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 .
Detailed Action
In amendments dated 12/15/25, Applicant amended claims 1-6, 9-16and 19-20, canceled no claims, and added no new claims. Claims 1-20 are presented for examination.
Objections
Claims 1, 11, and 20 are objected to because of the following informalities:
the second limitation recites “the individuals selected from the reference panel” but no individuals have been recited as selected from a reference panel, creating an antecedent basis problem;
the third limitation recites “a plurality of simulated genomic of simulated descendants” which is unclear language. Per specification paragraph 0003 (“simulated inheritance datasets”) Examiner construes this as reciting “a plurality of genomic datasets of simulated descendants;”
the fourth limitation recites “the plurality of simulated family trees that correspond to different populations” but each simulated family tree is recited as corresponding to a population and each population may be the same, in which case “the plurality of simulated family trees that correspond to different populations” lacks antecedent basis;
the fourth limitation recites “a plurality of simulated genomic representing a plurality of simulated descendants” which is unclear language. Per specification paragraph 0003 (“simulated inheritance datasets”) Examiner construes this language as “a plurality of simulated genomic datasets representing a plurality of simulated descendants;” and
the seventh limitation recites “applying the plurality of simulated genomic as training samples” which is grammatically unclear. Per specification paragraph 0003 (“simulated inheritance datasets”) Examiner construes this language as “applying the plurality of simulated genomic datasets as training samples.”
Claims 2 and 12 are objected to because of the following informality: the second limitation recites “a number of the individuals in a particular data-inheritance origin whose genomic match the inheritance dataset of the candidate” which is grammatically unclear. Per specification paragraph 0086 (“whose inheritance datasets …”) Examiner construes this language as “a number of the individuals in a particular data-inheritance origin whose genomic datasets match the inheritance dataset of the candidate.”
Rejections under 35 U.S.C. 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 1-20 are rejected under 35 U.S.C. 101 because the claimed invention is directed to mental processes without significantly more. Independent claims 1, 11, and 20 each recites constructing a plurality of simulated family trees, wherein each simulated family tree corresponds to a population that has a mixed origin and is built using the individuals selected from the reference panel based on an origin composition corresponding to the population, wherein building a simulated family tree comprises: generating simulated descendants of the individuals in the simulated family tree by simulating meiosis and recombination events based on the genetic markers of the genomic datasets of the reference panels to generate a plurality of simulated genomic of simulated descendants; generating, from the plurality of simulated family trees that correspond to different populations, a plurality of simulated genomic representing a plurality of simulated descendants, each descendant representing a simulated descendant in one of the simulated family trees; and training a machine learning model that is configured to determine inheritance labels, wherein training the machine learning model comprises: applying the plurality of simulated genomic as training samples; applying the machine learning model to predict inheritance labels of the training samples; comparing predicted inheritance labels to actual labels obtained from the plurality of simulated family trees; and adjusting the origin-specific weight parameters based on label comparisons. Constructing a plurality of simulated family trees, generating simulated descendants, and generating a plurality of simulated genomic datasets are recited broadly and are mental processes accomplishable in the human mind or on paper. Training a machine learning model is merely applying it and is not significantly more than the recited mental processes per Recentive Analytics v. Fox Broadcasting Corp. (134 F.4th 1205, 2025 U.S.P.Q.2d 628), and applying the genomic datasets as training data, comparing labels of data and adjusting weight parameters are also mental processes accomplishable in the human mind or on paper. Each claim recites additional elements of receiving a plurality of genomic datasets of reference panels of individuals, each reference panel corresponding to an origin and comprising a genomic dataset that includes genetic markers representative of the origin, which is a data gathering step and insignificant extra-solution activity; and initiating origin-specific weight parameters in the machine learning model, which is an input step and also insignificant extra-solution activity. Claim 11 recites one or more processors and memory configured to store instructions and claim 20 recites a non-transitory computer readable medium, which are generic components of a computer system. Examiner found several paragraphs in the specification that mentioned improving the training of or augmenting a machine learning model such as paragraphs 0099 and 0106-0109, in particular with the use of simulation and the use and adjustment of origin-specific weight parameters. Thus the claim steps do not recite a particular improvement in any technology or function of a computer per MPEP 2106.04(d) and do not recite any unconventional steps in the invention per MPEP 2106.05(a). Therefore, the recited mental processes are not integrated into a practical application. Taking the claims as a whole, receiving a plurality of genomic datasets is recited broadly and amounts to receiving data across a network per specification paragraphs 0249 and 0256 and figure 8 network 820, which is routine and conventional activity per the list of such activities in MPEP 2106.05(d) part II. Initiating parameters in a machine learning model is routine and conventional per Sprenkle (US 20250112126 paragraph 0060, initiating parameters in machine learning model and later adjusting said parameters based on the difference (comparison) between predicted outputs and ground-truth labels) and Guar et al (US 20240281887 paragraph 0080, adjusting weights in a machine learning based on comparison of labels associated with input data). The one or more processors, memory configured to store instructions, and non-transitory computer readable medium, are still generic components of a computer system. Therefore, the claims do not include additional elements that are sufficient to amount to significantly more than the recited mental processes.
Claims 2 and 12 each recites filtering a plurality of candidates based on inheritance labels (mental process accomplishable in the human mind or on paper); identifying, for each candidate, a number of the individuals in a particular data-inheritance origin whose genomic match the inheritance dataset of the candidate (identifying is evaluating and a mental process); and selecting a candidate to be added to the genomic datasets based on the number of matched individuals that correspond to the candidate compared to numbers of matched individuals of other candidates (selecting is evaluating and a mental process). Claims 3 and 13 each recites generating a candidate pool that include a plurality of candidates based on inheritance labels related to a particular origin (recited broadly and a mental process accomplishable in the human mind or on paper); determining that one of the candidates has a first number of matches associated with a particular origin and a second number of matches associated with a second origin, the second number of matches exceeding a threshold (determining is evaluating and a mental process); and removing said one of the candidates from the candidate pool (mental process accomplishable in the human mind or on paper).
Claims 4 and 14 each recites accessing a population composition of the geographical location, the population composition comprising information related to percentage of individuals with a plurality of origins (accessing a population composition is retrieving data and routine and conventional activity per the list of such activities in MPEP 2016.05(d) part II); sampling, based on the population composition of the geographical location, reference- panel datasets from the plurality of reference panels for the plurality of origins (sampling datasets is retrieving data and routine and conventional activity per the list of such activities in MPEP 2016.05(d) part II); and representing sampled reference-panel datasets in nodes of the particular simulated family tree (representing datasets in nodes is storing the datasets which is routine and conventional activity per the list of such activities in MPEP 2016.05(d) part II). Claims 5 and 15 each recites selecting placements of the sampled reference-panel datasets based on the generation- specific composition (selecting placements is evaluating and a mental process). Claims 6 and 16 each recites treating the particular simulated individual as a descendant individual of the genomic datasets that are placed in the particular simulated family tree (treating an entity as data is a mental process accomplishable in the human mind or on paper); simulating a plurality of inheritance events (simulating an event is a mental process accomplishable in the human mind or on paper); and generating the particular simulated inheritance dataset of the particular simulated individual based on the plurality of inheritance events (generating a dataset is recited broadly and is a mental process accomplishable in the human mind or on paper).
Claims 7 and 17 each recites wherein the machine learning model is a hidden Markov model with windows that represent segments of inheritance data, each window comprising a plurality of nodes and each node representing an origin, wherein the origin-specific weight parameters are associated with weights of the plurality of nodes (training a machine learning model is merely applying it and is not significantly more than the recited mental processes). Claims 8 and 18 each recites comparing the predicted inheritance labels to the actual labels to identifying under- represented origins and over-represented origins (comparing labels is evaluating and a mental process); for an under-represented origin, increasing a value of the origin-specific weight parameter corresponding to the under-represented origin (increasing a value is a mental process accomplishable in the human mind or on paper); and for an over-represented origin, decreasing a value of the origin-specific weight parameter corresponding to the over-represented origin (decreasing a value is a mental process accomplishable in the human mind or on paper).
Claims 9 and 19 each recites dividing a particular simulated inheritance dataset in a particular training sample into a plurality of windows (dividing data is a mental process accomplishable in the human mind or on paper); examining how a segment of the particular simulated inheritance dataset in a particular window is inherited in a particular simulated family tree (examining is recited broadly and is a mental process accomplishable in the human mind or on paper); identifying a reference-panel individual in the particular family tree who passes down the segment to the simulated inheritance dataset (identifying is evaluating and a mental process); determining an origin label of said reference-panel individual (determining is a mental process accomplishable in the human mind or on paper); and using the origin label as the actual label (using data is a mental process accomplishable in the human mind or on paper). Claim 10 recites wherein the training samples comprises admixed individuals that are simulated from plurality of simulated family trees and non-admixed individuals that are sampled from actual user datasets (data is a mental process accomplishable in the human mind or on paper).
Rejections under 35 U.S.C. 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.
Claims 1-20 are rejected under 35 U.S.C. 103 as being unpatentable over Montserrat et al (US 20230197204), hereafter Montserrat, in view of Curtis et al (US 20210257060), hereafter Curtis, in further view of McMaster-Schraiber et al (US 20210134387), hereafter McMaster-Schraiber.
With respect to claims 1, 11, and 20, Montserrat teaches:
receiving a plurality of genomic datasets of reference panels of individuals, each reference panel corresponding to an origin and comprising a genomic dataset that includes genetic markers representative of the origin (paragraphs 0011, 0043, 0104 the full genome data of a plurality of individuals organized - according to known ancestral origins, i.e. a plurality of data-inheritance origins, represent the plurality of reference panels);
constructing a plurality of simulated family trees, wherein each simulated family tree corresponds to a population that has a mixed origin and is built using the individuals selected from the reference panel based on an origin composition corresponding to the population ((paragraphs 0043, 0105; A plurality of simulated admixed descendants of the data of the plurality of individuals organized according to their known ancestral origins, i.e. of the reference panels, is generated based on Wright-Fisher forward - simulation over a series of generations),
generating, from the plurality of simulated family trees that correspond to different populations, a plurality of simulated genomic representing a plurality of simulated descendants, each descendant representing a simulated descendant in one of the simulated family trees;
(paragraphs 0043, 0105 the simulated admixed descendants correspond to the - descendant named entities whose genomic sequences represent the plurality of simulated inheritance datasets), and
training a machine learning model that is configured to determine inheritance labels (paragraphs 0043, 0044, 0104, 0105 figure 2B, the inheritance labels being the ancestral origin categories), wherein training the machine learning model comprises:
initiating origin-specific weight parameters in the machine learning model (paragraphs 0042, 0099 weights assigned to ancestral origin estimates);
applying the plurality of simulated inheritance datasets as training samples (paragraphs 0043, 0105, 0107 training data derived from simulated genomic sequences of admixed descendants from full genome data of ancestral origins);
applying the machine learning model to predict inheritance labels of the training samples (paragraphs 0043, 0107 the predicted ancestral origin categories);
comparing predicted inheritance labels to actual labels obtained from the plurality of simulated inheritance trees (paragraphs 0044, 0045, 0107 the predicted and true ancestral origin categories are compared); and
adjusting the origin-specific weight parameters based on label comparisons (paragraphs 0044, 0107 adjusting weights after comparison of predicted and true origin categories).
Montserrat does not teach constructing a plurality of simulated inheritance trees [emphasis added], wherein each simulated inheritance tree corresponds to a population and is built using the reference-panel datasets that are selected from the reference panel corresponding to the population. Curtis teaches this in building a tree structure of communities from genetic datasets (paragraph 0004 communities from genetic datasets) (paragraph 0084 figure 7A). It would have been obvious to have combined use of a data tree structure in Curtis with the simulated inheritance techniques in Montserrat to use the tree structure to define genetic communities and sub-communities in the dataset.
The combination of Montserrat and Curtis does not teach wherein building a simulated family tree comprises: generating simulated descendants of the individuals in the simulated family tree by simulating meiosis and recombination events based on the genetic markers of the genomic datasets of the reference panels to generate a plurality of simulated genomic of simulated descendants; [emphasis added]. McMaster-Schraiber teaches using meiosis and recombination with IBD segments identified from individuals inherited from a common ancestor, where meiosis and recombination actions are used to measure these segments for connecting communities (paragraphs 0089, 0092). It would have been obvious to have used meiosis and recombination actions as described in McMaster-Schraiber with the combination of a data tree structure described in Curtis and the simulated inheritance techniques in Montserrat to trace traits and other genetic information passed down along family lines in a community or population.
With respect to claims 2 and 12, all the limitations in claims 1 and 11 are addressed by the combination of Montserrat, Curtis, and McMaster-Schraiber above. Curtis also teaches:
filtering a plurality of candidates based on inheritance labels (paragraph 0077 filter training set based on label (birth year));
identifying, for each candidate, a number of the individuals in a particular data-inheritance origin whose genomic match the inheritance dataset of the candidate (paragraph 0077 identify individual A or B and a common ancestor); and
selecting a candidate to be added to the genomic datasets based on the number of matched individuals that correspond to the candidate compared to numbers of matched individuals of other candidates (paragraph 0077 select those individuals for training datasets).
With respect to claims 3 and 13, all the limitations in claims 1 and 11 are addressed by the combination of Montserrat, Curtis, and McMaster-Schraiber above. Curtis also teaches:
generating a candidate pool that include a plurality of candidates based on inheritance labels related to a particular origin (paragraph 0074 determine candidate community (ethnic origin from paragraph 0043) using metrics);
determining that one of the candidates has a first number of matches associated with a particular origin and a second number of matches associated with a second origin, the second number of matches exceeding a threshold (paragraph 0072 individuals with genetic segments above a threshold); and
removing said one of the candidates from the candidate pool (paragraph 0072 genetic segments above the threshold filtered from the dataset).
With respect to clams 4 and 14, all the limitations in claims 1 and 11 are addressed by the combination of Montserrat, Curtis, and McMaster-Schraiber above. Curtis also teaches:
accessing a population composition of the geographical location, the population composition comprising information related to percentage of individuals with a plurality of origins (paragraph 0083 filtering/accessing nodes per percent of target ethnic origin (geographic location));
sampling, based on the population composition of the geographical location, reference- panel datasets from the plurality of reference panels for the plurality of origins (paragraph 0083 sampling said nodes); and
representing sampled reference-panel datasets in nodes of the particular simulated family tree (paragraph 0083 representing nodes of said target ethnic origin).
With respect to claims 5 and 15, all the limitations in claims 1, 4, 11, and 14 are addressed by the combination of Montserrat, Curtis, and McMaster-Schraiber above. Curtis also teaches:
selecting placements of the sampled reference-panel datasets based on the generation- specific composition (paragraph 0076 generation is specific to individuals with IBD segments).
With respect to claims 6 and 16, all the limitations in claims 1 and 11 are addressed by the combination of Montserrat, Curtis, and McMaster-Schraiber above. Curtis also teaches:
treating the particular simulated individual as a descendant individual of the genomic datasets that are placed in the particular simulated family tree (paragraph 0083 treating individual as descendant of a dataset for a target ethnicity);
simulating a plurality of inheritance events (paragraph 0083 plurality of inheritance events with individuals); and
generating the particular simulated inheritance dataset of the particular simulated individual based on the plurality of inheritance events (paragraph 0083 generating dataset of target ethnicity).
With respect to claims 7 and 17, all the limitations in claims 1 and 11 are addressed by the combination of Montserrat, Curtis, and McMaster-Schraiber above. McMaster-Schraiber also teaches wherein the machine learning model is a hidden Markov model with windows that represent segments of inheritance data, each window comprising a plurality of nodes and each node representing an origin, wherein the origin-specific weight parameters are associated with weights of the plurality of nodes (paragraph 0119 Hidden Markov Model (HMM) with windows for ordered air of haploid states for a genome, paragraph 0180 weights assigned to each window, paragraph 0127 HMM has nodes, edges with weights (probabilities)).
With respect top claims 8 and 18, all the limitations in claims 1 and 11 are addressed by the combination of Montserrat, Curtis, and McMaster-Schraiber above. Montserrat also teaches:
comparing the predicted inheritance labels to the actual labels to identifying under-represented origins and over-represented origins (paragraph 0044 comparing predicted and true origin data, paragraph 0039 assign smaller/larger (increasing/decreasing) weights for prediction error of SNP origins versus actual origins);
for an under-represented origin, increasing a value of the origin-specific weight parameter corresponding to the under-represented origin (paragraph 0039 assign smaller/larger (increasing/decreasing) weights for prediction error of SNP origins versus actual origins); and
for an over-represented origin, decreasing a value of the origin-specific weight parameter corresponding to the over-represented origin (paragraph 0039 assign smaller/larger (increasing/decreasing) weights for prediction error of SNP origins versus actual origins).
With respect to claims 9 and 19, all the limitations in claims 1 and 11 are addressed by the combination of Montserrat, Curtis, and McMaster-Schraiber above. Montserrat also teaches:
dividing a particular simulated inheritance dataset in a particular training sample into a plurality of windows (paragraph 0042 dividing target origins into windows/subsets);
examining how a segment of the particular simulated inheritance dataset in a particular window is inherited in a particular simulated family tree (paragraph 0042 perform origin estimates (examining), paragraph 0043 examine segments of genomic sequences with known origins);
identifying a reference-panel individual in the particular family tree who passes down the segment to the simulated inheritance dataset (paragraph 0043 identify individuals to be classified in origin dataset (inheritance));
determining an origin label of said reference-panel individual (paragraph 0043 classifying origins into locations (label) for an individual); and
using the origin label as the actual label (paragraph 0043 using the origin data to validate and train a machine learning model).
With respect to claim 10, all the limitations in claims 1 and 11 are addressed by the combination of Montserrat, Curtis, and McMaster-Schraiber above. Montserrat also teaches wherein the training samples comprises admixed individuals that are simulated from plurality of simulated family trees and non-admixed individuals that are sampled from actual user datasets (paragraph 0043 training dataset comprising simulated admixed descendants and known origins (non-admixed)).
Responses to Applicant’s Remarks
Regarding rejections of claims 1-20 under 35 U.S.C. 101 for reciting mental processes without significantly more, Applicant’s arguments have been considered but are not persuasive. On page 14 of his Remarks Applicant asserts there are no mental processes recited in the claims. Examiner disagrees as shown in the rejections maintained above. MPEP 2106.04(2)(2)(III) states "the courts do not distinguish between mental processes that are performed entirely in the human mind and mental processes that require a human to use a physical aid (e.g., pen and paper or a slide rule) to perform the claim limitation" and "nor do the courts distinguish between claims that recite mental processes performed by humans and claims that recite mental processes performed on a computer." Examiner notes each limitation uses a generic computer as a tool to perform and each one (constructing a plurality of simulated family trees, generating simulated descendants, generating simulated genomic datasets, comparing inheritance labels, adjusting weight parameters) are recited broadly and, according to a BRI of each limitation, can be done with a physical aid such as pen and paper. Training a machine learning model and applying a machine learning model are established methods of machine learning and are not significantly more then mental processes per Recentive Analytics v. Fox Broadcasting Corp. (134 F.4th 1205, 2025 U.S.P.Q.2d 628). Furthermore, Applicant discusses the invention involving several techniques like probabilistic modeling, linkage disequilibrium, algorithmic manipulation, and numerical optimization which are not claimed. On page 15 of his Remarks Applicant asserts “the claim also recites an improvement in computer functionality, specifically an
enhancement to machine learning models used for determining inheritance labels in challenging
contexts such as admixed populations or populations tied to specific geographical locations with
mixed origin compositions.” Examiner notes the claims lacks inventive details that might show such an improvement to a computer’s functionality. The claims lack inventive details in simulated family trees and reproduction events such as those found in paragraph 0099 that also might show an improvement to a technical field.
Regarding rejections of claims 1-20 under 35 U.S.C. 103 by Montserrat and Curtis in further view of McMaster-Schraiber, Applicant’s arguments have been considered but are not persuasive. On page 16 of his Remarks Applicant asserts the references “do not teach or suggest generating a plurality of simulated family trees constructed from reference-panel individuals according to a population-specific, generation-specific composition, and then simulating reproduction events based on those population compositions to generate realistic, multi-generational simulated genomic datasets.“ Examiner disagrees as Montserrat teaches simulated family trees of admixed descendants according t their ancestral origins in paragraphs 0043 and 0105 as cited above. Applicant further asserts “None of the prior art suggests using realistic generational admixture simulation to create ground-truth data for improving origin-label prediction and expanding region coverage where reference samples are inadequate. The iterative tuning of origin-specific weight parameters using simulated datasets provides a technical improvement to the model's architecture and inference capability, enabling accurate ethnic composition estimation for complex populations such as Mexico or Victoria, Australia,
capabilities that are not achievable by merely applying the generic machine learning or genetic matching techniques taught in the prior art.” Examiner notes the clams do not recite ground-truth data, vectorized tuning of pre-population weights, iterative anything, expanding region coverage, or accurate ethnic composition estimation. Examiner believes the references teach the claims as shown.
Conclusion
THIS ACTION IS MADE FINAL. Applicant is reminded of the extension of time policy as set forth in 37 CFR 1.136(a).
A shortened statutory period for reply to this final action is set to expire THREE MONTHS from the mailing date of this action. In the event a first reply is filed within TWO MONTHS of the mailing date of this final action and the advisory action is not mailed until after the end of the THREE-MONTH shortened statutory period, then the shortened statutory period will expire on the date the advisory action is mailed, and any nonprovisional extension fee (37 CFR 1.17(a)) pursuant to 37 CFR 1.136(a) will be calculated from the mailing date of the advisory action. In no event, however, will the statutory period for reply expire later than SIX MONTHS from the mailing date of this final action.
Inquiry
Any inquiry concerning this communication or earlier communications from the examiner should be directed to BRUCE M MOSER whose telephone number is (571)270-1718. The examiner can normally be reached M-F 9a-5p.
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, Boris Gorney can be reached at 571 270-5626. 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.
/BRUCE M MOSER/Primary Examiner, Art Unit 2154 3/12/26