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 .
Election/Restrictions
Applicant’s election of Species 2 (Claim 12), drawn to a dictionary learning model (L0-regularized autoencoder) to create cellular constituent modules, in the reply filed on January 30th, 2026 is acknowledged.
Claim 8 is withdrawn from further consideration pursuant to 37 CFR 1.142(b) as being drawn to a nonelected correlation model (Leiden clustering or Louvain clustering) to create cellular constituent modules. Claim 1 is the generic or linking claim. Election was made without traverse in the reply filed on January 30th 2026.
Applicant is reminded that upon the cancelation of claims to a non-elected invention, the inventorship must be corrected in compliance with 37 CFR 1.48(a) if one or more of the currently named inventors is no longer an inventor of at least one claim remaining in the application. A request to correct inventorship under 37 CFR 1.48(a) must be accompanied by an application data sheet in accordance with 37 CFR 1.76 that identifies each inventor by his or her legal name and by the processing fee required under 37 CFR 1.17(i).
Claim Status
Claims 4-5, 7, 9-11, 13-15, 17, 21-24, 27, 29-30, 32-33, 35-36, 38-39, 42, 44-45, 49-60, 62-63, 66, 68, 70-72 are cancelled.
Claims 1-3, 6, 12, 16, 18-20, 25-26, 28, 31, 34, 37, 40-41, 43, 46-48, 61, 64-65, 67, and 69 are currently pending and under exam herein.
Claim 8 is withdrawn.
Claims 1-3, 6, 12, 16, 18-20, 25-26, 28, 31, 34, 37, 40-41, 43, 46-48, 61, 64-65, 67, and 69 are rejected.
Claim 25 is objected to.
Priority
The instant application claims benefits to U.S. Provisional Patent Application No. 63/210,710, entitled "SYSTEMS AND METHODS FOR TERRAFORMING," filed June 15th, 2021. The claim to domestic benefit is acknowledge. Thus, the effective filing date of claims 1-3, 6, 12, 16, 18-20, 25-26, 28, 31, 34, 37, 40-41, 43, 46-48, 61, 64-65, 67, and 69 is June 15th, 2021.
Information Disclosure Statement
The information disclosure statements (ISD) was filed on 01/30/2026. All references in the IDS have been considered by the examiner and attached in this office action.
Drawings
The Drawings filed on 06/15/2022 are accepted.
Specification
The specification amendments filed on 08/12/2022 is objected to, as the amendments do not align with original specification priority and the application datasheet filed 08/12/2022. Examiner will proceed with interpretation of priority to original provisional application No. 63/210,710, entitled "SYSTEMS AND METHODS FOR TERRAFORMING," filed June 15th, 2021.
The title of the invention is not descriptive. A new title is required that is clearly indicative of the invention to which the claims are directed. The following title is suggested: Systems and Methods for Associating Cellular Constituents with Cellular Processes.
The disclosure is objected to because of the following informalities: Acronyms not spelled out
[00120]: ADC
[0041], [0122], [0270]: BCI, ECFP4, EcFC, MDL, RNNS2S, SMILES, RNNS2S,
Appropriate correction is required.
The use of the trade names or marks used in commerce below, has been noted in this application.
[00151]: Bluetooth Wi-Fi, Wi-MAX
[00152]: DVD, ANDROID, iOS, DARWIN, RTXC, LINUX, UNIX, OS X, WINDOWS, VxWorks
The terms should be accompanied by the generic terminology; furthermore, the term should be capitalized wherever it appears or, where appropriate, include a proper symbol indicating use in commerce such as ™, SM , or ® following the term.
Although the use of trade names and marks used in commerce (i.e., trademarks, service marks, certification marks, and collective marks) are permissible in patent applications, the proprietary nature of the marks should be respected and every effort made to prevent their use in any manner which might adversely affect their validity as commercial marks.
The lengthy specification has not been checked to the extent necessary to determine the presence of all possible minor errors. Applicant’s cooperation is requested in correcting any errors of which applicant may become aware in the specification.
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.
Claims 16, 25, 34, and 37 are 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.
For claims 16, and 37, the dependent claims (16 and 37) are expanding on what the number of variables (cells, cellular constituents, cellular constituent modules) are in the independent claim (1), such that it is unclear what the metes and bounds of the dependent claims are supposed to be as now the ranges include numbers that were not previously presented.
Claim 1 recites a plurality of cells comprising of ≥20 cells, a plurality of cellular constituents comprising ≥50 cellular constituents, and a plurality of cellular constituent modules comprising >10 cellular constituent modules
Claim 16 refers to the method of claim 1, and recites wherein the plurality of cellular constituents consists between 20 - 10,000 cellular constituents
Claim 37 refers to the method of claim 1, and recites wherein the plurality of cellular constituent modules comprises ≥5 cellular constituent modules
The claims recite both broad and narrower statement of the range/limitation. The claims are considered indefinite because there is a question or doubt as to whether the feature introduced by such narrower language is (a) merely exemplary of the remainder of the claim, and therefore not required, or (b) a required feature of the claims.
Claim 25 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. Claim 25 recites the following methods to generating a fingerprint: Daylight, BCI, ECFP4, EcFC, MDL, TTFP, UNITY 2D, RNNS2S, GraphConv, or fingerprint SMILES Transformer. However, there are no clear definition/explanation of all the abbreviations within the claims or the Specification of the instant application, hence it is unclear what the metes and bounds of the claim are.
Claim 34 recites the limitation "the Lipinski rule". 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 1-3, 6, 12, 16, 18-20, 25-26, 28, 31, 34, 37, 40-41, 43, 46-48, 61, 64-65, 67, and 69 are rejected under 35 U.S.C. 101 because the claimed invention is directed to an abstract idea and natural phenomenon without significantly more.
In accordance with MPEP § 2106, claims found to recite statutory subject matter (Step 1: YES) are then analyzed to determine if the claims recite any concepts that equate to an abstract idea, law of nature or natural phenomenon (Step 2A, Prong 1). In the instant application, the claims recite the following limitations that equate to an abstract idea:
Claims 1, 67, and 69 recites a method, system, and program respectively, to train a model to identify a set of cellular constituents associated with a cellular process of interest (abstract idea; mental process). The method starts by (B) using vectors in a dataset to create modules arranged in latent representation (abstract idea; mental process, and/or mathematical concepts). Then, (D) matrix multiplication is performed between the latent representation and a second dataset to obtain an activation data structure (abstract idea; mental process and/or mathematical process). Finally, (E) a model is trained using the difference between a calculated activation and measured activation, and parameters are adjusted based on the result of the difference (abstract idea; mental process and/or mathematical process). The claims are merely obtaining data, rearranging the data into modules, before correlating them with a cellular process based on the constituents’ expected and measured activation. Although, the claims recite performing the steps in a generic computer environment, the limitations can be practically performed in the human mind as well. Therefore, the claim limitations constitute both mental processes and mathematical concepts, falling broadly under the category of abstract ideas. Please see MPEP § 2106.04(a)(2) for more details. In addition, the invention seeks to correlate the presence of cellular constituents (genes, RNA, protein, etc.) with a cellular process, which also further falls under a law of nature/natural phenomenon.
Claim 2 recites (F) identifying one or more cellular constituent modules that is associated with one or more covariates, and then correlating the cellular constituents in the identified module with a cellular process of interest (abstract idea; mental process). The process of comparing one or more modules and seeing if they are activated in certain covariate (phenotypes, disease, cellular processes) to correlate the constituents in the modules with a cellular process is a mental process that can be practically performed in the human mind or with pen and paper.
Claim 12 recites that the use of vectors to create modules is done through a dictionary learning model, specifically an L0-regularized autoencoder (abstract idea; mental process and/or mathematical concepts). Based on the broadest reasonable interpretation, a L0-regularized autoencoder utilizes a loss function to prune inactive component and cluster the most essential features for representation. This process utilizes math to cluster vectors of similar features, which could also be reasonably accomplish by the human mind.
Claim 28 recites that the training of the model (E) is done through categorical cross-entropy loss in a multi-task formulation, where each covariate corresponds to a cost function (abstract idea; mathematical process). Again, this training utilizes loss/cost functions to evaluate the performance of multiple classification tasks through calculating the difference between predicted and true labels, which is a mathematical concept that can also be performed by the human mind with pen and paper.
Claim 31 recites using a contextualization algorithm, a gene set enrichment analysis algorithm (GSEA), to associate a cellular constituent module with a known cellular pathway, biological process, transcription factor, cell receptor, or kinase (abstract idea, mental process) and then pruning the activation data structure by removing the modules that fail to associate (abstract idea, mental process). A GSEA works by ranking all the genes in a dataset based on their differential expression between two phenotypes, then calculating an enrichment score before determining the enriched (overexpress/under express) gene set. This is a process that relies heavily on mathematical concepts, and in the broadest sense also achievable by the human mind.
Claim 43 recites the model of claim 1 is a logistic regression model, a neural network model, a support vector machine model, a Naive Bayes model, a nearest neighbor model, a boosted trees model, a random forest model, a decision tree model, a multinomial logistic regression model, a linear model, or a linear regression model, which are all mathematical models that utilize mathematical concepts to detect patterns in data (abstract idea; mathematical concepts). In addition, based on the broadest reasonable interpretation, simple models such as linear regression can also be done by a human with pen and paper, making the limitation a mental process under abstract ideas as well.
Claim 64 recites that the model of claim 1 is an ensemble model where the component models in the ensemble provide a calculated activation for a different cellular constituent module in the plurality of cellular constituent modules responsive to input of the covariate into the respective component model (abstract idea; mental process and/or mathematical concepts). The process of taking an input, comparing it against known data, before outputting a prediction on how likely the component will be active is a mental process that can be practically performed in the human mind.
Claim 65 recites that the component models are a logistic regression model, a neural network model, a support vector machine model, a Naive Bayes model, a nearest neighbor model, a boosted trees model, a random forest model, a decision tree model, a multinomial logistic regression model, a linear model, or a linear regression model. Similar to claim 43, these limitations are all mathematical models to detect patterns, and some can be implemented by the human mind as well (abstract idea; mathematical concepts and mental process).
The limitations regarding correlating constituent and constituent modules to covariates and cellular processes describes a mathematical process to associate collected data to known data before making a judgment, making these limitations mental process that can be performed in the mind or with pen and paper. While the limitations regarding calculation for correlations, enrichment, and activation are mathematical relations that also fall under abstract ideas. These recitations are similar to the concepts of collecting information, analyzing it and displaying certain results of the collection and analysis in Electric Power Group, LLC, v. Alstom (830 F.3d 1350, 119 USPQ2d 1739 (Fed. Cir. 2016)), organizing and manipulating information through mathematical correlations in Digitech Image Techs., LLC v Electronics for Imaging, Inc. (758 F.3d 1344, 111 U.S.P.Q.2d 1717 (Fed. Cir. 2014)) and comparing information regarding a sample or test to a control or target data in Univ. of Utah Research Found. v. Ambry Genetics Corp. (774 F.3d 755, 113 U.S.P.Q.2d 1241 (Fed. Cir. 2014)) and Association for Molecular Pathology v. USPTO (689 F.3d 1303, 103 U.S.P.Q.2d 1681 (Fed. Cir. 2012)) that the courts have identified as concepts that can be practically performed in the human mind or mathematical relationships. Therefore, these limitations fall under the “Mental process” and “Mathematical concepts” groupings of abstract ideas. While claims 1-3, 6, 12, 16, 18-20, 25-26, 28, 31, 34, 37, 40-41, 43, 46-48, 61, 64-65, 67, and 69 recite performing some aspects of the analysis with a “computer system” comprising of processors, memory, and a non-transitory computer-readable medium, there are no additional limitations that indicate that this computer system requires anything other than carrying out the recited mental process or mathematical concept in a generic computer environment. Merely reciting that a mental process is being performed in a generic computer environment does not preclude the steps from being performed practically in the human mind or with pen and paper as claimed. If a claim limitation, under its broadest reasonable interpretation, covers performance of the limitation in the mind but for the recitation of generic computer components, then if falls within the “Mental processes” grouping of abstract ideas. As such, claims 1-3, 6, 12, 16, 18-20, 25-26, 28, 31, 34, 37, 40-41, 43, 46-48, 61, 64-65, 67, and 69 recite an abstract idea (Step 2A, Prong 1: YES).
Claims found to recite a judicial exception under Step 2A, Prong 1 are then further analyzed to determine if the claims as a whole integrate the recited judicial exception into a practical application or not (Step 2A, Prong 2). This judicial exception is not integrated into a practical application because the claims do not recite an additional element that reflects an improvement to technology or applies or uses the recited judicial exception in some other meaningful way. Rather, the instant claims recite additional elements that amount to mere instructions to implement the abstract idea in a generic computing environment or mere instructions to apply the recited judicial exception via a generic experiment. Specifically, the claims recite the following additional elements:
Claim 1 recites a computer system comprising a memory and one or more processors, which equates to a generic computing environment/computer
Claims 1, 67, and 69 recites (A) obtaining one or more first datasets in electronic form, (C) obtaining one or more second datasets in electronic form, which fall under the common task of data gathering.
Claim 3 defines the annotated cell state of claim 1 as an exposure of a cell to a compound under an exposure condition, wherein the exposure condition is a duration of exposure, a concentration of the compound, or a combination of a duration of exposure and a concentration of the compound, while this dependent claim further limits the type of data being received in claim 1, it does not change the fact that the additional elements are mere data gathering steps.
Claim 6 defines the cellular constituents of claim 1 as a particular gene, a particular mRNA associated with a gene, a carbohydrate, a lipid, an epigenetic feature, a metabolite, a protein, or a combination thereof, and the abundance of the constituent is determined by a colorimetric measurement, a fluorescence measurement, a luminescence measurement, a resonance energy transfer (FRET) measurement, single-cell ribonucleic acid (RNA) sequencing (scRNA-seq), scTag-seq, single-cell assay for transposase-accessible chromatin using sequencing (scATAC-seq), CyTOF/SCoP, E-MS/Abseq, miRNA-seq, CITE-seq, or any combination thereof, again while this dependent claim further limits the type of data being received in claim 1, it does not change the fact that the additional elements are mere data gathering steps.
Claim 16 recites that the plurality of cellular constituents consists of between 20 - 10,000 cellular constituents, although the claim further limits the type of data received in claim 1, it does not change the fact that the additional elements are mere data gathering steps
Claim 18 recites that each cellular constituent module in the plurality of cellular constituent modules consists of between 2 - 300 cellular constituents, although the claim further limits the range of the data being rearranged, it does not change the fact that the process of data rearrangement is a mental process/mathematical concept
Claim 19 defines the cellular process of interest of claim 1 as an aberrant cell process associated with a disease, and the first plurality of cells includes cells that are representative of the disease and cells that are not representative of the disease as indicated by the plurality of annotated cell states, while the claim further limits the type of data received in claim 1, it does not change the fact that the additional elements are mere data gathering steps
Claim 20 defines the covariates of claim 1 as a cell batch, a cell donor, a cell type, a disease status, or exposure to a compound, and the respective representations as a cell batch identification, an identification of the cell donor/a characteristic of the cell donor, a cell type identification, an indication of absence or presence of disease, or a fingerprint of the compound, while the claim further limits the type of data received in claim 1, it does not change the fact that the additional elements are mere data gathering steps
Claim 25 specifies that the fingerprint of the compound from claim 20 can be generated by common methods such as: Daylight, BCI, ECFP4, EcFC, MDL, TTFP, UNITY 2D, RNNS2S, GraphCony, or fingerprint SMILES Transformer, while the claim further limits the type of data received, it does not change the fact that the additional elements are mere data gathering steps
Claim 34 recites that the compound of claim 3 is an organic compound having a molecular weight of less than 2000 Daltons that satisfies at least three criteria of the Lipinski rule of five criteria, while the claim further limits the type of compound from claim 3, it does not change the fact that the additional elements are mere data gathering steps
Claim 37 recites that that the plurality of cellular constituent modules comprises 5 or more modules, although the claim further limits the range of the data being rearranged, it does not change the fact that the process of data rearrangement is a mental process/mathematical concept
Claim 40 recites that each cellular constituent module comprises 5 or more cellular constituents, although the claim further limits the range of the data being rearranged, it does not change the fact that the process of data rearrangement is a mental process/mathematical concept
Claim 41 recites that each cellular constituent module consists between 2-20 cellular constituents, although the claim further limits the range of the data being rearranged, it does not change the fact that the process of data rearrangement is a mental process/mathematical concept
Claim 47 recites that the cells of claim 1 consist of cells from organ, tissue, stem cells, primary human cells, human cell lines, umbilical cord blood, peripheral blood, bone marrow, solid tissue or differentiated cells, while the claim further limits the type of data received in claim 1, it does not change the fact that the additional elements are mere data gathering steps
Claim 48 further defines the organ, tissue, stem cells, primary human cells, solid tissue, differentiated cells of claim 47, while the claim further limits the type of data received in claim 1, it does not change the fact that the additional elements are mere data gathering steps
Claim 61 recites that the set of constituents and modules in claim 2 are 2-20 cellular constituents and 1 cellular constituent module, or 2 - 100 cellular constituents and ≥2 cellular constituent module, or 2-1000 cellular constituents and ≥5 cellular constituent module, although the claim further limits the range of the data being rearranged, it does not change the fact that the process of data rearrangement is a mental process/mathematical concept
Claim 67 recites a computer system, comprising one or more processors and memory, that stores instructions, which equates to a generic computing environment/computer
Claim 69 recites a non-transitory computer-readable medium storing one or more computer programs, executable by a computer, the computer comprising one or more processors and a memory, which equates to a generic computing environment/computer
All of these additional elements provide further definitions and clarifications to the limitations of the inventions. However, they do not help to integrate the judicial exceptions into a practical application or use. There are no limitations that indicate that the claimed computer system or the formats of the provided data require anything other than generic computing systems. As such, these limitations equate to mere instructions to implement the abstract idea on a generic computer that the courts have stated does not render an abstract idea eligible in Alice Corp., 573 U.S. at 223, 110 USPQ2d at 1983. See also 573 U.S. at 224, 110 USPQ2d at 1984. In general, linking the use of an abstract idea to a particular technological environment, such as a computer, does not integrate the abstract idea into a practical application based on MPEP 2106.05(h). Therefore, claims 1-3, 6, 12, 16, 18-20, 25-26, 28, 31, 34, 37, 40-41, 43, 46-48, 61, 64-65, 67, and 69 are directed to an abstract idea and/or natural phenomenon as the additional elements do not integrate the judicial exceptions into a practical application (Step 2A, Prong 2: NO).
Claims found to be directed to a judicial exception are then further evaluated to determine if the claims recite an inventive concept that provides significantly more than the judicial exception itself (Step 2B). In the instant application, claims 1-3, 6, 12, 16, 18-20, 25-26, 28, 31, 34, 37, 40-41, 43, 46-48, 61, 64-65, 67, and 69 do not recite further limitations or specification to the additional element that would indicate anything other than obtaining cellular and chemical data through process known in the art as stated in the Specification of the instant application, that is then analyze and outputted in a genic computer. For example, [00122] in the instant Specification cites various literatures to obtain chemical fingerprints through the known processes of BCI, Daylight, ECFC4, ECFP4, MDL, Unity 2D, atom pair fingerprints (APFP), and topological torsion fingerprints (TTFP), while [00198-00199] and [00211] talk about processes, known in the art, to obtain cellular constituent data from RNA and dimension reduction techniques to determine cellular constituent abundances. As discussed above, there are no additional limitations to indicate that the claimed analysis engine requires anything other than generic computer components in order to carry out the recited abstract idea in the claims. The additional elements do not comprise an inventive concept when considered individually or as an ordered combination that transforms the claimed judicial exception into a patent-eligible application of the judicial exception. Therefore, the claims do not amount to significantly more than the judicial exception itself (Step 2B: No). As such, claims 1-3, 6, 12, 16, 18-20, 25-26, 28, 31, 34, 37, 40-41, 43, 46-48, 61, 64-65, 67, and 69 are not patent eligible.
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 for establishing a background for determining obviousness under 35 U.S.C. 103 are summarized as follows:
1. Determining the scope and contents of the prior art.
2. Ascertaining the differences between the prior art and the claims at issue.
3. Resolving the level of ordinary skill in the pertinent art.
4. 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 1-3, 6, 16, 18-20, 26, 28, 34, 37, 41, 43, 46-48, 61, 64-65, 67, and 69 are rejected under 35 U.S.C. 103 over Narain et al. (US 2018/0260515 A1, Published 2018 Sep 13th) in view of Lotfollahi et al. (bioRxiv preprint doi.org/10/1101/478503 pgs. 1-27, Published 2018 Nov 28th). The limitations of the instant claim are italicized below.
With respect to claim 1, Narain et al. teaches a method for identifying a modulator of a disease process with a computer-implemented model (claim 1, A method of training a model to identify a set of cellular constituents associated with a cellular process of interest). Narain et al. starts by obtaining various datasets which include information representing the expression levels of a plurality of genes titled the “first data set” and “third data set” (Pg. 50, para [0146]-[0147], (A) obtaining one or more first datasets in electronic form … for each respective cellular constituent in a plurality of cellular constituents … a corresponding abundance of the respective cellular constituent in the respective cell,…each respective element in the corresponding plurality of elements having a corresponding count representing the corresponding abundance of the respective cellular constituent in a respective cell in the first plurality of cells). Narain et al. stated that number of genes selected could vary depending on the purpose of the model, but recited 50 or more conditions reflecting or stimulating different characteristic aspect of the biological system could be investigated (Pg. 62-63 para [0318], wherein the plurality of cellular constituents comprises 50 or more cellular constituents). These datasets of genes represent cells that may be diseases or experienced perturbations with an agent (claim 1, claim 18, for each respective cell in a first plurality of cells, wherein the first plurality of cells … (ii) collectively represents a plurality of annotated cell states). Narain et al. goes on to obtain a second set of data representing the measured functional activity or cellular response (claim 1, (C) obtaining one or more second datasets in electronic form, the one or more second datasets individually or collectively comprising: for each respective cell in a second plurality of cells, wherein the second plurality of cells comprises … (ii) collectively represents a plurality of covariates associated with the cellular process of interest: for each respective cellular constituent in the plurality of cellular constituents: a corresponding abundance of the respective cellular constituent in the respective cell). Narain et al. later specifies that multiple cells of the same or different origins can be included in the cell model, and at least 20 or more cells can be included (Pg. 62 para [0314], for each respective cell in a first plurality of cells, wherein the first plurality of cells (i) comprises twenty or more cells, wherein the second plurality of cells comprises (i) twenty or more cells). Next, Narain et al. states that these data sets can then be used to construct a network fragment library (step 214), wherein the fragments define quantitative, continuous relationships among all possible small sets of measured variables (Pg. 59 para [0279], (B) using the plurality of vectors to identify each respective cellular constituent module in a plurality of cellular constituent modules, each respective cellular constituent module in the plurality of cellular constituent modules including a subset of the plurality of cellular constituents, wherein the plurality of cellular constituent modules are arranged in a latent representation dimensioned by (i) the plurality of cellular constituent modules and (ii) the plurality of cellular constituents or a plurality of dimension reduction components representing the plurality of cellular constituents). In addition, Narain et al. teaches that the network fragments are utilized to create an ensemble of initial trial networks (1000 networks) that undergo optimization by adding, subtracting or substitution from the library based on their association to a molecular mechanism in a biological process (Pg. 51-52 para [0164], wherein the plurality of cellular constituent modules comprises more than ten cellular constituent modules). Afterwards, Narain et al. takes these inputs and gives it to a computer-implemented model to create “generated cell model networks”, that include quantitative probabilistic directional information regarding causal relationships between the expression level of the plurality of genes and the functional activity or cellular responses (Pg. 50 para [0153], (D) forming an activation data structure by combining the cellular constituent count data structure and the latent representation using the plurality of cellular constituents or the plurality of dimension reduction components as a common dimension). Lastly, Narain et al. compares the “generated cell model network” which contains calculated quantitative parameter relative to the cells being evaluated, and the “generated comparison cell model network” which contains parameters relative to control cells, with a differential (delta) network module to identify significantly different parameters for adjustment (Pg. 51 para [0159] and Pg. 60 para [0294], (E) training the model using, for each respective covariate in the plurality of covariates, a difference between (i) a calculated activation against each cellular constituent module represented by the model upon input of a representation of the respective covariate into the model and (ii) measured activation against each cellular constituent module represented by the model, wherein the training adjusts a plurality of covariate parameters associated with the model responsive to the difference, wherein each respective covariate parameter in the plurality of covariate parameters represents a covariate in the plurality of covariates).
Concerning claim 2, Narain et al. discloses that quantitative relationship information may be identified for each relationship in the generated cell model networks (step 30) and may include a direction indicating causality, and/or an expression of the quantitative magnitude of the strength of the relationship (Pg. 51 para [0162]). Narain et al. further teaches that the quantitative relationship information can be used to explore each hub of activity in the networks as a potential therapeutic target and/or biomarker (Pg. 51 para [0162], (F) identifying, using the plurality of covariate parameters upon training the model, one or more cellular constituent modules in the plurality of cellular constituent modules that is associated with one or more covariates in the plurality of covariates, thereby associating each cellular constituent in the set of cellular constituents from the plurality of cellular constituents, from among the cellular constituents in the identified one or more cellular constituent modules, with the cellular process of interest).
With regards to claim 3, Narain et al. discloses that the cells from the first dataset can represent cells that have experienced an environmental perturbation which comprises of a contact with an agent and/or a change in culture condition (Claim 17). Narain et al. goes on to give examples of cells models being exposed to different perturbation with Coenzyme 10 at varying concentrations of 50 uM or 100 uM (Pg. 96 para [0728], an annotated cell state in the plurality of annotated cell states is an exposure of a cell in the first plurality of cells to a compound under an exposure condition, wherein the exposure condition is a duration of exposure, a concentration of the compound, or a combination of a duration of exposure and a concentration of the compound).
Concerning claim 6, Narain et al. teaches the data in the first and third data set is usually related to the level of certain macromolecules, such as DNA, RNA, protein, lipid, etc. and gives the example of proteomic data (Pg. 57 para [0254], each cellular constituent in the plurality of cellular constituents is a particular gene, a particular mRNA associated with a gene, a carbohydrate, a lipid, an epigenetic feature, a metabolite, a protein, or a combination thereof). Narain et al. goes on to elaborate how commercial techniques and kits (flow cytometry, cell-based assays, functional assays, RNA isolation, qPCR) can be utilized to acquire the data (Pg. 42 para [0008] and Pg. 57 para [0255], the corresponding abundance of the respective cellular constituent in the respective cell in the first or second plurality of cells is determined by a colorimetric measurement, a fluorescence measurement, a luminescence measurement, a resonance energy transfer (FRET) measurement, single-cell ribonucleic acid (RNA) sequencing (scRNA-seq), scTag-seq, single-cell assay for transposase-accessible chromatin using sequencing (scATAC-seq), CyTOF/SCoP, E-MS/Abseq, miRNA-seq, CITE-seq, or any combination thereof).
With respect to claim 16, Narain et al. discloses the use of a plurality of genes in cells associated with a biological process, implying that numerous genes are associated in different processes (claim 1, Pg. 43 para [0013]). Figures 23, 28, and 40 also demonstrates the genes that were measured and utilized in the model, with the cumulative sum of the genes being over at least 20 but under 10,000 (the plurality of cellular constituents consists of between twenty and 10,000 cellular constituents).
Regarding claim 18, Narain et al. teaches that the network fragments define relationships among all possible small sets (e.g., 2-3 member sets or 2-4 member sets) of measured variables (Pg. 59 para [0279], each cellular constituent module in the plurality of cellular constituent modules consists of between two cellular constituents and three hundred cellular constituents).
Concerning claim 19, Narain et al. discloses that the invention is attempting to model biological systems associated with diseases and perturbations (Pg. 42 para [0008], the cellular process of interest is an aberrant cell process associated with a disease). Narain et al. goes on to state that the cell models can include one or more “control cells” which may be untreated or unperturbed cells or normal, non-diseased cells (Pg. 62 para [0316], the first plurality of cells includes cells that are representative of the disease and cells that are not representative of the disease as indicated by the plurality of annotated cell states).
With regards to claim 20, Narain et al. teaches that the invention is attempting to model biological systems associated with diseases and perturbations and utilizes cells representing these conditions in the analysis (Pg. 42 para [0008] and Pg. 62 para [0316], a respective covariate in the plurality of covariates comprises disease status and the representation of the respective covariate is an indication of absence or presence of the disease).
Regarding claim 26, Narain et al. discloses that the cells from the first dataset can represent cells that have experienced an environmental perturbation which comprises of a contact with an agent and/or a change in culture condition (Claim 17). Narain et al. goes on to give examples of cells models being exposed to different perturbation with Coenzyme 10 at varying concentrations of 50 uM or 100 uM (Pg. 96 para [0728], the representation of the respective covariate further comprises a duration of time the respective covariate was incubated with the respective cell, and wherein the representation of the respective covariate further comprises a concentration of the respective covariate used to incubate the respective cell).
With respect to claim 28, Narain et al. discloses the use of scoring functions in the model to evaluate how likely a model fits the input data (Pg. 59 - Pg. 60 para [0284]). The multivariate system utilizes a plurality of negative logarithm functions to score each individual network fragment, with higher scores indicating that the model fits the input data, and lower scores indicating otherwise (Pg. 59 - Pg. 60 para [0284], the training the model (E) is performed using a categorical cross-entropy loss in a multi-task formulation, in which each covariate in the plurality of covariates corresponds to a cost function in a plurality of cost functions and each respective cost function in the plurality of cost functions has a common weighting factor).
Concerning Claim 34, Narain et al. teaches the use of external stimulus components to perturb the cells and gives the example of “Multidimensional Intracellular Molecules (MIMs)” that induce signal transduction and/or gene expression mechanisms within a cell (Pg. 54 para [0203] - [0204]). Narain et al. further specifies an example of tyrosine, which is an organic compound (amino acid) that has a molecular weight of 181.19 Da (<500 Da for Lipinski rule), 3 hydrogen bond donor (<5 for Lipinski rule), and 4 hydrogen bond acceptors (<10 for Lipinski rule) (Pg. 54 para [0206], the compound is an organic compound having a molecular weight of less than 2000 Daltons that satisfies at least three criteria of the Lipinski rule of five criteria).
With regards to claim 37, Narain et al. teaches that the network fragments are utilized to create an ensemble of initial trial networks (1000 networks) that undergo optimization by adding, subtracting or substitution from the library based on their association to a molecular mechanism in a biological process (Pg. 51-52 para [0164], the plurality of cellular constituent modules comprises five or more cellular constituent modules).
Regarding claim 41, Narain et al. teaches that the network fragments define relationships among all possible small sets (e.g., 2-3 member sets or 2-4 member sets) of measured variables (Pg. 59 para [0279], the subset of the plurality of cellular constituents in the respective cellular constituent modules consists of between two and 20 cellular constituents in a molecular pathway associated with the cellular process of interest).
Concerning claim 43, Narain et al. discloses that the invention could utilize a variety of models including linear regression, logistic regression, polynomial regression, and Bayesian network (Pg. 59 para [0279], the model is a logistic regression model, a neural network model, a support vector machine model, a Naive Bayes model, a nearest neighbor model, a boosted trees model, a random forest model, a decision tree model, a multinomial logistic regression model, a linear model, or a linear regression model).
With respect to claim 46, Narain et al. discloses how commercial techniques and kits (flow cytometry, cell-based assays, functional assays, RNA isolation, qPCR) can be utilized to acquire data pertaining to the cells in the first and second datasets (Pg. 42 para [0008] and Pg. 57 para [0255], a corresponding abundance of the respective cellular constituent in the respective cell is determined using a cell-based assay).
Regarding claim 47, Narain et al. teaches that the cells used for analysis can be from a number of origins and includes non-limiting examples of cancer cell lines, immortal cell lines, normal cell lines, primary cells, cells freshly isolated from live tissue or organs (Pg. 54 para [0198], the first plurality of cells or the second plurality of cells comprises or consists of: cells from an organ, cells from a tissue, a plurality of stem cells, a plurality of primary human cells, a plurality of human cell lines, cells from umbilical cord blood, from peripheral blood, or from bone marrow, cells in or from a solid tissue, or a plurality of differentiated cells).
Concerning claim 48, Narain et al. discloses specific examples of cells utilized for analysis such as PaCa2 (human pancreatic cancer cell line), HepG2 (human liver epithelial cells, cancerous), PC3 (human prostate cancer cell line), MCF7 (human breast cancer cell line), THLE2 (human liver epithelial cell line), and HDFa (human skin fibroblast cell line) that were experimented on and for which data was collected (Pg. 56 para [0249], the organ is heart, liver, lung, muscle, brain, pancreas, spleen, kidney, small intestine, uterus, or bladder, the tissue is bone, cartilage, joint, tracheae, spinal cord, cornea, eye, skin, or blood vessel the plurality of stem cells is a plurality of embryonic stem cells, a plurality of adult stem cells, or a plurality of induced pluripotent stem cells (iPSC),the plurality of primary human cells are a plurality of CD34+ cells, a plurality of CD34+ hematopoietic stems, a plurality of progenitor cells (HSPC), a plurality of T-cells, a plurality of mesenchymal stem cells (MSC), a plurality of airway basal stem cells, or a plurality of induced pluripotent stem cells, the solid tissue is placenta, liver, heart, brain, kidney, or gastrointestinal tract, and the plurality of differentiated cells is a plurality of megakaryocytes, a plurality of osteoblasts, a plurality of chondrocytes, a plurality of adipocytes, a plurality of hepatocytes, a plurality of hepatic mesothelial cells, a plurality of biliary epithelial cells, a plurality of hepatic stellate cells, a plurality of hepatic sinusoid endothelial cells, a plurality of Kupffer cells, a plurality of pit cells, a plurality of vascular endothelial cells, a plurality of pancreatic duct epithelial cells, a plurality of pancreatic duct cells, a plurality of centroacinous cells, a plurality of acinar cells, a plurality of islets of Langerhans, a plurality of cardiac muscle cells, a plurality of fibroblasts, a plurality of keratinocytes, a plurality of smooth muscle cells, a plurality of type I alveolar epithelial cells, a plurality of type II alveolar epithelial cells, a plurality of Clara cells, a plurality of ciliated epithelial cells, a plurality of basal cells, a plurality of goblet cells, a plurality of neuroendocrine cells, a plurality of kultschitzky cells, a plurality of renal tubular epithelial cells, a plurality of urothelial cells, a plurality of columnar epithelial cells, a plurality of glomerular epithelial cells, a plurality of glomerular endothelial cells, a plurality of podocytes, a plurality of mesangium cells, a plurality of nerve cells, a plurality of astrocytes, a plurality of microglia, or a plurality of oligodendrocytes).
With respect to claim 61, Narain et al. discloses the use of a plurality of genes in cells associated with a biological process, implying that numerous genes are associated in different processes (claim 1, Pg. 43 para [0013]). Figures 23, 28, and 40 also demonstrates the genes that were measured and utilized in the model, with the cumulative sum of the genes being over at least 20 (the set of cellular constituents consists of between 2 and 1000 cellular constituents in the plurality of cellular constituents). In addition, Narain et al. teaches that the network fragments (subsets) are utilized to create an ensemble of initial trial networks (1000 networks fragments) that undergo optimization by adding, subtracting or substitution from the library based on their association to a molecular mechanism in a biological process (Pg. 51-52 para [0164], the one or more cellular constituent modules comprises five or more cellular constituent modules).
Regarding claim 64, Narain et al. teaches the use of an ensemble of Bayesian networks to generate casual relationships between the expression level of the plurality of genes and the functional activity or cellular responses, that include quantitative probabilistic directional information regarding the relations (Pg. 50 para [0153], the model is an ensemble model comprising a plurality of component models, and wherein each respective component model in the plurality of component models provides a calculated activation for a different cellular constituent module in the plurality of cellular constituent modules responsive to inputting the representation of the respective covariate into the respective component model).
Concerning claim 65, Narain et al. discloses that the invention could utilize a variety of models including linear regression, logistic regression, polynomial regression, and Bayesian network (Pg. 59 para [0279], wherein a component model in the plurality of component models is a logistic regression model, a neural network model, a support vector machine model, a Naive Bayes model, a nearest neighbor model, a boosted trees model, a random forest model, a decision tree model, a multinomial logistic regression model, a linear model, or a linear regression model).
Regarding claim 67, Narain et al. teaches a computing device (100), with a memory (106) and a processor (102) that executes software stored in the memory for a method for identifying a modulator of a disease process with a computer-implemented model (claim 1, Pg. 61 para [0303], A computer system, comprising one or more processors and memory, the memory storing instructions for performing a method for training a model to identify a set of cellular constituents associated with a cellular process of interest). Narain et al. starts by obtaining various datasets which include information representing the expression levels of a plurality of genes titled the “first data set” and “third data set” (Pg 50, para [0146]-[0147], (A) obtaining one or more first datasets in electronic form … for each respective cellular constituent in a plurality of cellular constituents … a corresponding abundance of the respective cellular constituent in the respective cell, thereby accessing or forming a plurality of vectors, each respective vector in the plurality of vectors (i) corresponding to a respective cellular constituent in the plurality of cellular constituents and (ii) comprising a corresponding plurality of elements, each respective element in the corresponding plurality of elements having a corresponding count representing the corresponding abundance of the respective cellular constituent in a respective cell in the first plurality of cells). Narain et al. stated that number of genes selected could vary depending on the purpose of the model, but recited 50 or more conditions reflecting or stimulating different characteristic aspect of the biological system could be investigated (Pg. 62-63 para [0318], wherein the plurality of cellular constituents comprises 50 or more cellular constituents). These datasets of genes represent cells that may be diseases or experienced perturbations with an agent (claim 1, claim 18, for each respective cell in a first plurality of cells, wherein the first plurality of cells … (ii) collectively represents a plurality of annotated cell states). Narain et al. goes on to obtain a second set of data representing the measured functional activity or cellular response (claim 1, (C) obtaining one or more second datasets in electronic form, the one or more second datasets individually or collectively comprising: for each respective cell in a second plurality of cells, wherein the second plurality of cells comprises … (ii) collectively represents a plurality of covariates associated with the cellular process of interest: for each respective cellular constituent in the plurality of cellular constituents: a corresponding abundance of the respective cellular constituent in the respective cell, thereby obtaining a cellular constituent count data structure dimensioned by (i) the second plurality of cells and (ii) the plurality of cellular constituents or the plurality of dimension reduction components). Narain et al. later specifies that multiple cells of the same or different origins can be included in the cell model, and at least 20 or more cells can be included (Pg. 62 para [0314], for each respective cell in a first plurality of cells, wherein the first plurality of cells (i) comprises twenty or more cells, wherein the second plurality of cells comprises (i) twenty or more cells). Next, Narain et al. states that these data sets can then be used to construct a network fragment library (step 214), wherein the fragments define quantitative, continuous relationships among all possible small sets of measured variables (Pg. 59 para [0279], (B) using the plurality of vectors to identify each respective cellular constituent module in a plurality of cellular constituent modules, each respective cellular constituent module in the plurality of cellular constituent modules including a subset of the plurality of cellular constituents, wherein the plurality of cellular constituent modules are arranged in a latent representation dimensioned by (i) the plurality of cellular constituent modules and (ii) the plurality of cellular constituents or a plurality of dimension reduction components representing the plurality of cellular constituents). In addition, Narain et al. teaches that the network fragments are utilized to create an ensemble of initial trial networks (1000 networks) that undergo optimization by adding, subtracting or substitution from the library based on their association to a molecular mechanism in a biological process (Pg. 51-52 para [0164], wherein the plurality of cellular constituent modules comprises more than ten cellular constituent modules). Afterwards, Narain et al. takes these inputs and gives it to a computer-implemented model to create “generated cell model networks”, that include quantitative probabilistic directional information regarding causal relationships between the expression level of the plurality of genes and the functional activity or cellular responses (Pg. 50 para [0153], (D) forming an activation data structure by combining the cellular constituent count data structure and the latent representation using the plurality of cellular constituents or the plurality of dimension reduction components as a common dimension, wherein the activation data structure comprises, for each cellular constituent module in the plurality of cellular constituent modules: for each cell in the second plurality of cells, a respective activation weight). Lastly, Narain et al. compares the “generated cell model network” to a “generated comparison cell model network” with a differential (delta) network module to identify significantly different parameters for adjustment (Pg. 51 para [0159] and Pg. 60 para [0294], (E) training the model using, for each respective covariate in the plurality of covariates, a difference between (i) a calculated activation against each cellular constituent module represented by the model upon input of a representation of the respective covariate into the model and (ii) measured activation against each cellular constituent module represented by the model, wherein the training adjusts a plurality of covariate parameters associated with the model responsive to the difference, wherein each respective covariate parameter in the plurality of covariate parameters represents a covariate in the plurality of covariates).
Concerning claim 69, Narain et al. teaches a computing system/environment that has a computing device (100), with a memory (106), a processor (102) and non-transitory computer readable media as storage (116) which stores code (150) for a method for identifying a modulator of a disease process with a computer-implemented model (claim 1, Pg. 60 para [0298] and Pg. 61 para [0303], A non-transitory computer-readable medium storing one or more computer programs, executable by a computer, a method for training a model to identify a set of cellular constituents associated with a cellular process of interest, the computer comprising one or more processors and a memory, the one or more computer programs collectively encoding computer executable instructions that perform a method). Narain et al. starts by obtaining various datasets which include information representing the expression levels of a plurality of genes titled the “first data set” and “third data set” (Pg 50, para [0146]-[0147], (A) obtaining one or more first datasets in electronic form … for each respective cellular constituent in a plurality of cellular constituents … a corresponding abundance of the respective cellular constituent in the respective cell, thereby accessing or forming a plurality of vectors, each respective vector in the plurality of vectors (i) corresponding to a respective cellular constituent in the plurality of cellular constituents and (ii) comprising a corresponding plurality of elements, each respective element in the corresponding plurality of elements having a corresponding count representing the corresponding abundance of the respective cellular constituent in a respective cell in the first plurality of cells). Narain et al. stated that number of genes selected could vary depending on the purpose of the model, but recited 50 or more conditions reflecting or stimulating different characteristic aspect of the biological system could be investigated (Pg. 62-63 para [0318], wherein the plurality of cellular constituents comprises 50 or more cellular constituents). These datasets of genes represent cells that may be diseases or experienced perturbations with an agent (claim 1, claim 18, for each respective cell in a first plurality of cells, wherein the first plurality of cells … (ii) collectively represents a plurality of annotated cell states). Narain et al. goes on to obtain a second set of data representing the measured functional activity or cellular response (claim 1, (C) obtaining one or more second datasets in electronic form, the one or more second datasets individually or collectively comprising: for each respective cell in a second plurality of cells, wherein the second plurality of cells comprises … (ii) collectively represents a plurality of covariates associated with the cellular process of interest: for each respective cellular constituent in the plurality of cellular constituents: a corresponding abundance of the respective cellular constituent in the respective cell, thereby obtaining a cellular constituent count data structure dimensioned by (i) the second plurality of cells and (ii) the plurality of cellular constituents or the plurality of dimension reduction components). Narain et al. later specifies that multiple cells of the same or different origins can be included in the cell model, and at least 20 or more cells can be included (Pg. 62 para [0314], for each respective cell in a first plurality of cells, wherein the first plurality of cells (i) comprises twenty or more cells, wherein the second plurality of cells comprises (i) twenty or more cells). Next, Narain et al. states that these data sets can then be used to construct a network fragment library (step 214), wherein the fragments define quantitative, continuous relationships among all possible small sets of measured variables (Pg. 59 para [0279], (B) using the plurality of vectors to identify each respective cellular constituent module in a plurality of cellular constituent modules, each respective cellular constituent module in the plurality of cellular constituent modules including a subset of the plurality of cellular constituents, wherein the plurality of cellular constituent modules are arranged in a latent representation dimensioned by (i) the plurality of cellular constituent modules and (ii) the plurality of cellular constituents or a plurality of dimension reduction components representing the plurality of cellular constituents). In addition, Narain et al. teaches that the network fragments are utilized to create an ensemble of initial trial networks (1000 networks) that undergo optimization by adding, subtracting or substitution from the library based on their association to a molecular mechanism in a biological process (Pg. 51-52 para [0164], wherein the plurality of cellular constituent modules comprises more than ten cellular constituent modules). Afterwards, Narain et al. takes these inputs and gives it to a computer-implemented model to create “generated cell model networks”, that include quantitative probabilistic directional information regarding causal relationships between the expression level of the plurality of genes and the functional activity or cellular responses (Pg. 50 para [0153], (D) forming an activation data structure by combining the cellular constituent count data structure and the latent representation using the plurality of cellular constituents or the plurality of dimension reduction components as a common dimension, wherein the activation data structure comprises, for each cellular constituent module in the plurality of cellular constituent modules: for each cell in the second plurality of cells, a respective activation weight). Lastly, Narain et al. compares the “generated cell model network” to a “generated comparison cell model network” with a differential (delta) network module to identify significantly different parameters for adjustment (Pg. 51 para [0159] and Pg. 60 para [0294], (E) training the model using, for each respective covariate in the plurality of covariates, a difference between (i) a calculated activation against each cellular constituent module represented by the model upon input of a representation of the respective covariate into the model and (ii) measured activation against each cellular constituent module represented by the model, wherein the training adjusts a plurality of covariate parameters associated with the model responsive to the difference, wherein each respective covariate parameter in the plurality of covariate parameters represents a covariate in the plurality of covariates).
However, Narain et al. does not teach the specific use of single-cell RNA-sequencing (scRNA-seq) data and their matrices for data analysis. Hence, Narain et al. does not disclose the use of matrices and vectors to define the modules. However, the use of scRNA-seq data in the form of count matrices representing gene expression were known in the art before the effective filing date of the instant application as taught by Lotfollahi et al.
With respect to claim 1, Lotfollahi et al. teaches the specific use of scRNA-seq datasets in latent space to predict single-cell perturbation response across cell types (pg. 1 Abstract). Lotfollahi et al. teaches the use of a first sc-RNA seq dataset representing two groups of peripheral blood mononuclear cells (PMBCs), 14446 cells were stimulated with interferon-β and 14619 were control cells (pg. 14 para 3). Lotfollahi et al. clarifies that the dataset is from Kang et al. (Nature Biotechnology 36, pgs. 89-94, 2017) with gene expression omnibus code GSE96583 (pg. 14 para 3). Through accessing the raw dataset, specifically the mat files, one can see that there are count matrices with the genes as rows, and cells as columns (thereby accessing or forming a plurality of vectors, each respective vector in the plurality of vectors (i) corresponding to a respective cellular constituent in the plurality of cellular constituents and (ii) comprising a corresponding plurality of elements). Lotfollahi et al. then discloses that a Spearman correlation between every single cell and 8 cluster averages was calculated and each cell was assigned to the cell type that it had a maximum correlation with (pg. 14 para 3, (B) using the plurality of vectors to identify each respective cellular constituent module in a plurality of cellular constituent modules). In addition, Lotfollahi et al. also teaches the use of a second PMBC dataset from Zheng et al. (Nature Communications 8, 14049, 2017) which again is a sc-RNA seq dataset with similar count matrices to Kang et al. (pg. 16 para 3, thereby obtaining a cellular constituent count data structure dimensioned by (i) the second plurality of cells and (ii) the plurality of cellular constituents). Then, Lotfollahi et al. teaches the combination of the datasets into latent space (pg. 15 para 3). Lastly, Lotfollahi et al. teaches the calculation of a delta parameter (δ), that represents the difference between cells in condition 0 and 1 (unperturbed/control vs. perturb/stimulated) from the gene expression datasets, which can be used to estimate the activation in gene expression from a cell in condition 0 to a cell in condition 1 (pg. 13 para 3, wherein the activation data structure comprises, for each cellular constituent module in the plurality of cellular constituent modules: for each cell in the second plurality of cells, a respective activation weight).
With respect to claim 67, Lotfollahi et al. teaches the specific use of scRNA-seq datasets in latent space to predict single-cell perturbation response across cell types (pg. 1 Abstract). Lotfollahi et al. teaches the use of a first sc-RNA seq dataset representing two groups of peripheral blood mononuclear cells (PMBCs), 14446 cells were stimulated with interferon-β and 14619 were control cells (pg. 14 para 3). Lotfollahi et al. clarifies that the dataset is from Kang et al. (Nature Biotechnology 36, pgs. 89-94, 2017) with gene expression omnibus code GSE96583 (pg. 14 para 3). Through accessing the raw dataset, specifically the mat files, one can see that there are count matrices with the genes as rows, and cells as columns (thereby accessing or forming a plurality of vectors, each respective vector in the plurality of vectors (i) corresponding to a respective cellular constituent in the plurality of cellular constituents and (ii) comprising a corresponding plurality of elements). Lotfollahi et al. then discloses that a Spearman correlation between every single cell and 8 cluster averages was calculated and each cell was assigned to the cell type that it had a maximum correlation with (pg. 14 para 3, (B) using the plurality of vectors to identify each respective cellular constituent module in a plurality of cellular constituent modules). In addition, Lotfollahi et al. also teaches the use of a second PMBC dataset from Zheng et al. (Nature Communications 8, 14049, 2017) which again is a sc-RNA seq dataset with similar count matrices to Kang et al. (pg. 16 para 3, thereby obtaining a cellular constituent count data structure dimensioned by (i) the second plurality of cells and (ii) the plurality of cellular constituents). Then, Lotfollahi et al. teaches the combination of the datasets into latent space (pg. 15 para 3). Lastly, Lotfollahi et al. teaches the calculation of a delta parameter (δ), that represents the difference between cells in condition 0 and 1 (unperturbed/control vs. perturb/stimulated) from the gene expression datasets, which can be used to estimate the activation in gene expression from a cell in condition 0 to a cell in condition 1 (pg. 13 para 3, wherein the activation data structure comprises, for each cellular constituent module in the plurality of cellular constituent modules: for each cell in the second plurality of cells, a respective activation weight).
With respect to claim 69, Lotfollahi et al. teaches the specific use of scRNA-seq datasets in latent space to predict single-cell perturbation response across cell types (pg. 1 Abstract). Lotfollahi et al. teaches the use of a first sc-RNA seq dataset representing two groups of peripheral blood mononuclear cells (PMBCs), 14446 cells were stimulated with interferon-β and 14619 were control cells (pg. 14 para 3). Lotfollahi et al. clarifies that the dataset is from Kang et al. (Nature Biotechnology 36, pgs. 89-94, 2017) with gene expression omnibus code GSE96583 (pg. 14 para 3). Through accessing the raw dataset, specifically the mat files, one can see that there are count matrices with the genes as rows, and cells as columns (thereby accessing or forming a plurality of vectors, each respective vector in the plurality of vectors (i) corresponding to a respective cellular constituent in the plurality of cellular constituents and (ii) comprising a corresponding plurality of elements). Lotfollahi et al. then discloses that a Spearman correlation between every single cell and 8 cluster averages was calculated and each cell was assigned to the cell type that it had a maximum correlation with (pg. 14 para 3, (B) using the plurality of vectors to identify each respective cellular constituent module in a plurality of cellular constituent modules). In addition, Lotfollahi et al. also teaches the use of a second PMBC dataset from Zheng et al. (Nature Communications 8, 14049, 2017) which again is a sc-RNA seq dataset with similar count matrices to Kang et al. (pg. 16 para 3, thereby obtaining a cellular constituent count data structure dimensioned by (i) the second plurality of cells and (ii) the plurality of cellular constituents). Then, Lotfollahi et al. teaches the combination of the datasets into latent space (pg. 15 para 3). Lastly, Lotfollahi et al. teaches the calculation of a delta parameter (δ), that represents the difference between cells in condition 0 and 1 (unperturbed/control vs. perturb/stimulated) from the gene expression datasets, which can be used to estimate the activation in gene expression from a cell in condition 0 to a cell in condition 1 (pg. 13 para 3, wherein the activation data structure comprises, for each cellular constituent module in the plurality of cellular constituent modules: for each cell in the second plurality of cells, a respective activation weight).
It would have been prima facie obvious to one of ordinary skill in the art at the effective filing date of the invention to implement the scRNA-seq datasets and preprocessing steps of Lotfollahi et al. with the network model of Narain et al. to allow for higher sensitivity and more in-depth detection of minute changes in cellular activity. scRNA sequencing also presents as an attractive alternative to the proteomic process (used in Narain et al.), as scRNA sequencing is relatively low cost (does not require antibodies), has higher throughput and a well-established experimentation pipeline. One of ordinary skill in the art would have been motivated to utilize the scRNA-seq data of Lotfollahi et al. for more sensitive predictions as scRNA-seq data can capture low-abundance transcripts (mRNA) that proteomics can struggle to capture, and the variational encoder would have given the model the ability to process the high-dimensional data into latent space for easier interpretation. Furthermore, one of skill in the art before the effective filing date of the claimed invention would have a reasonable expectation of success at incorporating scRNA-seq dataset and variational autoencoder into the network model of Narain et al. as the addition of neural network layers within a multi-layer network is a well-known and established technique.
Claims 12 and 40 are rejected under 35 U.S.C. 103 as being unpatentable over Narain et al. and Lotfollahi et al. as applied to claim 1 above, and further in view of Liu et al. (Computational and Mathematical Methods in Medicine, Published October 24th 2016, Article 3456153). The limitations of the instant application are italicized below.
The limitations of claim 1 has been taught by Narain et al. and Lotfollahi et al. above.
With respect to claim 12, although Narain et al. classifies the gene expression variables into network fragments for selection, they do not explicitly utilize a dictionary learning model such as the L0-regularized autoencoder to do so. Instead, Narain et al. opted to utilized the Bayesian information criterion (BIC) and Bayesian fragment enumeration method (Pg. 51 para [0163]). However, Liu et al. does teach the use of L0-regularized autoencoders in variable selection for high-dimensional big data like genomic data and states that their proposed algorithm is efficient at constructing biologically important networks with high-dimensional big data (Abstract). Liu et al. further emphasized that exhaustive search with BIC over all possible combinations of features is computationally infeasible with high dimensional big data, hence the motivation for their model (Pg 1 right col first para). Liu et al. demonstrates that the L0-regularized regression was able to successfully identify biologically meaningful subnetworks (modules) within real ovarian cancer data (multisource gene expression data) that indicate patient survival (Pg 9 left col first para, using the plurality of vectors to identify each cellular constituent module in the plurality of cellular constituent modules comprises a dictionary learning model that produces the plurality of dimension reduction components, wherein the dictionary learning model is an L0-regularized autoencoder).
Regarding claim 40, although Narain et al. classifies the gene expression variables into network fragments for selection, they do not explicitly state that there are more than 5 variables per network. However, Liu et al. demonstrates all of the above in addition to showing that subnetworks (modules) can be made up of more than 5 variables (Pg 9 Fig 2). Specifically, in the case of the ovarian cancer data set, Liu et al. generated a subnetwork of 22 genes that were later identified to be biologically relevant (Pg 9 right col para 1, the plurality of cellular constituents in the respective cellular constituent module comprises five or more cellular constituents).
It would have been prima facie obvious to one of ordinary skill in the art at the effective filing date of the invention to implement the L0-regularized regression of Liu et al. with the network model of Narain et al. to allow for more efficient and accurate transformation of high dimensional data, such as gene expression, into sparse and more interpretable representations. As stated in Liu et al. one of ordinary skill in the art would have been motivated to incorporate the L0-regularized regressions in order to overcome BIC’s challenge of not being able to conduct an exhaustive search. In addition, the Specification of the instant application stated that one would also be motivated to incorporate the L0-regularized autoencoders in order to not enforce a 1:1 correspondence between modules and cellular constituents, allowing the constituents to appear in several modules at the same time ([00229]). This would have also allowed for an increased number of constituents per modules as seen in claim 40. Furthermore, one of skill in the art before the effective filing date of the claimed invention would have a reasonable expectation of success at incorporating the L0-regularized regression into the network model of Narain et al., as Liu et al. has already demonstrated success at incorporating the L0-regularized regression into known gene networks to create subnetworks for analysis.
Claim 25 is rejected under 35 U.S.C. 103 as being unpatentable over Narain et al. and Lotfollahi et al. as applied to claims 1 and 20 above, and further in view of Gao et al. (Phys Chem Phys., Publish March 20th 2020, 22(16), pgs. 8373-8390).
The limitations of claim 1 and 20 have been taught by Narain et al. and Lotfollahi et al. above.
With regards to claim 25, although Narain et al. teaches the use of external stimulus component in the cellular models, it does not teach the generation of fingerprints from their chemical structures. However, Gao et al. teaches how 2D fingerprints of compounds generated through common implementations such as Daylight and ECFP4, could hold predictive power on important drug related properties of toxicity and binding affinity (Pg 8374 right col para 3 – Pg 8375 left col para 1, pg. 8382 right col para 3, generating the fingerprint from a chemical structure of the compound using Daylight, BCI, ECFP4, EcFC, MDL, TTFP, UNITY 2D, RNNS2S, GraphCony, or fingerprint SMILES Transformer). In addition, Gao et al. disclose the use of these various fingerprints in multitask neural networks for toxicity predictions, which resemble the network models in Narain et al., demonstrating successful integration of the fingerprints into a network model (Pg 8376 left col para 1, Pg8377 left col para 2).
It would have been prima facie obvious to one of ordinary skill in the art at the effective filing date of the invention to implement the chemical fingerprint of Gao et al. into the cellular model of Narain et al. to create more representative interpretation of how the perturbation agents could influence cellular responses. As stated in Gao et al., one of ordinary skill in the art would have been motivated to incorporate these easy and fast to generate compound fingerprints in order to gain predictive insights into factors that might affect cellular responses such as toxicity (Pg 8382 right col para 3). Furthermore, one of skill in the art before the effective filing date of the claimed invention would have a reasonable expectation of success at incorporating the chemical fingerprints into the network model of Narain et al., as Gao et al. has already successfully demonstrated the integration of chemical fingerprints into network models for analysis.
Claim 31 is rejected under 35 U.S.C. 103 as being unpatentable over Narain et al. and Lotfollahi et al. as applied to claim 1 above, and further in view of Tang et al. (Journal of Molecular Neuroscience Vol 50 Pg 324-332 Published March 15th 2013).
The limitations of claim 1 has been taught by Narain et al. and Lotfollahi et al. above.
Regarding claim 31, Narain et al. teaches the how the trial network can be changed by the deletion of a network fragment and then scored, and if the score is improved the change is kept, if not the change is rejected, which allows optimization of the network over iterations (Pg. 60 para [0285], pruning the activation data structure by removing from the activation data structure one or more cellular constituent modules that fail to associate with a known cellular pathway, a biological process, a transcription factor, a cell receptor, or a kinase that is implicated in the cellular process of interest). In addition, Narain et al. teaches the step of identifying a gene associated with a unique causal relationship as a modulator of a biological system (Pg. 43 para [0013], for each respective cellular constituent module in the plurality of cellular constituent modules, using the identity of each cellular constituent in the respective cellular constituent module to associate the respective cellular constituent module with a known cellular pathway, a biological process, a transcription factor, a cell receptor, or a kinase).
However, Narain et al. does not explicitly teach the use of a gene set enrichment analysis algorithm or a contextualization algorithm to associate the genes with biological systems and processes. However, the use of gene set enrichment analysis and contextualization algorithms to associate a gene with a cellular process were known in the art at the time of the effective filing date of the invention as taught by Tang et al.
Tang et al. teaches the use of gene set enrichment analysis (GSEA) to identify significant genes sets and pathways present in glioma (Pg 325 right col para 2). Analyzing the gene datasets, Tang et al. were able to identify the genes involved in six significant pathways, that were upregulated and concerned with cellular processes, cell growth and death, environmental information processing, and signal transduction (Pg 326 right col para 2, using the identity of each cellular constituent in the respective cellular constituent module to associate the respective cellular constituent module with a known cellular pathway, a biological process, a transcription factor, a cell receptor, or a kinase using a contextualization algorithm wherein the contextualization algorithm is a gene set enrichment analysis algorithm).
It would have been prima facie obvious to one of ordinary skill in the art at the effective filing date of the invention to incorporate the gene set enrichment analysis of Tang et al. into the cellular model of Narain et al. to create more representative interpretations of how the perturbation agents could influence cellular responses. As stated in Tang et al., one of ordinary skill in the art would have been motivated to utilize GSEA to determine key genes and pathways associated with the development of disease, helping to uncover collective behavior of genes in states of health and/or disease (Pg 325 left col para 2). Furthermore, one of skill in the art before the effective filing date of the claimed invention would have a reasonable expectation of success at incorporating GSEA into the network model of Narain et al. as GSEA utilization and analysis was a well-established field and had shown success with varying types of datasets as shown by Tang et al.
Conclusion
The prior art made of record and not relied upon is considered pertinent to applicant's disclosure.
Patel-Murray, N.L., Adam, M., Huynh, N. et al. A Multi-Omics Interpretable Machine Learning Model Reveals Modes of Action of Small Molecules. Sci Rep 10, 954 (2020).
A. P. Athreya et al., "Machine Learning Helps Identify New Drug Mechanisms in Triple-Negative Breast Cancer," in IEEE Transactions on NanoBioscience, vol. 17, no. 3, pp. 251-259, July 2018,
Lotfollahi M et al., Theis FJ. Predicting cellular responses to complex perturbations in high-throughput screens. Mol Syst Biol. 2023 Jun 12;19(6): e11517
Any inquiry concerning this communication or earlier communications from the examiner should be directed to WENYU YANG whose telephone number is (571)272-0035. The examiner can normally be reached 8:00am - 5:00 pm.
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, Olivia Wise can be reached at (571) 272-2249. 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.
/W.Y./Examiner, Art Unit 1685
/OLIVIA M. WISE/Supervisory Patent Examiner, Art Unit 1685
Prosecution Insights