-person, and video conferencing using a USPTO supplied web-based collaboration tool. To schedule an interview, applicant is encouraged to use the USPTO Automated Interview Request (AIR) at http://www.uspto.gov/interviewpractice.
If attempts to reach the examiner by telephone are unsuccessful, the examiner’s supervisor, Jason B. Dunham, can be reached on (571) 272-8109. The fax phone number for the organization where this application or proceeding is assigned is 571-273-8300.
Information regarding the status of published or unpublished applications may be obtained from Patent Center. Unpublished application information in Patent Center is available to registered users. To file and manage patent submissions in Patent Center, visit: https://patentcenter.uspto.gov. Visit https://www.uspto.gov/patents/apply/patent-center for more information about Patent Center and https://www.uspto.gov/patents/docx for information about filing in DOCX format. For additional questions, contact the Electronic Business Center (EBC) at 866-217-9197 (toll-free). If you would like assistance from a USPTO Customer Service Representative, call 800-786-9199 (IN USA OR CANADA) or 571-272-1000.
/V.C.I./Examiner, Art Unit 3686
/DEVIN C HEIN/Examiner, Art Unit 3686
Prosecution Insights