DETAILED ACTION
Applicant’s response, 06 June 2025, has been fully considered. The following rejections and/or objections are either reiterated or newly applied. They constitute the complete set presently being applied to the instant application.
Herein, "the previous Office action" refers to the final rejection of 31 May 2024.
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 .
Continued Examination Under 37 CFR 1.114
A request for continued examination under 37 CFR 1.114, including the fee set forth in 37 CFR 1.17(e), was filed in this application after final rejection. Since this application is eligible for continued examination under 37 CFR 1.114, and the fee set forth in 37 CFR 1.17(e) has been timely paid, the finality of the previous Office action has been withdrawn pursuant to 37 CFR 1.114. Applicant's submission filed on 06 June 2025 has been entered.
Priority
Applicant’s claim under 35 USC § 119(e) for the benefit of prior-filed Provisional Application No. 62/486692 is acknowledged.
In this action, all claims are examined as though they had an effective filing date of 18 Apr 2017. In future actions, the effective filing date of one or more claims may change, due to amendments to the claims, or further analysis of the disclosure of the priority application.
Status of Claims
Claims 12-13 are cancelled.
Claims 1-11 and 14-32 are pending.
Claims 1-11 and 14-32 are rejected to.
Claims 8 and 15 are objected to.
Claim Objections
Claims 8 and 15 are objected to because of the following informalities. This objection is newly recited.
Claim 8 recites “…wherein the one or more additional target proteins is…”, which is a grammatical error and should recite “…the one or more additional target proteins [[is]] are…”.
Claim 15 recites “wherein the chemical fingerprint analysis is…”, which should be amended to recite “…wherein the chemical fingerprints are…”, to use consistent language with claim 14 and increase clarity.
Appropriate correction is required.
Claim Rejections - 35 USC § 112(b)
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.
Claims 7, 18, and 22 are rejected under 35 U.S.C. 112(b) as being indefinite for failing to particularly point out and distinctly claim the subject matter which the inventor or a joint inventor regards as the invention. This rejection is newly recited.
Claim 7 is indefinite for recitation of “The method of claim 6, wherein the candidate compound list is capable of being displayed and ranked by the selectivity of the candidate compound for the target protein over the one or more additional target proteins”. Claim 1, from which claim 7 ultimately depends, recites “(e) outputting a list of the candidate compounds displayed and prioritized by estimated binding interactions according to each compound’s predicted binding interaction value”. Examples of claim language, although not exhaustive, that may raise a question as to the limiting effect of the language in a claim include “wherein clauses”. The court noted that a "‘whereby clause in a method claim is not given weight when it simply expresses the intended result of a process step positively recited.’" Id. (quoting Minton v. Nat’l Ass’n of Securities Dealers, Inc., 336 F.3d 1373, 1381, 67 USPQ2d 1614, 1620 (Fed. Cir. 2003)). In the instant case, due to word “capable” if the limitation relating displaying and ranking the candidate compound list by selectivity for the target protein over the additional target proteins is merely expressing an intended result of outputting the list of candidate compounds of claim 1 (i.e. the list could be displayed and ranked in another order), or if claim 7 is requiring some particular feature of the outputted list (e.g. is the list required to include the selectivity information, is the list required to have sorting functions, etc.). Therefore, the metes and bounds of the claim are unclear. For purpose of examination, claim 7 is interpreted to recite an intended result or use of the outputted list of claim 1.
Claim 18 is indefinite for recitation of “The method of claim 16, wherein the candidate compound list is capable of being displayed and ranked by the believability score…”, which is indefinite for the same reasons discussed above for claim 7. For purpose of examination, claim 18 is interpreted to recite an intended result or use of the outputted list of claim 1.
Claim 22 is indefinite for recitation of “The method of claim 16, wherein the candidate compound list is capable of being displayed and ranked by the one or more property findings…”, which is indefinite for the same reasons discussed above for claim 7. For purpose of examination, claim 22 is interpreted to recite an intended result or use of the outputted list of claim 1.
Claim Rejections - 35 USC § 112(d)
The following is a quotation of 35 U.S.C. 112(d):
(d) REFERENCE IN DEPENDENT FORMS.—Subject to subsection (e), a claim in dependent form shall contain a reference to a claim previously set forth and then specify a further limitation of the subject matter claimed. A claim in dependent form shall be construed to incorporate by reference all the limitations of the claim to which it refers.
Claims 7, 18, and 22 are rejected under 35 U.S.C. 112(d) as being of improper dependent form for failing to further limit the subject matter of the claim upon which it depends, or for failing to include all the limitations of the claim upon which it depends. This rejection is newly recited.
Claim 7 recites “The method of claim 6, wherein the candidate compound list is capable of being displayed and ranked by the selectivity of the candidate compound for the target protein over the one or more additional target proteins”, which is interpreted to recite an intended result of outputting the list in claim 1. However, claim 7 fails to further limit the subject matter of claims 1 and 6, from which it depends.
Claim 18 recites “The method of claim 16, wherein the candidate compound list is capable of being displayed and ranked by the believability score…”, which is interpreted to recite an intended result of outputting the list in claim 1. However, claim 18 fails to further limit the subject matter of claims 1 and 16, from which it depends.
Claim 22 recites “The method of claim 19, wherein the candidate compound list is capable of being displayed and ranked by the one or more property findings…”, which is interpreted to recite an intended result of outputting the list in claim 1. However, claim 22 fails to further limit the subject matter of claims 1 and 19, from which it depends.
Applicant may cancel the claim(s), amend the claim(s) to place the claim(s) in proper dependent form, rewrite the claim(s) in independent form, or present a sufficient showing that the dependent claim(s) complies with the statutory requirements.
Claim Rejections - 35 USC § 101
The rejection of claims 12-13 under 35 U.S.C. 101 in the previous Office action has been withdrawn in view of the cancelation of these claims received 06 June 2025.
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.
Claim 25 is rejected under 35 U.S.C. 101 because the claimed invention is directed to non-statutory subject matter. This rejection is newly recited.
The claim(s) does/do not fall within at least one of the four categories of patent eligible subject matter because:
Claim 25 recites “A computer readable medium having stored hereon executable instructions…”. Therefore, the subject matter of claim 25 is a computer program on computer readable media. A review of the specification does not show a definition of computer readable media that excludes an embodiment that is information in a signal. As such, an embodiment of the claims read on non-statutory subject matter (In re Nuijten 84 USPQ2d 1495 (2007)). The applicants may overcome the rejection by 1) amendment of the claims to be limited to physical forms of computer readable storage media described in the specification or 2) by amending the claimed subject matter to be limited to “non-transitory computer readable medium”, see the notice regarding Computer Readable Media (1351 OG 212 (23 February 2010)).
Claims 1-11 and 14-32 are rejected under 35 USC § 101 because the claimed inventions is directed to a judicial exception without significantly more. Any newly recited portions are necessitated by claim amendment.
The Supreme Court has established a two-step framework for this analysis, wherein a claim does not satisfy § 101 if (1) it is “directed to” a patent-ineligible concept, i.e., a law of nature, natural phenomenon, or abstract idea, and (2), if so, the particular elements of the claim, considered “both individually and as an ordered combination,” do not add enough to “transform the nature of the claim into a patent-eligible application.” Elec. Power Grp., LLC v. Alstom S.A., 830 F.3d 1350, 1353 (Fed. Cir. 2016) (quoting Alice, 134 S. Ct. at 2355). Applicant is also directed to MPEP 2106.
Step 1: The instantly claimed invention (claims 1 and 26 being representative) is directed to a method and product. Therefore, the instantly claimed invention falls into one of the four statutory categories. [Step 1: YES] While claim 25 is not directed to statutory subject matter, in the interest of compact prosecution, the claim is being examined under Step 2A and Step 2B, below.
Step 2A: First it is determined in Prong One whether a claim recites a judicial exception, and if so, then it is determined in in Prong Two if the recited judicial exception is integrated into a practical application of that exception.
Step 2A, Prong 1: Under the MPEP § 2106.04, the Step 2A (Prong 1) analysis requires determining whether a claim recites an abstract idea, law of nature, or natural phenomenon.
Claims 1 and 25-26 recite the following steps which fall under the mathematical concepts and/or mental processes groupings of abstract ideas:
providing a plurality of binding interaction findings for a target protein…, wherein each binding interaction finding is a binding interaction, or lack of binding interaction, between the target protein and a training set compound that is determined experimentally in an affinity-based compound library selection, wherein the training set compound is one of a plurality of training set compounds and wherein the specificity of a binding interaction between a training set compound and the target protein is assessed by enrichment of the training set compound in the affinity-based compound library selection under conditions for specific target protein binding, further wherein each training set compound comprises a nucleotide tag encoding the identity of the training set compound, and wherein at least 90% of the binding interaction findings in the plurality are representative of a binding interaction between the target protein and a training set compound, and wherein the training set compounds do not include candidate compounds, and further wherein the plurality of binding interaction findings comprise at least 250,000 binding interaction findings (mental process);
generating a training data set that comprises the plurality of at least 250,000 binding interaction findings and further comprises training set compound molecular structure information (mental process);
training a model using…machine learning and the training data set, wherein the training comprises analysis of comparisons between each of the at least 250,000 binding interaction findings in association with the training set compound molecular structure information (mathematical concept);
using….the model to generate estimated binding interactions between the target protein and the set of candidate compounds….wherein the candidate compounds differ from the training set compounds (mathematical concept);
wherein generating the estimated binding interaction findings comprises providing input data to the model comprising molecular representation data for the candidate compounds and generation, for each candidate compound of the set of candidate compounds, a predicted binding interaction value based at least in part on the molecular representation data (mathematical concept).
The identified claim limitations falls into one of the groups of abstract ideas of mathematical concepts and/or mental processes for the following reasons. First, the steps of providing information and generating a training set are similar to the concepts of "collecting information, analyzing it, and displaying certain results of the collection and analysis," where the data analysis steps are recited at a high level of generality such that they could practically be performed in the human mind, Electric Power Group v. Alstom, S.A., 830 F.3d 1350, 1353-54, 119 USPQ2d 1739, 1741-42 (Fed. Cir. 2016). The steps only require broadly collecting/providing particular information relating to protein and target binding interactions, and then analyzing and organizing the information to form a training set, which can be practically performed in the mind aided with pen and paper. See MPEP 2106.04(a)(2) III.
Furthermore, the limitations of training a model using machine learning and the training set to estimate interactions between the target protein and compounds (i.e. associating binding interaction findings with molecular structure information during training) and then using the trained model to generate estimated binding interactions between the target protein and candidate compounds by providing input data to the model which then outputs a predicted binding interaction recite a mathematical concept because they amount to a textual equivalent to performing mathematical calculations. A claim that recites a mathematical calculation, when the claim is given its broadest reasonable interpretation in light of the specification, will be considered as falling within the "mathematical concepts" grouping. For example, a step of "determining" a variable or number using mathematical methods or "performing" a mathematical operation may also be considered mathematical calculations when the broadest reasonable interpretation of the claim in light of the specification encompasses a mathematical calculation. See MPEP 2106.04(a)(2) I. C. In the instant case, in light of Applicant’s specification at para. [0131]-[0132] and [0139], the training and application of a machine learning model includes training and using a Naïve Bayes algorithm (e.g. a probabilistic model) which requires calculating posterior probabilities and analyzing statistics to mathematical determine the patterns in the training data and furthermore takes numerical values as input and then performs mathematical calculations to generate a numerical output of the predicted binding interaction value. Therefore, these limitations recite mathematical concepts.
Dependent claims 2-9, 14-22, and 27-32 further recite an abstract idea and/or are part of the abstract idea identified in claims 1 and 25-26 above. Dependent claim 2 further limits the mental process of providing binding interaction findings to provide one million findings. Dependent claims 3-4 further limit the mental process of providing binding interaction findings between the target protein and a training set compound with a nucleotide tag encoding the identified of the compound. Dependent claim 5 defines the process in which the binding interaction findings were previously determined, and thus are part of the abstract idea of providing the interaction findings discussed above. Dependent claim 6 further recites the mental process of providing additional binding interaction findings for additional target proteins that are a mutant or isoform of the target protein, which is a mental process for the same reasons regarding providing information discussed above with respect to claim 1. Dependent claim 7 fails to further limit the subject matter of claims and 1 and 6, and thus are part of the abstract idea of claims 1 and 6 above. Dependent claim 8 further limits the mental process of providing information in claim 6. Dependent claim 9 further recites the mental process of providing additional binding interaction findings for negative control experiments, which is a mental process for the same reasons regarding providing information discussed above with respect to claim 1. Dependent claims 14-14 further limits the mathematical concept of the input into the machine learning model to include chemical fingerprints, such as a Morgan fingerprint (e.g. numerical data). Dependent claim 16-17 further recites the mental process and mathematical concept of generating a believability score for each estimated binding interaction by using a principal component analysis (i.e. math) and using chemical structure comparisons of the candidate compounds and one or more compounds from the plurality of binding interactions for the target protein. Dependent claim 18 fails to further limit the subject matter of claim 16 and thus is part of the abstract idea of claim 16. Dependent claims 19-20 further recites the mental process of providing property findings, including molecular weight and/or clogP, for candidate compounds. Dependent claim 21 further limits the mathematical concept of estimating the estimated binding interactions to use the one or more property findings. Dependent claim 22 fails to further limit the subject matter of claim 19, and thus is part of the abstract idea of claim 19. Dependent claims 27-30 further limit the abstract idea of providing binding interaction findings to provide at least 25 million binding interaction findings. Dependent claims 31-32 further limit the mathematical concept of training and using the model to use disynthon compound analysis, which uses disynthons (e.g. a substructure of the compound) to generate statistics in the machine learning process (see Applicant’s specification at para. [0129]).
Therefore, claims 1-11 and 14-32 recite an abstract idea. [Step 2A, Prong 1: YES]
Step 2A: Prong 2: Under the MPEP § 2106.04, the Step 2A, Prong 2 analysis requires identifying whether there are any additional elements recited in the claim beyond the judicial exception(s), and evaluating those additional elements to determine whether they integrate the exception into a practical application of the exception. This judicial exception is not integrated into a practical application for the following reasons.
Dependent claims 2-9, 11, 14-22, and 27-32 do not recite any elements in addition to the judicial exception and thus are part of the judicial exception.
The additional elements of claims 1 and 25-26 include:
a physical computing device/computer;
a computer readable medium;
a physical computing device having a representation of a set of candidate compounds (i.e. storing data);
outputting a list of the candidate compounds displayed and prioritized by estimated binding interactions according to each compound’s predicted binding interaction value (i.e. data output).
The additional element of claim 10 includes:
transmitting the candidate compound list over the internet or to a display device (i.e. transmitting data).
The additional elements of a computer, memory, storing data, outputting data, and transmitting data are generic computer components and/or functions. The courts have found the use of a computer or other machinery in its ordinary capacity for economic or other tasks (e.g., to receive, store, or transmit data) or simply adding a general purpose computer or computer components after the fact to an abstract idea (e.g., a fundamental economic practice or mathematical equation) does not integrate a judicial exception into a practical application. See MPEP 2106.05(f).
The additional element of claims 23-24 include:
synthesizing one or more of the candidate compounds from the molecular from the candidate compound list (claim 23).; and
contacting the one or more synthesizing candidate compounds with the target protein to determine one or more experimental binding interactions (claim 24).
Adding the generic technology of synthesizing and assaying the compounds to the abstract idea imposes no meaningful limits on how the abstract idea itself is performed or implemented. Similarly, adding the abstract idea to the steps of synthesis and assay does not impose any meaningful limits on how the technologies of compound synthesis or assay operate. The compound is synthesized and assayed in exactly the same manner regardless of whether a practitioner designed the compound using the claimed abstract idea, some other design procedure, or designed the compound arbitrarily. Hence, these steps only generally links the abstract idea to the technological environment of synthesizing and assaying compounds, rather than integrating the abstract idea into a practical application (see MPEP 2106.04(d) § I; and MPEP 2106.05(h)).
Therefore, the additionally recited elements amount to insignificant extra-solution activity and, as such, the claims as a whole do no integrate the abstract idea into practical application. Thus, claims 1-11 and 14-32 are directed to an abstract idea. [Step 2A, Prong 2: NO]
Step 2B: In the second step it is determined whether the claimed subject matter includes additional elements that amount to significantly more than the judicial exception. See MPEP § 2106.05.
The claims do not include any additional steps appended to the judicial exception that are sufficient to amount to significantly more than the judicial exception for the following reasons.
Dependent claims 2-9, 11, 14-22, and 27-32 do not recite any elements in addition to the judicial exception and thus are part of the judicial exception.
The additional elements of claims 1 and 25-26 include:
a physical computing device/computer;
a computer readable medium;
a physical computing device having a representation of a set of candidate compounds (i.e. storing data);
outputting a list of the candidate compounds displayed and prioritized by estimated binding interactions according to each compound’s predicted binding interaction value (i.e. data output).
The additional element of claim 10 includes:
transmitting the candidate compound list over the internet or to a display device (i.e. transmitting data).
The additional elements of a computer, memory, storing data, outputting data, and transmitting data are conventional computer components and/or functions. The courts have found the use of a computer or other machinery in its ordinary capacity for economic or other tasks (e.g., to receive, store, or transmit data) or simply adding a general purpose computer or computer components after the fact to an abstract idea (e.g., a fundamental economic practice or mathematical equation) does not provide significantly more. See Affinity Labs v. DirecTV, 838 F.3d 1253, 1262, 120 USPQ2d 1201, 1207 (Fed. Cir. 2016) (cellular telephone); TLI Communications LLC v. AV Auto, LLC, 823 F.3d 607, 613, 118 USPQ2d 1744, 1748 (Fed. Cir. 2016) (computer server and telephone unit).
The additional element of claims 23-24 include:
synthesizing one or more of the candidate compounds from the molecular from the candidate compound list (claim 23).; and
contacting the one or more synthesizing candidate compounds with the target protein to determine one or more experimental binding interactions (claim 24).
Various prior art publications of record (e.g. Buller, et al. Bioconjugate Chemistry 2010; Decurtins, et al. Nature Protocols 2016; Dumelin, et al. QSAR & Combinatorial Science 2006; Melkko, et al. Drug Discovery Today 2007) demonstrate that synthesizing and screening a DNA-encoded chemical library was a well-understood, routine and conventional practice in the art prior to the time of invention. Hence, these elements, when considered individually, are insufficient to constitute inventive concepts that would render the claims significantly more than an abstract idea (see MPEP 2106.05(d)).
Taken alone, the additional elements do not amount to significantly more than the above-identified judicial exception(s). Even when viewed as a combination, the additional elements fail to transform the exception into a patent-eligible application of that exception. Thus, the claims as a whole do not amount to significantly more than the exception itself. [Step 2B: NO]
Therefore, the instantly rejected claims are not drawn to eligible subject matter as they are directed to an abstract idea (and/or natural correlation) without significantly more. For additional guidance, applicant is directed generally to applicant is directed generally to the MPEP § 2106.
Response to Arguments
Applicant's arguments filed 06 June 2025 regarding 35 U.S.C. 101 have been fully considered but they are not persuasive.
Applicant remarks the present claims are analogous to the patent-eligible subject matter of PEG Example 39, in which a computer implemented method for facial detection is claimed which utilizes sets of digital facial images to train a neural network and does not recite a judicial exception, and further remarks like in Example 39, Applicant’s methods begin with a complex data set derived from at least 250,000 binding interaction findings to train a model using machine learning model, analogous to Example 39’s use of facial images to train a neural network, and the model is then utilized to detect compounds with particular binding properties, analogous to Examples 39’s use of the model to detect faces with particular features (Applicant’s remarks at pg. 9, para. 5 to pg. 10, para. 3). Applicant further remarks the comparative analysis claimed would require performing 31 billion pairwise interactions given the dataset has 250,000 binding interaction findings, and this level of complexity is analogous to example 39’s collection and comparison of digital facial images (pg. 11, para. 4 10 pg. 12, para. 2). Applicant further remarks steps (c) and (d) do not recite a mathematical concept because claim limitations may be based on mathematical concepts as long as the mathematical concepts are not recited in the claims, and in this respect the present claims parallel the patent-eligible claim of example 39, and further remarks the complexity of the current training set and process of training a model are entirely analogous to the collection and comparison of digital images in example 39 (Applicant’s remarks at pg. 12, para. 6 to pg. 13, para. 3).
This argument is not persuasive. Example 39, which has been incorporated into MPEP 2106.04(a)(1), recites a method comprising “collecting a set of digital facial images, applying one or more transformations to the digital images, creating a first training set including the modified set of digital facial images; training the neural network in a first stage using the first training set, creating a second training set including digital non-facial images that are incorrectly detected as facial images in the first stage of training; and training the neural network in a second stage using the second training set”.
This is not analogous to the instant claims for the following reasons. First, the instant claims train the model on a training data that is derived from “a plurality of binding interaction findings for a target protein…, wherein each binding interaction finding is a binding interaction or a lack of binding interaction…wherein the plurality of binding interaction findings comprises at least 250,000 binding interaction findings”. In light of Appellant’s specification at pg. 29 lines 1-4 and 13-17 the binding interaction findings are simply numerical values including binary values or a probability score from 0 to 1. It is not persuasive that the human mind cannot observe numerical values, even 250,000 numerical values, and analyze said values to provide interaction findings. For comparison, books often contain more than 250,000 words; the human mind is capable of observing or reading and analyzing large numbers of words. Therefore, the data used in the instant claims is not analogous to the digital images used in Example 39.
Furthermore, the instant claims simply require “training a model using computer-implemented machine learning”, which was found to recite a mathematical concept in the Final Rejection mailed 31 May 2024. MPEP 2106.04(a)(2) I. C. states a claim that recites a mathematical calculation, when the claim is given its broadest reasonable interpretation in light of the specification, will be considered as falling within the "mathematical concepts" grouping. A mathematical calculation is a mathematical operation (such as multiplication) or an act of calculating using mathematical methods to determine a variable or number, e.g., performing an arithmetic operation such as exponentiation. There is no particular word or set of words that indicates a claim recites a mathematical calculation.
Applicant’s specification as originally filed at pg. 26, lines 39-41, discloses non-limiting machine learning methods include Naïve Bayes, Random Forest, Decision Tree, and Support Vector Machines. Other machine learning models generally known in the art include models such as linear regression models. Thus the instant claims encompass utilizing a statistical classification model as the machine learning model which only requires utilizing multiplication and division on numerical values in a Naïve Bayes model, or performing weighted addition with numerical values in a linear regression model. Similarly training such a model only requires calculating a loss using a loss function (i.e. performing subtraction) using an estimated and a known interaction finding, adjusting the value of a parameter in the model to reduce the loss, and repeating the process. Thus, given the broadest reasonable interpretation of the claim, in light of the specification, these steps encompass a mathematical calculation and amount to a textual equivalent of performing mathematical calculations. Furthermore, the comparison step is part of the mathematical concept of training of the machine learning model, and there is no requirement that a mathematical calculation recited in a claim must be practically performable in the mind to qualify as a mathematical calculation as defined in MPEP 2106.04(a)(2) I. Regardless, it is noted that the training a model with a large number of interaction findings would only require repeating a calculation of an estimated interaction from molecular descriptors (e.g. 0s and 1s as described at pg. 29 of the specification) of a single molecule of the training set using the Naïve Bayes Model for each molecule, which could be performed mentally aided with pen and paper to track the results of each calculation. It is further noted nothing in the claims do not require billions of comparisons, and instead only compare each of the 250,000 binding interactions with structure information (250,000 comparisons).
Therefore, the instant claims encompass training a statistical Naïve Bayes and/or linear regression model using numerical values, which is not analogous to Example 39 which requires training a neural network on digital images, and the instant claims can be readily distinguished from those of example 39.
Applicant remarks the claimed methods require more than observing values or reading large numbers or words, and the claimed analysis involves the complex identification of patterns of binding interactions, which contradicts the assertion that the analysis required by the present claims is similar in nature to reading and analyzing 250,000 words (Applicant’s remarks at pg. 12, para. 3-5).
This argument is not persuasive. As discussed above, the analysis of the binding interaction findings in the training of a machine learning model is a mathematical concept. Whether a mathematical concept can be practically performed in the human mind is not a consideration in determining whether the limitation recites the abstract idea of a mathematical concept. With respect to simply “providing a plurality of binding interaction findings”, including 250,000 binding interaction findings does not require any particular “analysis” and instead encompasses reading and writing 250,000 already generated interaction findings. This limitation recites a mental process and is not “complex” as discussed in the above rejection.
Claim Rejections - 35 USC § 103
The rejection of claims 12-13 under 35 U.S.C. 103 in the previous Office action have been withdrawn in view of the cancellation of these claims received 06 June 2025.
In the event the determination of the status of the application as subject to AIA 35 U.S.C. 102 and 103 (or as subject to pre-AIA 35 U.S.C. 102 and 103) is incorrect, any correction of the statutory basis for the rejection will not be considered a new ground of rejection if the prior art relied upon, and the rationale supporting the rejection, would be the same under either status.
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.
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 1-5, 10-11, 14, and 19-32 are rejected under 35 U.S.C. 103 as being unpatentable over Agrafiotis (2003) and Litovchick (2015). Any newly recited portion is necessitated by claim amendment.
Cited documents:
Agrafiotis et al., US 2003/0014191 A1; previously cited; and
Litovchick et al., Encoded Library Synthesis Using Chemical Ligation and the Discovery of she Inhibitors from a 334-Million Member library, 2015, Scientific Reports, pg. 1-8; previously cited.
Regarding claims 1 and 25-26, Agrafiotis discloses “an automatic, partially automatic, and/or manual iterative system comprising a computer, method and/or computer program product for generating chemical entities having desired or specified physical, chemical, functional, and/or bioactive properties” ([0016]; FIG. 3), which includes the following functions:
Agrafiotis discloses (a) assaying the chemical compounds in a directed diversity libraries to obtain physical properties data ([0405]), wherein the physical properties data include inhibition data relating to enzyme activity (i.e. binding interaction findings between a target protein and compound) ([0406]). Regarding the process in which each binding interaction was determined experimentally in an affinity-based compound library selection by enrichment of the training set compound in the affinity-based library selection under conditions for specific target protein binding, this limitation only serves to define the process in which the plurality of binding interactions were previously determined. See MPEP 2113 I. Here, the interaction data of Agrafiotis is the same product as recited in the claims, even if the data was generated by a different process, and thus Agrafiotis discloses this limitation.
Agrafiotis discloses that the physical properties data includes data for multiple compounds ([0407]) and one receptor ([0408]), so all “of the binding interaction findings … are representative of a binding interaction between the target protein and a compound [in the set of candidate compounds]”, and further discloses an example of a compound library for compounds that inhibit thrombin (i.e. all compounds inhibit the target protein) [0099]), thereby satisfying the >90% limitation of claim 1.
Agrafiotis further teaches that this system is advantageous for “efficiently and effectively generating new leads designed for specific utilities” (0015). Agrafiotis further discloses the physical properties data of inhibition data is used in a training set of structure-activity pairs ([0166]), and that a trained model can be used to estimate properties for new compounds ([0285]), demonstrating the training compounds do not include the candidate compounds. Agrafiotis additionally discloses using historical structure-property data in the training set (i.e. not including the candidate compounds) (FIG. 1, #128; [106]; [0132]).
Agrafiotis discloses (b) selecting (i.e. generating) a training set of interaction data ([0180]); each training case in the training set is a structure-property or structure-activity pair (i.e. interaction finding and molecular structure information of compound) ([0166]) from the physical properties data ([0105]).
Agrafiotis discloses using the physical properties data including the interaction data from the training set to generate a QSAR model ([0253]-[0254]). Agrafiotis discloses the trained QSAR model takes compounds as input and estimates molecular properties, including binding affinity for new compounds that have not yet been assayed ([0285]), demonstrating the training of the model compares the relationships between the structure and activity of the compounds in the training set (e.g. trained using structure-activity pairs).
Agrafiotis discloses (d) using the QSAR model to estimate molecular properties, including binding affinity, for new compounds (i.e. candidate compound) that have not yet been assayed ([0285]). Given these new compounds have not yet been assayed, they are different than the assayed set of training compounds.
Agrafiotis further discloses the estimating of the binding interactions for the new compound comprises inputting physicochemical and structural features (i.e. molecular representation data) of the new compound into the model ([0211]; [0212]).
Agrafiotis discloses producing (i.e. outputting) a list of compounds that best satisfy selection criteria in a manner specified by the objective function ([0312]), wherein the selection criteria are established for new compounds based on estimated molecular properties, which include predicted binding interactions ([0108]; [0116]-[0117]) and an objective function that evaluates the compounds ([0306]-[0307]). Thus the outputted list of compounds are displayed and prioritized based on each compounds predicted binding interaction.
Agrafiotis does not disclose the following limitations:
Regarding claims 1 and 25-26, Agrafiotis does not teach that each training set chemical compound comprises a nucleotide tag encoding the identity of the compound.
However, Litovchick discloses "a chemical ligation method for construction of DNA-encoded small-molecule libraries" (Abstract). These molecules include a DNA tag that encodes the structure of the molecule (e.g. p. 4, Fig. 2). 334 million compounds were generated. Litovchick teaches screening these molecules using protein capture on an immobilized resin (bot. of p. 5 – top of p. 6). Litovchick teaches that "we have demonstrated that chemical ligation is a feasible approach to the catenation of encoding oligonucleotides in DNA-encoded chemistry. … We believe this chemical ligation methodology represents an advance over previous methods and will expand the scope and diversity of chemistry addressable using DNA-encoded library synthesis" (top of p. 7). Litovchick further discloses DNA-encoding of small molecule libraries is an attractive strategy for the discovery of novel ligands to biological targets and allows the interrogation of a vast numbers of compounds (pg. 1, para. 1).
It would have been prima facie obvious to one of ordinary skill in the art, before the effective filing date of the claimed invention, to have modified the training set of chemical compounds of Agrafiotis to have included a nucleotide tag encoding the identify of each respective compound, as shown by Litovchick. One of ordinary skill in the art would have been motivated to combine the methods of Agrafiotis and Litovchick in order to facilitate the discovery of novel ligands to biological targets and allow the interrogation of vast numbers of compounds, as shown by Litovchick (pg. 1, para. 1), thus facilitating the method for discovering and developing bioactive compounds of Agrafiotis. This modification would have had a reasonable expectation of success because Agrafiotis discloses the system is amenable for use with any kind of compound and assay ([0406], and Litovchick discloses how to synthesize and assay a DNA-encoded chemical library, as discussed above, such that the assay of Litovchick is applicable to the methods of Agrafiotis.
Regarding claims 1 and 25-26, Agrafiotis further does not disclose the training set includes at least 250,000 binding interaction findings.
However, Agrafiotis does disclose that conventional chemical libraries total an estimated 9 million identified compounds, which reflect a small sampling of all possible organic compounds, and that conventional approaches for the identification of lead compounds is limited by the relative small pool of compounds ([0006]), that combinatorial chemical libraries can include millions of compounds ([0009]). Agrafiotis further discloses that conventional, manual procedures for generating lead compounds operates to impose a limit on the number of compounds that can be evaluated as new drug leads, which is an inefficient, labor-intensive, and time consuming process of limited scope ([0008]), demonstrating the desire to screen for a large number of compounds. Agrafiotis further discloses analyzing subsets of a compound library which includes chemical compounds that already exists or that can be synthesized on demand either individually or combinatorically ([0024]; [0083]-[0084]), using historical property-structure data in the training set ([0132]; [0141]), and that the training sets anticipated to be used in a typical operation of the system of the present invention are large ([0178]). Agrafiotis discloses the present invention enables the automatic screening of very large numbers of chemical compounds ([0134]).
Furthermore, Litovchick additionally discloses an example of a 334 million compound library (pg. 5, para. 4). Litovchick further discloses DNA-encoding of small molecule libraries is an attractive strategy for the discovery of novel ligands to biological targets and allows the interrogation of a vast numbers of compounds, exceeding by orders of magnitude the traditional capacity of “one compound per well” screening approaches (pg. 1, para. 1). Litovchick screens these compounds for enzyme activity (pg. 5, para. 4, e.g. library assayed for affinity; pg. 7, para. 2). Accordingly, Litovchick provides a screen procedure that allows for the screening of a large 334 million member library that overcomes labor-intensive and time-consuming processes relating to only using one compound per well, and provides structure-property data for 334 million compounds.
It would have been prima facie obvious, before the effective filing date of the claimed invention, to have modified the method of Agrafiotis to have used a historical training set comprising millions of compounds, as suggested by Agrafiotis, such as the 334 million compound library with enzyme affinity information of Litovchick. One of ordinary skill in the art would have been motivated to use a large training set comprising millions of compounds, in order to increase the scope of compounds being screened, thus overcoming the limited scope of small compound libraries, as shown by Agrafiotis ([0006]-[0009]), and to further allow the model to learn from historical data, as shown by Agrafiotis ([0132]). This modification would have had a reasonable expectation of success given Litovchick provides an approach for generating a 334 million member library through the use of DNA-encoding used for screening small molecules (pg. 1, para. 1), and furthermore, Agrafiotis states the training sets expected to be used in the invention are very large and may use data from other independent experiments ([0178]; [0374]).
Regarding the dependent claims:
Regarding claims 2 and 27-30, the size of the training set affects only how many times the claimed steps are performed, not which steps are performed. Repeating steps to handle a bigger training set does not patentably distinguish the invention from the prior art. Furthermore, Agrafiotis in view of Litovchick make obvious using a training set comprising at least 25 million binding interaction findings, as applied to claim 1 above.
Regarding claims 3 and 4, Agrafiotis discloses that the physical properties data includes data for multiple compounds (0407) and one receptor (0408), so all "of the binding interaction findings … are representative of a binding interaction between the target protein and a compound [in the set of candidate compounds]". Agrafiotis further discloses an embodiment of a compound library for compounds that inhibit thrombin (i.e. 100% compounds inhibit the target protein) [0099]), thereby satisfying the >95% limitation of claim 3, and the >99% limitation of claim 4.
Regarding claim 5, Agrafiotis in view of Litovchick make obvious providing binding interaction findings between a target protein and compound comprising a nucleotide tag encoding the identity of the compound, as applied to claim 1 above. The process in which the binding interaction findings were determined by contacting the training set compounds with a target protein only serves to define the process in which the interaction findings were previously determined, but does not require an active step of contacting. See MPEP 2113 I. Here, the interaction data of Agrafiotis in view of Litovchick is the same product as recited in the claims, even if the data was generated by a different process, and thus Agrafiotis discloses this limitation. Regardless, it is further noted Agrafiotis does disclose assaying the compounds to determine enzyme activity data and determining dissociation constants from the assay ([0102]; [0346]), demonstrating the compounds are contacted with the enzymes.
Regarding claim 10, Agrafiotis discloses an output device for displaying any of the system information (0335), which includes an output list ([0312]).
Regarding claim 11, Agrafiotis discloses accessing database servers (0311), suggesting access over the Internet. Agrafiotis further discloses the computer system can include a communications interface that allows software and data to be transferred between computer systems and external devices, and include a network interface ([0322]), demonstrating the access and operation over the internet.
Regarding claim 14, Agrafiotis discloses the binding interaction predictions can be based on molecular similarity calculations that use hashed fingerprints to represent molecular structure (i.e. molecular structure information comprises chemical fingerprints) ([0273]; [0278]–[0279]).
Regarding claims 19-22, Agrafiotis discloses that the binding interaction predictions can be based on molecular similarity calculations that use physicochemical properties, including octanol-water partition coefficient ([0148]); i.e.
c
log
p
. Compounds are capable of being ranked and displayed by any criteria.
Regarding claim 23, Agrafiotis discloses synthesizing candidate compounds from the candidate compound list ([0336]-[0339]; FIG. 2, e.g. compounds synthesized from directed diversity library).
Regarding claim 24¸ Agrafiotis discloses assaying the synthesized compounds to determine enzyme activity data ([0346]; [0405]-[0406]; FIG. 2, e.g. directed diversity library compounds assayed).
Regarding claims 31 and 32, Agrafiotis discloses tetrameric and pentameric compound libraries (0009), and that "each inhibitor is generally composed of, but not restricted to, three chemical building blocks" (0099); i.e. a trisynthon. Elsewhere, Agrafiotis teaches "replacing a building block of one or more compounds currently in the state" (0380); "a building block" is a single synthon. Litovchick also teaches testing compounds that have been synthesized by one, two or three synthetic cycles (p. 5). However, one of the pools in the final cycle was an "encoded null", which did not receive the final synthon (p. 5, caption of Fig. 4). Hence, some of the compounds taught by Litovchick were only disynthons. Collectively, these indica