DETAILED ACTION
Notice of Pre-AIA or AIA Status
1. The present application, filed on or after March 16, 2013, is being examined under the first inventor to file provisions of the AIA .
Notice to Applicant
2. This communication is in response to the communication filed 11/12/2024. Claims 5-8, 11, 22-25, 28, 39-42, and 45 are cancelled. Claims 1-4, 9-10, 12-21, 26-27, 29-38, 43-44, and 46-51 are currently pending.
Claim Rejections - 35 USC § 101
3. 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.
3.1. Claims 1-4, 9-10, 12-21, 26-27, 29-38, 43-44, and 46-51 are rejected under 35 U.S.C. § 101 because while the claims (1) are to a statutory category (i.e., process, machine, manufacture or composition of matter, the claims (2A1) recite an abstract idea (i.e., a law of nature, a natural phenomenon); (2A2) do not recite additional elements that integrate the abstract idea into a practical application; and (2B) are not directed to significantly more than the abstract idea itself.
In regards to (1), the claims are to a statutory category (i.e., statutory categories including a process, machine, manufacture or composition of matter). In particular, independent claims 1, 18 and 35, and their respective dependent claims are directed, in part, to a method, system and medium for modeling population frequency for variant pathogenicity estimation.
In regards to (2A1), the claims, as a whole, recite and are directed to an abstract idea because the claims include one or more limitations that correspond to an abstract idea including mathematical concepts, mental processes and/or certain methods of organizing human activity which encompasses both certain activity of a single person, certain activity that involves multiple people, and certain activity between a person and a computer. For example, independent claims 1, 18 and 35, as a whole, are directed to configuring a machine learning model to model population frequency for variant classification for use in formulating a diagnosis for a patient by, inter alia, applying a logistic regression model to a first set of population data for a first set of genes; for each set of population data evaluating a variant classification prediction; outputting variant pathogenicity estimates, etc. which are human activities and/or interactions and therefore, certain methods of organizing human activity which encompasses both certain activity of a single person, certain activity that involves multiple people, and certain activity between a person and a computer. The dependent claims include all of the limitations of their respective independent claims and thus are directed to the same abstract idea identified for the independent claims but further describe the elements and/or recite field of use limitations.
The claims, as a whole, are also directed to mathematical concepts. For example, a logistic regression model predicts, via calculations, probabilities. Here, the logistic regression model is applied to population data to quantify predictive value of allele frequency in a gene, estimating pathogenicity, and the like, which are mathematical concepts.
Furthermore, assuming arguendo, the claims are not directed to certain methods of organizing human activities, the claims, nevertheless, are directed to an abstract idea because the claims, except for certain limitations (* identified below in bold), under the broadest reasonable interpretation, can be reasonably and practically performed in the human mind and/or with pen and paper using observation, evaluation, judgment and/or opinion. That is, other than reciting the certain additional elements, nothing in the claims precludes the limitations from being practically performed in the mind and/or with pen and paper.
CLAIM 1
A method for configuring a machine learning model to model population frequency for variant classification, the method comprising:
applying a logistic regression model to a first set of population data for a first set of genes, wherein an item of the first set of population data comprises, for a variant located at a position within a gene of the first set of genes, a set of features comprising at least one gene-level feature, at least one variant-level feature, and at least one population frequency meta-feature, and a reference label that indicates whether the variant is benign or pathogenic, wherein the at least one population frequency meta-feature quantifies predictive value of allele frequency in the gene;
for each item of the first set of population data, evaluating a variant classification prediction output by the logistic regression model based on an expected variant classification indicated by the reference label; and
iteratively adjusting a value of at least one parameter or coefficient of the logistic regression model until output of a loss function computed based on the variant classification prediction output by the logistic regression model satisfies at least one first performance criterion, to produce a trained logistic regression model, wherein the trained logistic regression model is capable of outputting variant pathogenicity estimates that satisfy at least one second performance criterion.
CLAIM 2
The method of claim 1, further comprising:
using the trained logistic regression model to generate a prediction as to whether the variant is benign or pathogenic.
CLAIM 3
The method of claim 2, further comprising:
providing the prediction as to whether the variant is benign or pathogenic to a clinician for use in formulating, by the clinician, a diagnosis of a patient.
CLAIM 4
The method of claim 1, further comprising:
applying the trained logistic regression model to a second set of population data for a plurality of variants of a second plurality of genes; and
for each variant of the second plurality of genes, receiving a variant classification prediction output by the trained logistic regression model and, in response to the variant classification prediction satisfying at least the second performance criterion, storing the variant classification prediction in association with the variant for retrieval via at least one query.
CLAIM 9
The method of claim 1, further comprising:
selecting, as the set of features, not more than thirty features; and
applying the logistic regression model to the selected set of not more than thirty features,
wherein the trained logistic regression model is capable of outputting variant pathogenicity estimates that satisfy the at least one second performance criterion based on the selected set of not more than thirty features.
CLAIM 10
The method of claim 1, further comprising:
computing a fixation index, wherein the fixation index comprises sub-population frequency data;
including the fixation index in the set of features; and
applying the logistic regression model to the set of features including the fixation index, wherein the trained logistic regression model is capable of outputting variant pathogenicity estimates that satisfy the at least one second performance criterion based on the set of features including the fixation index.
CLAIM 12
The method of claim 1, wherein iteratively adjusting the value of at least one parameter of the logistic regression model comprises adjusting a C value of the logistic regression model until the output of the loss function satisfies the at least one first performance criterion, to produce the trained logistic regression model.
CLAIM 13
The method of claim 1, further comprising:
configuring the logistic regression model to model population frequency for variant classification using L1 regularization.
CLAIM 14
The method of claim 1, further comprising:
estimating the first performance criterion using a means square error or area under the receiver operating characteristic curve.
CLAIM 15
The method of claim 1, further comprising:
determining the second performance criterion using at least one of decision boundary inspection, feature weight inspection, or a comparison to a benchmark variant classification.
CLAIM 16
The method of claim 1, further comprising:
using variant classification estimates output by the trained logistic regression model as an input to a variant classification framework.
CLAIM 17
The method of claim 1, wherein the at least one population frequency meta-feature comprises expected frequency distributions of known benign variants and known pathogenic variants within the gene.
CLAIM 18
A system, comprising:
at least one processor; and
at least one memory coupled to the at least one processor, wherein the at least one memory comprises at least one instruction that when executed by the at least one processor causes the at least one processor to perform at least one operation comprising:
applying a logistic regression model to a first set of population data for a first set of genes, wherein an item of the first set of population data comprises, for a variant located at a position within a gene of the first set of genes, a set of features comprising at least one gene-level feature, at least one variant-level feature, and at least one population frequency meta-feature, and a reference label that indicates whether the variant is benign or pathogenic, wherein the at least one population frequency meta-feature quantifies predictive value of allele frequency in the gene;
for each item of the first set of population data, evaluating a variant classification prediction output by the logistic regression model based on an expected variant classification indicated by the reference label; and
adjusting a value of at least one parameter or coefficient of the logistic regression model until at least one first performance criterion is satisfied to produce a trained logistic regression model, wherein the trained logistic regression model is capable of outputting variant pathogenicity estimates that satisfy at least one second performance criterion.
CLAIM 19
The system of claim 18, wherein the at least one memory further comprises at least one instruction that when executed by the at least one processor causes the at least one processor to perform at least one operation comprising:
using the trained logistic regression model to generate a prediction as to whether the variant is benign or pathogenic.
CLAIM 20
The system of claim 19, wherein the at least one memory further comprises at least one instruction that when executed by the at least one processor causes the at least one processor to perform at least one operation comprising:
providing the prediction as to whether the variant is benign or pathogenic to a clinician for use in
formulating, by the clinician, a diagnosis of a patient.
CLAIM 21
The system of claim 18, wherein the at least one memory further comprises at least one instruction that when executed by the at least one processor causes the at least one processor to perform at least one operation comprising:
applying the trained logistic regression model to a second set of population data for a plurality of variants of a second plurality of genes; and
for each variant of the second plurality of genes, receiving a variant classification prediction output by the trained logistic regression model and, in response to the variant classification prediction satisfying at least the second performance criterion, storing the variant classification prediction in association with the variant for retrieval via at least one query.
CLAIM 26
The system of claim 18, wherein the at least one memory further comprises at least one instruction that when executed by the at least one processor causes the at least one processor to perform at least one operation comprising:
selecting, as the set of features, not more than thirty features; and
applying the logistic regression model to the selected set of not more than thirty features, wherein the trained logistic regression model is capable of outputting variant pathogenicity estimates that satisfy the at least one second performance criterion based on the selected set of not more than thirty features.
CLAIM 27
The system of claim 18, wherein the at least one memory further comprises at least one instruction that when executed by the at least one processor causes the at least one processor to perform at least one operation comprising:
computing a fixation index, wherein the fixation index comprises sub-population frequency data;
including the fixation index in the set of features; and
applying the logistic regression model to the set of features including the fixation index, wherein the trained logistic regression model is capable of outputting variant pathogenicity estimates that satisfy the at least one second performance criterion based on the set of features including the fixation index.
CLAIM 29
The system of claim 18, wherein iteratively adjusting the value of at least one parameter of the logistic regression model comprises adjusting a C value of the logistic regression model until the output of the loss function satisfices the at least one first performance criterion, to produce the trained logistic regression model.
CLAIM 30
The system of claim 18, wherein the at least one memory further comprises at least one instruction that when executed by the at least one processor causes the at least one processor to perform at least one operation comprising:
configuring the logistic regression model to model population frequency for variant classification using L1 regularization.
CLAIM 31
The system of claim 18, wherein the at least one memory further comprises at least one instruction that when executed by the at least one processor causes the at least one processor to perform at least one operation comprising:
estimating the first performance criterion using a means square error or area under the receiver operating characteristic curve.
CLAIM 32
The system of claim 18, wherein the at least one memory further comprises at least one instruction that when executed by the at least one processor causes the at least one processor to perform at least one operation comprising:
determining the second performance criterion using at least one of decision boundary inspection, feature weight inspection, or a comparison to a benchmark variant classification.
CLAIM 33
The system of claim 18, wherein the at least one memory further comprises at least one instruction that when executed by the at least one processor causes the at least one processor to perform at least one operation comprising:
using variant classification estimates output by the trained logistic regression model as an input to a variant classification framework.
CLAIM 34
The system of claim 18, wherein the at least one population frequency meta-feature comprises expected frequency distributions of known benign variants and known pathogenic variants within the gene.
CLAIM 35
At least one non-transitory machine-readable medium comprising at least one instruction that when executed by at least one processor causes the at least one processor to perform operations comprising:
applying a logistic regression model to a first set of population data for a first set of genes, wherein an item of the first set of population data comprises, for a variant located at a position within a gene of the first set of genes, a set of features comprising at least one population frequency meta-feature and a reference label that indicates whether the variant is benign or pathogenic, wherein the at least one population frequency meta-feature quantifies predictive value of allele frequency in the gene;
for each item of the first set of population data, evaluating a variant classification prediction output by the logistic regression model based on an expected variant classification indicated by the reference label; and
adjusting a value of at least one parameter or coefficient of the logistic regression model until at least one first performance criterion is satisfied to produce a trained logistic regression model, wherein the trained logistic regression model is capable of outputting variant pathogenicity estimates that satisfy at least one second performance criterion.
CLAIM 36
The at least one non-transitory machine-readable medium of claim 35, further comprising at least one instruction that when executed by the at least one processor causes the at least one processor to perform at least one operation comprising:
using the trained logistic regression model to generate a prediction as to whether the variant is benign or pathogenic.
CLAIM 37
The at least one non-transitory machine-readable medium of claim 36, further comprising at least one instruction that when executed by the at least one processor causes the at least one processor to perform at least one operation comprising:
providing the prediction as to whether the variant is benign or pathogenic to a clinician for use in formulating, by the clinician, a diagnosis of a patient.
CLAIM 38
The at least one non-transitory machine-readable medium of claim 35, further comprising at least one instruction that when executed by the at least one processor causes the at least one processor to perform at least one operation comprising:
applying the trained logistic regression model to a second set of population data for a plurality of variants of a second plurality of genes; and
for each variant of the second plurality of genes, receiving a variant classification prediction output by the trained logistic regression model and, in response to the variant classification prediction satisfying at least the second performance criterion, storing the variant classification prediction in association with the variant for retrieval via at least one query.
CLAIM 43
The at least one non-transitory machine-readable medium of claim 35, further comprising at least one instruction that when executed by the at least one processor causes the at least one processor to perform at least one operation comprising:
selecting, as the set of features, not more than thirty features; and
applying the logistic regression model to the selected set of not more than thirty features, wherein the trained logistic regression model is capable of outputting variant pathogenicity estimates that satisfy the at least one second performance criterion based on the selected set of not more than thirty features.
CLAIM 44
The at least one non-transitory machine-readable medium of claim 35 further comprising at least one instruction that when executed by the at least one processor causes the at least one processor to perform at least one operation comprising:
computing a fixation index, wherein the fixation index comprises sub-population frequency data;
including the fixation index in the set of features; and
applying the logistic regression model to the set of features including the fixation index, wherein the trained logistic regression model is capable of outputting variant pathogenicity estimates that satisfy the at least one second performance criterion based on the set of features including the fixation index.
CLAIM 46
The at least one non-transitory machine-readable medium of claim 35, wherein adjusting the value of at least one parameter of the logistic regression model comprises adjusting a C value of the logistic regression model until output of a loss function satisfices the at least one first performance criterion, to produce the trained logistic regression model.
CLAIM 47
The at least one non-transitory machine-readable medium of claim 35, further comprising at least one instruction that when executed by the at least one processor causes the at least one processor to perform at least one operation comprising:
configuring the logistic regression model to model population frequency for variant classification using L1 regularization.
CLAIM 48
The at least one non-transitory machine-readable medium of claim 35, further comprising at least one instruction that when executed by the at least one processor causes the at least one processor to perform at least one operation comprising:
estimating the at least one first performance criterion using a means square error or area under the receiver operating characteristic curve.
CLAIM 49
The at least one non-transitory machine-readable medium of claim 35, further comprising at least one instruction that when executed by the at least one processor causes the at least one processor to perform at least one operation comprising:
determining the second performance criterion using at least one of decision boundary inspection, feature weight inspection, or a comparison to a benchmark variant classification.
CLAIM 50
The at least one non-transitory machine-readable medium of claim 35, further comprising at least one instruction that when executed by the at least one processor causes the at least one processor to perform at least one operation comprising:
using variant classification estimates output by the trained logistic regression model as an input to a variant classification framework.
CLAIM 51
The at least one non-transitory machine-readable medium of claim 35, wherein the at least one population frequency meta-feature comprises expected frequency distributions of known benign variants and known pathogenic variants within the gene.
* The limitations that are in bold are considered “additional elements” that are further analyzed below in subsequent steps of the 101 analysis. The limitations that are not in bold are abstract and/or can be reasonably and practically performed in the human mind and/or with pen paper.
In regards to (2A2), the claims do not recite additional elements that integrate the abstract idea into a practical application. The additional elements in the claims (i.e., * identified above in bold) do not integrate the abstract idea into a practical application because the additional elements merely add insignificant extra-solution activity to the abstract idea; merely link the use of the judicial exception to a particular technological environment or field of use; and/or simply append technologies and functions, specified at a high level of generality, to the abstract idea (i.e., the additional elements do not amount to more than a recitation of the words “apply it” (or an equivalent) or are more than mere instructions to implement an abstract idea or other exception on a computer).
Here, the additional elements (e.g., machine learning model, processor, memory, medium, etc.) are recited at a high-level of generality such that it amounts to no more than mere instructions to apply the abstract idea using generic computer technologies. Moreover, the claims recite “by the logistic model”, “causes the at least one processor to perform”, etc. devoid of any meaningful technological improvement details and thus, further evidence the additional elements are merely being used to leverage generic technologies to automate what otherwise could be done manually. Accordingly, the additional elements do not integrate the abstract idea into a practical application because they do not impose any meaningful limits on practicing the abstract idea.
Furthermore, the additional elements do not recite improvements to the functioning of a computer, or to any other technology or technical field—the additional elements merely recite general purpose computer technology; the additional elements do not recite applying or using a judicial exception to effect a particular treatment or prophylaxis for disease or medical condition—there is no actual administration of a particular treatment; the additional elements do not recite applying the judicial exception with, or by use of, a particular machine—the additional elements merely recite general purpose computer technology; the additional elements do not recite limitations effecting a transformation or reduction of a particular article to a different state or thing—the additional elements do not recite transformation such as a rubber mold process; the additional elements do not recite applying or using the judicial exception in some other meaningful way beyond generally linking the use of the judicial exception to a particular technological environment—the additional elements merely leverage general purpose computer technology to link the abstract idea to a technological environment.
In regards to (2B), the claims, individually, as a whole and in combination with one another, do not include additional elements that are sufficient to amount to significantly more than the judicial exception because the additional elements or combination of elements in the claims, other than the abstract idea per se, amount to no more than a recitation of (A) a generic computer structure(s) that serves to perform computer functions that serve to merely link the abstract idea to a particular technological environment (i.e., computers); and/or (B) functions that are well-understood, routine, and conventional activities previously known to the pertinent industry.
Here, as discussed above with respect to integration of the abstract idea into a practical application, the additional elements amount to no more than mere instructions to apply the exception using generic computer technologies. Mere instructions to apply an exception using generic computer technologies cannot provide an inventive concept.
Moreover, paragraphs [0175]-[0177] of applicant's specification (US 2025/0069702) recites that the system/method/medium is implemented using a personal computer (PC), a smart phone, a tablet PC, a set-top box (STB), a Personal Digital Assistant (PDA), a cellular telephone, a web appliance, a server, or any machine capable of executing a set of instructions (sequential or otherwise) that specify actions to be taken by that machine which are well-known general purpose or generic-type computers and/or technologies. The use of generic computer components recited at a high level of generality to process information through an unspecified processor/computer does not impose any meaningful limit on the computer implementation of the abstract idea. Thus, taken alone, the additional elements do not amount to significantly more than the above-identified judicial exception (the abstract idea). Looking at the limitations as an ordered combination adds nothing that is not already present when looking at the elements taken individually. There is no indication that the combination of elements improves the functioning of a computer or improves any other technology. Their collective functions merely provide conventional computer implementation.
Furthermore, the additional elements are merely well-known general purpose computers, components and/or technologies that receive, transmit, store, display, generate and otherwise process information which are akin to functions that courts consider well-understood, routine, and conventional activities previously known to the pertinent industry, such as, performing repetitive calculations; receiving or transmitting data over a network; electronic recordkeeping; retrieving and storing information in memory; and sorting information (See, for example, MPEP § 2106).
Therefore, the claims are not patent-eligible under 35 U.S.C. § 101.
Claim Rejections - 35 USC § 103
4. 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.
4.1. Claims 1-4, 9, 14-21, 26, 31-38, 43, and 48-51 are rejected under 35 U.S.C. 103 as being unpatentable over Gao et al. (US 2020/0279157), in view of Lai et al. (Non-Patent Literature (NPL): “LEAP: Using machine learning to Support variant classification in a clinical setting”).
CLAIM 1
Gao teaches a method for configuring a machine learning model to model population frequency for variant classification, (Gao: abstract), the method comprising:
applying a model to a first set of population data for a first set of genes, wherein an item of the first set of population data comprises, for a variant located at a position within a gene of the first set of genes, a set of features comprising at least one gene-level feature, at least one variant-level feature, and at least one population frequency meta-feature, and a reference label that indicates whether the variant is benign or pathogenic, wherein the at least one population frequency meta-feature quantifies predictive value of allele frequency in the gene (Gao: abstract “”a convolutional neural network-based classifier for variant classification; ¶¶ [0181]-[0187] “a deep learning network for variant pathogenicity classification. The importance of variant classification for clinical applications has inspired numerous attempts to use supervised machine learning to address the problem, but these efforts have been hindered by the lack of an adequately sized truth dataset containing confidently labeled benign and pathogenic variants for training”, [0226]-[0228], [0359], [0390]; FIGS. 1-55);
for each item of the first set of population data, evaluating a variant classification prediction output by the model based on an expected variant classification indicated by the reference label (Gao: abstract; ¶¶ [0203] “classification accuracy shown is the average of the true positive and true negative error rates, using the threshold where the classifier would predict the same number of pathogenic and benign variants as expected based on the enrichment”, [0206] “FIG. 37C shows classification accuracy and AUC for the PrimateAI network and the 20 classifiers listed above. The classification accuracy shown is the average of the true positive and true negative rates, using the threshold where the classifier would predict the same number of pathogenic and benign variants as expected based on the enrichment”; FIGS. 34E, 37C); and
iteratively adjusting a value of at least one parameter or coefficient of the logistic regression model until output of a loss function computed based on the variant classification prediction output by the model satisfies at least one first performance criterion, to produce a trained model, wherein the trained model is capable of outputting variant pathogenicity estimates that satisfy at least one second performance criterion (Gao: abstract; ¶¶ [0096] “convolutional neural network comprises convolution layers which perform the convolution operation between the input values and convolution filters (matrix of weights) that are learned over many gradient update iterations during the training”, [0170]-[0172], [0177]).
Gao does not appear to explicitly teach the following:
logistic regression model.
Lai, however, teaches the following:
logistic regression model (Lai: abstract; pg. 4, col. 2; pg. 7, col. 2).
It would have been obvious to one of ordinary skill in the art, before the effective filing date of the claimed invention, to include the logistic regression model, as taught by Lai, with convolutional neural network for variant classification, as taught by Gao, with the motivation of improving estimation accuracy (Lai: abstract).
CLAIM 2
Gao teaches the method of claim 1, further comprising: using the trained logistic regression model to generate a prediction as to whether the variant is benign or pathogenic (Gao: abstract; ¶¶ [0212]-[0213] “”predict the same number of pathogenic and benign variants).
CLAIM 3
Gao teaches the method of claim 2, further comprising: providing the prediction as to whether the variant is benign or pathogenic to a clinician for use in formulating, by the clinician, a diagnosis of a patient (Gao: abstract; ¶¶ [0166] “help researchers focus on the likely pathogenic genetic variants and accelerate the pace of diagnosis and cure of many diseases”, [0316]-[0317]).
CLAIM 4
Gao teaches the method of claim 1, further comprising: applying the trained logistic regression model to a second set of population data for a plurality of variants of a second plurality of genes; and for each variant of the second plurality of genes, receiving a variant classification prediction output by the trained logistic regression model and, in response to the variant classification prediction satisfying at least the second performance criterion, storing the variant classification prediction in association with the variant for retrieval via at least one query (Gao: abstract; ¶¶ [0169] “use semi-supervised algorithms to construct deep learning-based pathogenicity classifiers that accurately predict pathogenicity of variants”).
CLAIM 9
Gao teaches the method of claim 1, further comprising: selecting, as the set of features, not more than thirty features; and applying the logistic regression model to the selected set of not more than thirty features, wherein the trained logistic regression model is capable of outputting variant pathogenicity estimates that satisfy the at least one second performance criterion based on the selected set of not more than thirty features (Gao: abstract; ¶¶ [0226] “25 amino acids to each side of the variant”, [0267]-[0268]).
CLAIM 14
Gao teaches the method of claim 1, further comprising: estimating the first performance criterion using a means square error or area under the receiver operating characteristic curve (Gao: abstract; ¶¶ [0203], [0206] “FIG. 37B compares different classifiers at separating de novo missense variants in cases versus controls within the 605 genes, shown on a receiver operator characteristic (ROC) curve, with area under the curve (AUC) indicated for each classifier”).
CLAIM 15
Gao teaches the method of claim 1, further comprising: determining the second performance criterion using at least one of decision boundary inspection, feature weight inspection, or a comparison to a benchmark variant classification (Gao: abstract; ¶¶ [0217] “used the 10,000 withheld primate variants in the test dataset to benchmark the deep learning network as well as the other 20 classifiers”, [0301]).
CLAIM 16
Gao teaches the method of claim 1, further comprising: using variant classification estimates output by the trained logistic regression model as an input to a variant classification framework (Gao: abstract; ¶¶ [0252] “we measured the fraction of primate variants in the labeled benign dataset versus the unlabeled dataset. We calculated the probability that a primate variant is a member of the labeled benign dataset, based only on sequencing coverage, using linear regression (FIG. 24). When selecting unlabeled variants to match the primate variants in the labeled benign dataset”, [0319], [0326]-[0329] “deep convolutional neural network-based variant pathogenicity classifier to identify candidate variants in the particular gene that are pathogenic, determining a baseline number of mutations for the particular gene based on summing observed trinucleotide mutation rates of the candidate variants and multiplying the sum with a transmission count and a size of the cohort, applying the deep convolutional neural network-based variant pathogenicity classifier to identify de novo missense variants in the particular gene that are pathogenic, and comparing the baseline number of mutations with a count of the de novo missense variants. Based on an output of the comparison, the per-gene enrichment analysis confirms that the particular gene is associated with the genetic disorder and that the de novo missense variants are pathogenic”; FIG. 24).
CLAIM 17
Gao teaches the method of claim 1, wherein the at least one population frequency meta-feature comprises expected frequency distributions of known benign variants and known pathogenic variants within the gene (Gao: abstract; ¶¶ [0188]-[0189], [0230]).
CLAIM 18
Gao teaches a system, comprising: at least one processor; and at least one memory coupled to the at least one processor, wherein the at least one memory comprises at least one instruction that when executed by the at least one processor causes the at least one processor to perform at least one operation (Gao: abstract; ¶¶ [0038], [0351]).
The remainder of claim 18 repeats substantially the same limitations as those in claim 1. As such, remainder of claim 18 is rejected for substantially the same reasons given for claim 1 and are incorporated herein.
CLAIMS 19-21, 26 AND 31-34
Claims 19-21, 26 and 31-34 repeat substantially the same limitations as those in claims 2-4, 9 and 14-17. As such, claims 19-21, 26 and 31-34 are rejected for substantially the same reasons given for claims 2-4, 9 and 14-17 and are incorporated herein.
CLAIM 35
Gao teaches the least one non-transitory machine-readable medium comprising at least one instruction that when executed by at least one processor causes the at least one processor to perform operations (Gao: abstract; ¶¶ [0038]. [0351]).
The remainder of claim 35 repeats substantially the same limitations as those in claim 1. As such, remainder of claim 35 is rejected for substantially the same reasons given for claim 1 and are incorporated herein.
CLAIMS 36-38, 43 AND 48-51
Claims 36-38, 43, and 48-51 repeat substantially the same limitations as those in claims 2-4, 9, and 14-17. As such, claims 36-38, 43, and 48-51 are rejected for substantially the same reasons given for claims 2-4, 9, and 14-17and are incorporated herein.
4.2. Claims 10, 27, and 44 are rejected under 35 U.S.C. 103 as being unpatentable over Gao et al. (US 2020/0279157), in view of Lai et al. (Non-Patent Literature (NPL): “LEAP: Using machine learning to Support variant classification in a clinical setting”, and further in view of Rocha et al. (Non-Patent Literature (NPL): “The Extent and Impact of Variation in ADME Genes in Sub-Saharan African Populations”).
CLAIM 10
Gao and Lai do not appear to explicitly teach the method of claim 1, further comprising: computing a fixation index, wherein the fixation index comprises sub-population frequency data; including the fixation index in the set of features and applying the logistic regression model to the set of features including the fixation index, wherein the trained logistic regression model is capable of outputting variant pathogenicity estimates that satisfy the at least one second performance criterion based on the set of features including the fixation index.
Rocha, however, teaches computing a fixation index, wherein the fixation index comprises sub-population frequency data (Rocha: pg. 5); including the fixation index in the set of features and applying the logistic regression model to the set of features including the fixation index, wherein the trained logistic regression model is capable of outputting variant pathogenicity estimates that satisfy the at least one second performance criterion based on the set of features including the fixation index (Rocha: pg. 10).
It would have been obvious to one of ordinary skill in the art, before the effective filing date of the claimed invention, to include the fixation index, as taught by Rocha, with the logistic regression model, as taught by Lai, with convolutional neural network for variant classification, as taught by Gao, with the motivation of improving the accuracy of the estimated variant class (Rocha: pg. 5, 10, 13).
CLAIM 27
Claim 27 repeats substantially the same limitations as those in claim 10. As such, claim 27 is rejected for substantially the same reasons given for claim 10 and are incorporated herein.
CLAIM 44
Claim 44 repeats substantially the same limitations as those in claim 10. As such, claim 44 is rejected for substantially the same reasons given for claim 10 and are incorporated herein.
4.3. Claims 12, 29, and 46 are rejected under 35 U.S.C. 103 as being unpatentable over Gao et al. (US 2020/0279157), in view of Lai et al. (Non-Patent Literature (NPL): “LEAP: Using machine learning to Support variant classification in a clinical setting”, and further in view of Keerthi et al. (Non-Patent Literature (NPL): “A Modified Finite Newton Method for Fast Solution of Large Scale Linear SVMs”).
CLAIM 12
Gao and Lai do not appear to explicitly teach the method of claim 1, wherein iteratively adjusting the value of at least one parameter of the logistic regression model comprises adjusting a C value of the logistic regression model until the output of the loss function satisfies the at least one first performance criterion, to produce the trained logistic regression model.
Keerthi, however, teaches wherein iteratively adjusting the value of at least one parameter of the logistic regression model comprises adjusting a C value of the logistic regression model until the output of the loss function satisfies the at least one first performance criterion, to produce the trained logistic regression model (Keerthi: abstract; pg. 6, 16).
It would have been obvious to one of ordinary skill in the art, before the effective filing date of the claimed invention, to include adjusting a C value of the logistic regression model, as taught by Keerthi, with the logistic regression model, as taught by Lai, with convolutional neural network for variant classification, as taught by Gao, with the motivation of improving accuracy (Keerthi: introduction, pg. 6, 8, 16).
CLAIM 29
Claim 29 repeat substantially the same limitations as those in claim 12. As such, claim 29 is rejected for substantially the same reasons given for claim 12 and are incorporated herein.
CLAIM 46
Claim 46 repeats substantially the same limitations as those in claim 12. As such, claim 46 is rejected for substantially the same reasons given for claim 12 and are incorporated herein.
4.4. Claims 13, 30, and 47 are rejected under 35 U.S.C. 103 as being unpatentable over Gao et al. (US 2020/0279157), in view of Lai et al. (Non-Patent Literature (NPL): “LEAP: Using machine learning to Support variant classification in a clinical setting”, and further in view of Lei et al. (US 2020/0111109).
CLAIM 13
Gao and Lai do not appear to explicitly teach the method of claim 1, further comprising: configuring the logistic regression model to model population frequency for variant classification using L1 regularization.
Lei, however, teaches configuring the logistic regression model to model population frequency for variant classification using L1 regularization (Lei: abstract; ¶¶ [0047]-[0050]).
It would have been obvious to one of ordinary skill in the art, before the effective filing date of the claimed invention, to include using L1 regularization to configure a logistic regression model to model population frequency for variant classification, as taught by Lei, with the logistic regression model, as taught by Lai, with convolutional neural network for variant classification, as taught by Gao, with the motivation of improving model accuracy (Lei: abstract; ¶¶ [0085]).
CLAIM 30
Claim 30 repeats substantially the same limitations as those in claim 13. As such, claim 30 is rejected for substantially the same reasons given for claim 13 and are incorporated herein.
CLAIM 47
Claim 47 repeats substantially the same limitations as those in claim 13. As such, claim 47 is rejected for substantially the same reasons given for claim 13 and are incorporated herein.
Double Patenting
5. The nonstatutory double patenting rejection is based on a judicially created doctrine grounded in public policy (a policy reflected in the statute) so as to prevent the unjustified or improper timewise extension of the “right to exclude” granted by a patent and to prevent possible harassment by multiple assignees. A nonstatutory double patenting rejection is appropriate where the conflicting claims are not identical, but at least one examined application claim is not patentably distinct from the reference claim(s) because the examined application claim is either anticipated by, or would have been obvious over, the reference claim(s). See, e.g., In re Berg, 140 F.3d 1428, 46 USPQ2d 1226 (Fed. Cir. 1998); In re Goodman, 11 F.3d 1046, 29 USPQ2d 2010 (Fed. Cir. 1993); In re Longi, 759 F.2d 887, 225 USPQ 645 (Fed. Cir. 1985); In re Van Ornum, 686 F.2d 937, 214 USPQ 761 (CCPA 1982); In re Vogel, 422 F.2d 438, 164 USPQ 619 (CCPA 1970); In re Thorington, 418 F.2d 528, 163 USPQ 644 (CCPA 1969).
A timely filed terminal disclaimer in compliance with 37 CFR 1.321(c) or 1.321(d) may be used to overcome an actual or provisional rejection based on nonstatutory double patenting provided the reference application or patent either is shown to be commonly owned with the examined application, or claims an invention made as a result of activities undertaken within the scope of a joint research agreement. See MPEP § 717.02 for applications subject to examination under the first inventor to file provisions of the AIA as explained in MPEP § 2159. See MPEP § 2146 et seq. for applications not subject to examination under the first inventor to file provisions of the AIA . A terminal disclaimer must be signed in compliance with 37 CFR 1.321(b).
The filing of a terminal disclaimer by itself is not a complete reply to a nonstatutory double patenting (NSDP) rejection. A complete reply requires that the terminal disclaimer be accompanied by a reply requesting reconsideration of the prior Office action. Even where the NSDP rejection is provisional the reply must be complete. See MPEP § 804, subsection I.B.1. For a reply to a non-final Office action, see 37 CFR 1.111(a). For a reply to final Office action, see 37 CFR 1.113(c). A request for reconsideration while not provided for in 37 CFR 1.113(c) may be filed after final for consideration. See MPEP §§ 706.07(e) and 714.13.
The USPTO Internet website contains terminal disclaimer forms which may be used. Please visit www.uspto.gov/patent/patents-forms. The actual filing date of the application in which the form is filed determines what form (e.g., PTO/SB/25, PTO/SB/26, PTO/AIA /25, or PTO/AIA /26) should be used. A web-based eTerminal Disclaimer may be filled out completely online using web-screens. An eTerminal Disclaimer that meets all requirements is auto-processed and approved immediately upon submission. For more information about eTerminal Disclaimers, refer to www.uspto.gov/patents/apply/applying-online/eterminal-disclaimer.
5.1. Claims 1-4, 9-10, 12-21, 26-27, 29-38, 44, and 46-51 are rejected on the ground of nonstatutory double patenting as being unpatentable over claims 1-4, 8-9, 11-20, 24-25, 27-36, 41 and 43-48, respectively, of U.S. Patent No. 12,191,001. Although the claims at issue are not identical, they are not patentably distinct from each other because the examined application claim(s) is/are either anticipated by, or would have been obvious over, the reference claim(s) because all of the limitations of claims 1-4, 9-10, 12-21, 26-27, 29-38, 43-44, and 46-51 of instant pending patent application 18/943513 correspond to limitations recited in claims 1-4, 8-9, 11-20, 24-25, 27-36, 41 and 43-48 of U.S. Patent 12,191,001. Any claim limitation differences are not substantively significant and/or are obvious under a broad and reasonable interpretation, as detailed in the comparison chart below.
PENDING CLAIMS (18/943513)
US PATENT 12,191,001
CLAIM 1
A method for configuring a machine learning model to model population frequency for variant classification, the method comprising:
applying a logistic regression model to a first set of population data for a first set of genes, wherein an item of the first set of population data comprises, for a variant located at a position within a gene of the first set of genes, a set of features comprising at least one gene-level feature, at least one variant-level feature, and at least one population frequency meta-feature, and a reference label that indicates whether the variant is benign or pathogenic, wherein the at least one population frequency meta-feature quantifies predictive value of allele frequency in the gene;
for each item of the first set of population data, evaluating a variant classification prediction output by the logistic regression model based on an expected variant classification indicated by the reference label; and
iteratively adjusting a value of at least one parameter or coefficient of the logistic regression model until output of a loss function computed based on the variant classification prediction output by the logistic regression model satisfies at least one first performance criterion, to produce a trained logistic regression model, wherein the trained logistic regression model is capable of outputting variant pathogenicity estimates that satisfy at least one second performance criterion.
CLAIM 1
A method for configuring a machine learning model to model population frequency for variant classification, the method comprising:
applying a logistic regression model to a first set of population data for a first set of genes, wherein an item of the first set of population data comprises, for a variant located at a position within a gene of the first set of genes, a set of features comprising at least one gene-level feature, at least one variant-level feature, and at least one population frequency meta-feature, and a reference label that indicates whether the variant is benign or pathogenic, wherein the at least one population frequency meta-feature quantifies predictive value of allele frequency in the gene, wherein the applying comprises computing a gene-level constraint; including the gene-level constraint in the at least one gene-level feature; computing an allele frequency; including the allele frequency in the at least one variant-level feature; including, in the at least one population frequency meta-feature, a mathematical combination of the gene-level constraint and the allele frequency; and
applying the logistic regression model to the set of features including the mathematical combination of the gene-level constraint and the allele frequency; wherein the trained logistic regression model is capable of outputting variant pathogenicity estimates that satisfy the at least one second performance criterion based on the set of features including the mathematical combination of the gene-level constraint and the allele frequency;
for each item of the first set of population data, evaluating a variant classification prediction output by the logistic regression model based on an expected variant classification indicated by the reference label; and
iteratively adjusting a value of at least one parameter or coefficient of the logistic regression model until output of a loss function computed based on the variant classification prediction output by the logistic regression model satisfies at least one first performance criterion, to produce a trained logistic regression model, wherein the trained logistic regression model is capable of outputting variant pathogenicity estimates that satisfy at least one second performance criterion.
CLAIM 2
The method of claim 1, further comprising: using the trained logistic regression model to generate a prediction as to whether the variant is benign or pathogenic.
CLAIM 2
The method of claim 1, further comprising: using the trained logistic regression model to generate a prediction as to whether the variant is benign or pathogenic.
CLAIM 3
The method of claim 2, further comprising: providing the prediction as to whether the variant is benign or pathogenic to a clinician for use in formulating, by the clinician, a diagnosis of a patient.
CLAIM 3
The method of claim 2, further comprising: providing the prediction as to whether the variant is benign or pathogenic to a clinician for use in formulating, by the clinician, a diagnosis of a patient.
CLAIM 4
The method of claim 1, further comprising:
applying the trained logistic regression model to a second set of population data for a plurality of variants of a second plurality of genes; and
for each variant of the second plurality of genes, receiving a variant classification prediction output by the trained logistic regression model and, in response to the variant classification prediction satisfying at least the second performance criterion, storing the variant classification prediction in association with the variant for retrieval via at least one query.
CLAIM 4
The method of claim 1, further comprising:
applying the trained logistic regression model to a second set of population data for a plurality of variants of a second plurality of genes; and
for each variant of the second plurality of genes, receiving a variant classification prediction output by the trained logistic regression model and, in response to the variant classification prediction satisfying at least the second performance criterion, storing the variant classification prediction in association with the variant for retrieval via at least one query.
CLAIM 9
The method of claim 1, further comprising:
selecting, as the set of features, not more than thirty features; and
applying the logistic regression model to the selected set of not more than thirty features,
wherein the trained logistic regression model is capable of outputting variant pathogenicity estimates that satisfy the at least one second performance criterion based on the selected set of not more than thirty features.
CLAIM 8
The method of claim 1, further comprising:
selecting, as the set of features, not more than thirty features; and
applying the logistic regression model to the selected set of not more than thirty features,
wherein the trained logistic regression model is capable of outputting variant pathogenicity estimates that satisfy the at least one second performance criterion based on the selected set of not more than thirty features.
CLAIM 10
The method of claim 1, further comprising:
computing a fixation index, wherein the fixation index comprises sub-population frequency data;
including the fixation index in the set of features; and
applying the logistic regression model to the set of features including the fixation index, wherein the trained logistic regression model is capable of outputting variant pathogenicity estimates that satisfy the at least one second performance criterion based on the set of features including the fixation index.
CLAIM 9
The method of claim 1, further comprising:
computing a fixation index, wherein the fixation index comprises sub-population frequency data;
including the fixation index in the set of features; and
applying the logistic regression model to the set of features including the fixation index, wherein the trained logistic regression model is capable of outputting variant pathogenicity estimates that satisfy the at least one second performance criterion based on the set of features including the fixation index.
CLAIM 12
The method of claim 1, wherein iteratively adjusting the value of at least one parameter of the logistic regression model comprises adjusting a C value of the logistic regression model until the output of the loss function satisfies the at least one first performance criterion, to produce the trained logistic regression model.
CLAIM 11
The method of claim 1, wherein iteratively adjusting the value of at least one parameter of the logistic regression model comprises adjusting a C value of the logistic regression model until the output of the loss function satisfies the at least one first performance criterion, to produce the trained logistic regression model.
CLAIM 13
The method of claim 1, further comprising:
configuring the logistic regression model to model population frequency for variant classification using L1 regularization.
CLAIM 12
The method of claim 1, further comprising:
configuring the logistic regression model to model population frequency for variant classification using L1 regularization.
CLAIM 14
The method of claim 1, further comprising:
estimating the first performance criterion using a means square error or area under the receiver operating characteristic curve.
CLAIM 13
The method of claim 1, further comprising:
estimating the first performance criterion using a means square error or area under the receiver operating characteristic curve.
CLAIM 15
The method of claim 1, further comprising:
determining the second performance criterion using at least one of decision boundary inspection, feature weight inspection, or a comparison to a benchmark variant classification.
CLAIM 14
The method of claim 1, further comprising:
determining the second performance criterion using at least one of decision boundary inspection, feature weight inspection, or a comparison to a benchmark variant classification.
CLAIM 16
The method of claim 1, further comprising:
using variant classification estimates output by the trained logistic regression model as an input to a variant classification framework.
CLAIM 15
The method of claim 1, further comprising:
using variant classification estimates output by the trained logistic regression model as an input to a variant classification framework.
CLAIM 17
The method of claim 1, wherein the at least one population frequency meta-feature comprises expected frequency distributions of known benign variants and known pathogenic variants within the gene.
CLAIM 16
The method of claim 1, wherein the at least one population frequency meta-feature comprises expected frequency distributions of known benign variants and known pathogenic variants within the gene.
CLAIM 18
A system, comprising:
at least one processor; and
at least one memory coupled to the at least one processor, wherein the at least one memory comprises at least one instruction that when executed by the at least one processor causes the at least one processor to perform at least one operation comprising:
applying a logistic regression model to a first set of population data for a first set of genes, wherein an item of the first set of population data comprises, for a variant located at a position within a gene of the first set of genes, a set of features comprising at least one gene-level feature, at least one variant-level feature, and at least one population frequency meta-feature, and a reference label that indicates whether the variant is benign or pathogenic, wherein the at least one population frequency meta-feature quantifies predictive value of allele frequency in the gene;
for each item of the first set of population data, evaluating a variant classification prediction output by the logistic regression model based on an expected variant classification indicated by the reference label; and
adjusting a value of at least one parameter or coefficient of the logistic regression model until at least one first performance criterion is satisfied to produce a trained logistic regression model, wherein the trained logistic regression model is capable of outputting variant pathogenicity estimates that satisfy at least one second performance criterion.
CLAIM 17
A system, comprising:
at least one processor; and
at least one memory coupled to the at least one processor, wherein the at least one memory comprises at least one instruction that when executed by the at least one processor causes the at least one processor to perform at least one operation comprising:
applying a logistic regression model to a first set of population data for a first set of genes, wherein an item of the first set of population data comprises, for a variant located at a position within a gene of the first set of genes, a set of features comprising at least one gene-level feature, at least one variant-level feature, and at least one population frequency meta-feature, and a reference label that indicates whether the variant is benign or pathogenic, wherein the at least one population frequency meta-feature quantifies predictive value of allele frequency in the gene, wherein the applying comprises computing a gene-level constraint; including the gene-level constraint in the at least one gene-level feature; computing an allele frequency;
including the allele frequency in the at least one variant-level feature; including, in the at least one population frequency meta-feature, a mathematical combination of the gene-level constraint and the allele frequency; and
applying the logistic regression model to the set of features including the mathematical combination of the gene-level constraint and the allele frequency; wherein the trained logistic regression model is capable of outputting variant pathogenicity estimates that satisfy the at least one second performance criterion based on the set of features including the mathematical combination of the gene-level constraint and the allele frequency;
for each item of the first set of population data, evaluating a variant classification prediction indicated by the reference label; and
adjusting a value of at least one parameter or coefficient of the logistic regression model until at least one first performance criterion is satisfied to produce a trained logistic regression model, wherein the trained logistic regression model is capable of outputting variant pathogenicity estimates that satisfy at least one second performance criterion.
CLAIM 19
The system of claim 18, wherein the at least one memory further comprises at least one instruction that when executed by the at least one processor causes the at least one processor to perform at least one operation comprising:
using the trained logistic regression model to generate a prediction as to whether the variant is benign or pathogenic.
CLAIM 18
The system of claim 17, wherein the at least one memory further comprises at least one instruction that when executed by the at least one processor causes the at least one processor to perform at least one operation comprising:
using the trained logistic regression model to generate a prediction as to whether the variant is benign or pathogenic.
CLAIM 20
The system of claim 19, wherein the at least one memory further comprises at least one instruction that when executed by the at least one processor causes the at least one processor to perform at least one operation comprising:
providing the prediction as to whether the variant is benign or pathogenic to a clinician for use in formulating, by the clinician, a diagnosis of a patient.
CLAIM 19
The system of claim 18, wherein the at least one memory further comprises at least one instruction that when executed by the at least one processor causes the at least one processor to perform at least one operation comprising:
providing the prediction as to whether the variant is benign or pathogenic to a clinician for use in formulating, by the clinician, a diagnosis of a patient.
CLAIM 21
The system of claim 18, wherein the at least one memory further comprises at least one instruction that when executed by the at least one processor causes the at least one processor to perform at least one operation comprising:
applying the trained logistic regression model to a second set of population data for a plurality of variants of a second plurality of genes; and
for each variant of the second plurality of genes, receiving a variant classification prediction output by the trained logistic regression model and, in response to the variant classification prediction satisfying at least the second performance criterion, storing the variant classification prediction in association with the variant for retrieval via at least one query.
CLAIM 20
The system of claim 17, wherein the at least one memory further comprises at least one instruction that when executed by the at least one processor causes the at least one processor to perform at least one operation comprising:
applying the trained logistic regression model to a second set of population data for a plurality of variants of a second plurality of genes; and
for each variant of the second plurality of genes, receiving a variant classification prediction output by the trained logistic regression model and, in response to the variant classification prediction satisfying at least the second performance criterion, storing the variant classification prediction in association with the variant for retrieval via at least one query.
CLAIM 26
The system of claim 18, wherein the at least one memory further comprises at least one instruction that when executed by the at least one processor causes the at least one processor to perform at least one operation comprising:
selecting, as the set of features, not more than thirty features; and
applying the logistic regression model to the selected set of not more than thirty features, wherein the trained logistic regression model is capable of outputting variant pathogenicity estimates that satisfy the at least one second performance criterion based on the selected set of not more than thirty features.
CLAIM 24
The system of claim 17, wherein the at least one memory further comprises at least one instruction that when executed by the at least one processor causes the at least one processor to perform at least one operation comprising:
selecting, as the set of features, not more than thirty features; and
applying the logistic regression model to the selected set of not more than thirty features, wherein the trained logistic regression model is capable of outputting variant pathogenicity estimates that satisfy the at least one second performance criterion based on the selected set of not more than thirty features.
CLAIM 27
The system of claim 18, wherein the at least one memory further comprises at least one instruction that when executed by the at least one processor causes the at least one processor to perform at least one operation comprising:
computing a fixation index, wherein the fixation index comprises sub-population frequency data;
including the fixation index in the set of features; and
applying the logistic regression model to the set of features including the fixation index, wherein the trained logistic regression model is capable of outputting variant pathogenicity estimates that satisfy the at least one second performance criterion based on the set of features including the fixation index.
CLAIM 25
The system of claim 17, wherein the at least one memory further comprises at least one instruction that when executed by the at least one processor causes the at least one processor to perform at least one operation comprising:
computing a fixation index, wherein the fixation index comprises sub-population frequency data;
including the fixation index in the set of features; and
applying the logistic regression model to the set of features including the fixation index, wherein the trained logistic regression model is capable of outputting variant pathogenicity estimates that satisfy the at least one second performance criterion based on the set of features including the fixation index.
CLAIM 29
The system of claim 18, wherein iteratively adjusting the value of at least one parameter of the logistic regression model comprises adjusting a C value of the logistic regression model until the output of the loss function satisfices the at least one first performance criterion, to produce the trained logistic regression model.
CLAIM 27
The system of claim 17, wherein iteratively adjusting the value of at least one parameter of the logistic regression model comprises adjusting a C value of the logistic regression model until the output of the loss function satisfies the at least one first performance criterion, to produce the trained logistic regression model.
CLAIM 30
The system of claim 18, wherein the at least one memory further comprises at least one instruction that when executed by the at least one processor causes the at least one processor to perform at least one operation comprising:
configuring the logistic regression model to model population frequency for variant classification using L1 regularization.
CLAIM 28
The system of claim 17, wherein the at least one memory further comprises at least one instruction that when executed by the at least one processor causes the at least one processor to perform at least one operation comprising:
configuring the logistic regression model to model population frequency for variant classification using L1 regularization.
CLAIM 31
The system of claim 18, wherein the at least one memory further comprises at least one instruction that when executed by the at least one processor causes the at least one processor to perform at least one operation comprising:
estimating the first performance criterion using a means square error or area under the receiver operating characteristic curve.
CLAIM 29
The system of claim 17, wherein the at least one memory further comprises at least one instruction that when executed by the at least one processor causes the at least one processor to perform at least one operation comprising:
estimating the first performance criterion using a means square error or area under the receiver operating characteristic curve.
CLAIM 32
The system of claim 18, wherein the at least one memory further comprises at least one instruction that when executed by the at least one processor causes the at least one processor to perform at least one operation comprising:
determining the second performance criterion using at least one of decision boundary inspection, feature weight inspection, or a comparison to a benchmark variant classification.
CLAIM 30
The system of claim 17, wherein the at least one memory further comprises at least one instruction that when executed by the at least one processor causes the at least one processor to perform at least one operation comprising:
determining the second performance criterion using at least one of decision boundary inspection, feature weight inspection, or a comparison to a benchmark variant classification.
CLAIM 33
The system of claim 18, wherein the at least one memory further comprises at least one instruction that when executed by the at least one processor causes the at least one processor to perform at least one operation comprising:
using variant classification estimates output by the trained logistic regression model as an input to a variant classification framework.
CLAIM 31
The system of claim 17, wherein the at least one memory further comprises at least one instruction that when executed by the at least one processor causes the at least one processor to perform at least one operation comprising:
using variant classification estimates output by the trained logistic regression model as an input to a variant classification framework.
CLAIM 34
The system of claim 18, wherein the at least one population frequency meta-feature comprises expected frequency distributions of known benign variants and known pathogenic variants within the gene.
CLAIM 32
The system of claim 17, wherein the at least one population frequency meta-feature comprises expected frequency distributions of known benign variants and known pathogenic variants within the gene.
CLAIM 35
At least one non-transitory machine-readable medium comprising at least one instruction that when executed by at least one processor causes the at least one processor to perform operations comprising:
applying a logistic regression model to a first set of population data for a first set of genes, wherein an item of the first set of population data comprises, for a variant located at a position within a gene of the first set of genes, a set of features comprising at least one population frequency meta-feature and a reference label that indicates whether the variant is benign or pathogenic, wherein the at least one population frequency meta-feature quantifies predictive value of allele frequency in the gene;
for each item of the first set of population data, evaluating a variant classification prediction output by the logistic regression model based on an expected variant classification indicated by the reference label; and
adjusting a value of at least one parameter or coefficient of the logistic regression model until at least one first performance criterion is satisfied to produce a trained logistic regression model, wherein the trained logistic regression model is capable of outputting variant pathogenicity estimates that satisfy at least one second performance criterion.
CLAIM 33
At least one non-transitory machine-readable medium comprising at least one instruction that when executed by at least one processor causes the at least one processor to perform operations comprising:
applying a logistic regression model to a first set of population data for a first set of genes, wherein an item of the first set of population data comprises, for a variant located at a position within a gene of the first set of genes, a set of features comprising at least one population frequency meta-feature and a reference label that indicates whether the variant is benign or pathogenic, wherein the at least one population frequency meta-feature quantifies predictive value of allele frequency in the gene, wherein the applying comprises computing a gene-level constraint;
including the gene-level constraint in the at least one gene-level feature; computing an allele frequency; including the allele frequency in the at least one variant-level feature; including, in the at least one population frequency meta-feature, a mathematical combination of the gene-level constraint and the allele frequency; and
applying the logistic regression model to the set of features including the mathematical combination of the gene-level constraint and the allele frequency; wherein the trained logistic regression model is capable of outputting variant pathogenicity estimates that satisfy the at least one second performance criterion based on the set of features including the mathematical combination of the gene-level constraint and the allele frequency;
for each item of the first set of population data, evaluating a variant classification prediction indicated by the reference label; and
adjusting a value of at least one parameter or coefficient of the logistic regression model until at least one first performance criterion is satisfied to produce a trained logistic regression model, wherein the trained logistic regression model is capable of outputting variant pathogenicity estimates that satisfy at least one second performance criterion.
CLAIM 36
The at least one non-transitory machine-readable medium of claim 35, further comprising at least one instruction that when executed by the at least one processor causes the at least one processor to perform at least one operation comprising:
using the trained logistic regression model to generate a prediction as to whether the variant is benign or pathogenic.
CLAIM 34
The at least one non-transitory machine-readable medium of claim 33, further comprising at least one instruction that when executed by the at least one processor causes the at least one processor to perform at least one operation comprising:
using the trained logistic regression model to generate a prediction as to whether the variant is benign or pathogenic.
CLAIM 37
The at least one non-transitory machine-readable medium of claim 36, further comprising at least one instruction that when executed by the at least one processor causes the at least one processor to perform at least one operation comprising:
providing the prediction as to whether the variant is benign or pathogenic to a clinician for use in formulating, by the clinician, a diagnosis of a patient.
CLAIM 35
The at least one non-transitory machine-readable medium of claim 34, further comprising at least one instruction that when executed by the at least one processor causes the at least one processor to perform at least one operation comprising:
providing the prediction as to whether the variant is benign or pathogenic to a clinician for use in formulating, by the clinician, a diagnosis of a patient.
CLAIM 38
The at least one non-transitory machine-readable medium of claim 35, further comprising at least one instruction that when executed by the at least one processor causes the at least one processor to perform at least one operation comprising:
applying the trained logistic regression model to a second set of population data for a plurality of variants of a second plurality of genes; and
for each variant of the second plurality of genes, receiving a variant classification prediction output by the trained logistic regression model and, in response to the variant classification prediction satisfying at least the second performance criterion, storing the variant classification prediction in association with the variant for retrieval via at least one query.
CLAIM 36
The at least one non-transitory machine-readable medium of claim 33, further comprising at least one instruction that when executed by the at least one processor causes the at least one processor to perform at least one operation comprising:
applying the trained logistic regression model to a second set of population data for a plurality of variants of a second plurality of genes; and
for each variant of the second plurality of genes, receiving a variant classification prediction output by the trained logistic regression model and, in response to the variant classification prediction satisfying at least the second performance criterion, storing the variant classification prediction in association with the variant for retrieval via at least one query.
CLAIM 43
The at least one non-transitory machine-readable medium of claim 35, further comprising at least one instruction that when executed by the at least one processor causes the at least one processor to perform at least one operation comprising:
selecting, as the set of features, not more than thirty features; and
applying the logistic regression model to the selected set of not more than thirty features, wherein the trained logistic regression model is capable of outputting variant pathogenicity estimates that satisfy the at least one second performance criterion based on the selected set of not more than thirty features.
CLAIM 44
The at least one non-transitory machine-readable medium of claim 35 further comprising at least one instruction that when executed by the at least one processor causes the at least one processor to perform at least one operation comprising:
computing a fixation index, wherein the fixation index comprises sub-population frequency data;
including the fixation index in the set of features; and
applying the logistic regression model to the set of features including the fixation index, wherein the trained logistic regression model is capable of outputting variant pathogenicity estimates that satisfy the at least one second performance criterion based on the set of features including the fixation index.
CLAIM 41
The at least one non-transitory machine-readable medium of claim 33 further comprising at least one instruction that when executed by the at least one processor causes the at least one processor to perform at least one operation comprising:
[See CLAIM 9 and 25]
including the fixation index in the set of features; and
applying the logistic regression model to the set of features including the fixation index, wherein the trained logistic regression model is capable of outputting variant pathogenicity estimates that satisfy the at least one second performance criterion based on the set of features including the fixation index.
CLAIM 46
The at least one non-transitory machine-readable medium of claim 35, wherein adjusting the value of at least one parameter of the logistic regression model comprises adjusting a C value of the logistic regression model until output of a loss function satisfices the at least one first performance criterion, to produce the trained logistic regression model.
CLAIM 43
The at least one non-transitory machine-readable medium of claim 33, wherein adjusting the value of at least one parameter of the logistic regression model comprises adjusting a C value of the logistic regression model until output of a loss function satisfies the at least one first performance criterion, to produce the trained logistic regression model.
CLAIM 47
The at least one non-transitory machine-readable medium of claim 35, further comprising at least one instruction that when executed by the at least one processor causes the at least one processor to perform at least one operation comprising:
configuring the logistic regression model to model population frequency for variant classification using L1 regularization.
CLAIM 44
The at least one non-transitory machine-readable medium of claim 33, further comprising at least one instruction that when executed by the at least one processor causes the at least one processor to perform at least one operation comprising:
configuring the logistic regression model to model population frequency for variant classification using L1 regularization.
CLAIM 48
The at least one non-transitory machine-readable medium of claim 35, further comprising at least one instruction that when executed by the at least one processor causes the at least one processor to perform at least one operation comprising:
estimating the at least one first performance criterion using a means square error or area under the receiver operating characteristic curve.
CLAIM 45
The at least one non-transitory machine-readable medium of claim 33, further comprising at least one instruction that when executed by the at least one processor causes the at least one processor to perform at least one operation comprising:
estimating the at least one first performance criterion using a means square error or area under the receiver operating characteristic curve.
CLAIM 49
The at least one non-transitory machine-readable medium of claim 35, further comprising at least one instruction that when executed by the at least one processor causes the at least one processor to perform at least one operation comprising:
determining the second performance criterion using at least one of decision boundary inspection, feature weight inspection, or a comparison to a benchmark variant classification.
CLAIM 46
The at least one non-transitory machine-readable medium of claim 33, further comprising at least one instruction that when executed by the at least one processor causes the at least one processor to perform at least one operation comprising:
determining the second performance criterion using at least one of decision boundary inspection, feature weight inspection, or a comparison to a benchmark variant classification.
CLAIM 50
The at least one non-transitory machine-readable medium of claim 35, further comprising at least one instruction that when executed by the at least one processor causes the at least one processor to perform at least one operation comprising:
using variant classification estimates output by the trained logistic regression model as an input to a variant classification framework.
CLAIM 47
The at least one non-transitory machine-readable medium of claim 33, further comprising at least one instruction that when executed by the at least one processor causes the at least one processor to perform at least one operation comprising:
using variant classification estimates output by the trained logistic regression model as an input to a variant classification framework.
CLAIM 51
The at least one non-transitory machine-readable medium of claim 35, wherein the at least one population frequency meta-feature comprises expected frequency distributions of known benign variants and known pathogenic variants within the gene.
CLAIM 48
The at least one non-transitory machine-readable medium of claim 33, wherein the at least one population frequency meta-feature comprises expected frequency distributions of known benign variants and known pathogenic variants within the gene.
Conclusion
6. Any inquiry concerning this communication or earlier communications from the examiner should be directed to Michael Tomaszewski whose telephone number is (313)446-4863. The examiner can normally be reached M-F 5:30 am - 2:30 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, Peter H Choi can be reached at (469) 295-9171. 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.
/MICHAEL TOMASZEWSKI/Primary Examiner, Art Unit 3681
Prosecution Insights