DETAILED ACTION
Notice of Pre-AIA or AIA Status
The present application, filed on or after March 16, 2013, is being examined under the first inventor to file provisions of the AIA .
Priority
Instant application eligible to benefit from the Foreign Application as claimed by applicant on 12/23/2021 and effective filling date was considered as 12/23/2021.
Information Disclosure Statement
IDS submitted on 11/24/2022 has been considered by the examiner.
Drawings
The drawings filed on 11/03/2022 are accepted.
Specification
The specification filed on 11/24/2022 is accepted.
Claim status
Claims 1-20 are pending and examined on the merits.
Claim 1-20 are rejected.
Claim Interpretation
The following is a quotation of 35 U.S.C. 112(f):
(f) Element in Claim for a Combination. – An element in a claim for a combination may be expressed as a means or step for performing a specified function without the recital of structure, material, or acts in support thereof, and such claim shall be construed to cover the corresponding structure, material, or acts described in the specification and equivalents thereof.
The following is a quotation of pre-AIA 35 U.S.C. 112, sixth paragraph:
An element in a claim for a combination may be expressed as a means or step for performing a specified function without the recital of structure, material, or acts in support thereof, and such claim shall be construed to cover the corresponding structure, material, or acts described in the specification and equivalents thereof.
Under MPEP 2181, section I, a limitation invokes 112(f) if it meets the following:
Uses "means" or a generic placeholder (nonce word) for structure.
The term is modified by functional language.
The term is not modified by sufficient structure to perform the function.
Based on the three-prong analysis for evaluating 35 U.S.C. 112(f) (means-plus-function) limitations, as outlined in MPEP § 2181, the claim 10 and 20 triggers 112(f) issues, specifically for the limitations “scale-invariant feature transform unit,” “polymorphism phenotype analysis unit,” and “site hazard prediction unit.”
Scale-invariant feature transform unit
Prong A (Generic Placeholder): The term “unit” is a nonce word used as a substitute for “means”.
Prong B (Functional Language): The unit is modified by functional language: “configured to performed to produce the functional score”
Prong C (Insufficient Structure): The term “unit” does not support specific, known physical structure, such as “sensor”. The description “scale-invariant feature transform unit,” describes what it does (predicts) rather than what it is.
Therefore, 112(f) Invoked.
Polymorphism phenotype analysis unit
Prong A (Generic Placeholder): "Unit" is a nonce word.
Prong B (Functional Language): Modified by “performed to produce the functional score”.
Prong C (Insufficient Structure): No specific structure (e.g., specific hardware) is disclosed within the claim term itself to perform the functional score.
Therefore, 112(f) Invoked.
Site hazard prediction unit
Prong A (Generic Placeholder): “Unit” is a nonce word.
Prong B (Functional Language): Modified by “performed to produce the functional score”.
Prong C (Insufficient Structure): “site hazard prediction unit” is a pure functional term for a software. No specific structure (e.g., specific hardware) is disclosed within the claim term itself to perform the functional score.
Therefore, 112(f) Invoked.
In the specification applicant does not discloses for each of the units with sufficient structure”. The specification paragraphs [0014, 0025] describes the procedure for generating a functional score using the hazard prediction tool begins by determining whether the mutation sites in the provided variant sequences are missense or splicing variants. For missense variants, the tool utilizes scale-invariant feature transform (SIFT) and polymorphism phenotype analysis to compute the score. For splicing variants, it applies site hazard prediction to determine the functional score.
Because the claim, when read in view of the specification, does not connect the “units” to a specific structure, it does not satisfy the requirements for definiteness and structure. So, claim 10 is not patentable under U.S.C. 112(a) for lack of written description of the structures of the unit.
Claim Rejections - 35 USC § 112(b)
The following is a quotation of 35 U.S.C. 112(b):
(b) CONCLUSION. —The specification shall conclude with one or more claims particularly pointing out and distinctly claiming the subject matter which the inventor or a joint inventor regards as the invention.
The following is a quotation of 35 U.S.C. 112 (pre-AIA ), second paragraph:
The specification shall conclude with one or more claims particularly pointing out and distinctly claiming the subject matter which the applicant regards as his invention.
Claims 1-20 are rejected under 35 U.S.C. 112(b) or 35 U.S.C. 112 (pre-AIA ), second paragraph, as being indefinite for failing to particularly point out and distinctly claim the subject matter which the inventor or a joint inventor (or for applications subject to pre-AIA 35 U.S.C. 112, the applicant), regards as the invention.
The claim 1 and 11 recites that “mutation sites suffer from a disease.” This language is internally inconsistent and nonsensical. “Mutation sites” are specific locations in a genetic sequence (i.e., loci or alleles), which are inert, biochemical markers. They are physically incapable of “suffering” from a disease, as a disease is a pathological condition experienced by a living organism or cell.
Applicant attempts to clarify this in the reasoning by stating: “when the pathogenic score is higher, the probability of the mutation sites suffering from the corresponding disease is higher." However, this explanation compounds the indefiniteness. It conflates the probability of a subject (such as a patient) developing a disease with the literal sites themselves experiencing a medical condition.
Because it is unclear whether the applicant intends to claim the risk of a patient developing a disease, the probability of a mutation being pathogenic, or the clinical diagnosis of a specific pathology, the metes and bounds of the claim scope are rendered unclear. One skilled in the art is left to speculate as to the exact boundaries of what constitutes infringement (e.g., whether infringement occurs when calculating a score, or when an individual actually manifests a disease state).
Regarding claim 2, and 12 The term “special target disease or gene selection” is indefinite. One of ordinary skill in the art would not know the bounds of what constitutes a “special” selection versus a general selection. The term “special” is a term of degree that does not provide a standard for measuring the scope (MPEP 2173.05(b).
The specification fails to provide a clear definition, test, or objective standard for what qualifies as a “special” selection. Without an explicit definition provided in the written description, the scope of the claim is left to subjective interpretation.
One skilled in the art would not recognize the limitation with reasonable certainty, as there is no known objective criterion in the art to distinguish between a "special" selection and a standard or general one. Therefore, the boundaries of the protected subject matter are unclear.
Regarding claim 9, and 19 recites “judging whether it is a patient based on the related information...”.
In the step of “judging whether it is a patient based on the related information,” the pronoun “it” lacks proper antecedent basis. It is entirely unclear what specific subject, sample, or individual "it" refers to (e.g., a subject, a biological sample, or a specific genetic profile). Because the antecedent for "it" is missing, the claim fails to provide a clear, definite boundary of the steps required in the classification method. One of ordinary skill in the art would not understand the scope of the limitation with reasonable certainty. Because there is no criterion for “it”, the boundaries of the claim remain ill-defined. Therefore, the lack of clear boundaries makes it impossible to determine the scope of the claimed method [MPEP 2173.05(b)]
Claim Rejections - 35 USC § 101
35 U.S.C. 101 reads as follows:
Whoever invents or discovers any new and useful process, machine, manufacture, or composition of matter, or any new and useful improvement thereof, may obtain a patent therefor, subject to the conditions and requirements of this title.
Claims 1-20 are rejected under 35 U.S.C. 101 because the claimed invention is directed to an abstract idea without significantly more.
Step 2A, Prong 1
In accordance with MPEP § 2106, the instant claims 1-10, are drawn to a process (method), claims 11-20 are drawn to a CRM, and therefore are found to recite statutory subject matter (Step 1: YES). The instant claims are then analyzed to determine if the claims recite any concepts that equate to an abstract idea, law of nature or natural phenomenon (Step 2A, Prong 1). The instant claims recite the following limitations that equate to an abstract idea:
Claims 1 and 11 recite
Producing a population score … based on the related information. (Mathematical Calculation)
Producing a variant type score …. tool based on the related information. (Mathematical Calculation)
Producing a clinical score …. related information. (Mathematical Calculation)
Producing a functional score …. tool based on the related information. (Mathematical Calculation)
Summing the population score …. to produce a pathogenic score. (Mathematical Calculation)
Determining probability … based on the pathogenic score, wherein …. disease is higher. (Mental process and mathematical concept and law of nature)
The steps: scoring using databases, and summing scores are mathematical algorithms or mental processes and can be performed mentally or with the help of a pencil and paper, focusing on the manipulation of data rather than a specific physical process.
Claim 2 and 12 recites “the related information further comprises loss-of-function test data, …. gene selection, or a combination thereof”. (Natural phenomena or correlations occurring in nature).
Claim 3 recites
Performing a frequency variation analysis …. to generate a first population score. (Mathematical concept)
Performing a homozygous observational analysis …. second population score. (Mathematical concept)
Summing the first population score and …. to obtain the population score. (Mathematical concept)
Above steps in claim 3 are essentially mathematical calculations or relationships applied to data.
Claim 4 and 14 recites “performing the frequency variation analysis … database is greater than a predetermined threshold number, … to perform the frequency variation analysis”. (Mathematical concept and Mental process)
Claim 5 and 15 recites “Performing the homozygous observational analysis …. are greater than a predetermined threshold number, ….. observational analysis is not performed.” (Mathematical concept and Mental process)
For claim 4 and 5 are a “classification method” that relies on data manipulation, comparing, and database selection based on thresholds, which closely resembles gathering and comparing information—a process considered to be an abstract idea [MPEP 2106.04(a)(2)].
Claim 6 and 16 recites
Producing gene sequence variation hazard information using the variation pattern prediction tool ….. gene loss-of-function index. (Mathematical concept or mental process)
Performing a null variant analysis, ….. a copy-number variation analysis, to obtain the variant type score. (Mathematical concept)
The claim 6 directed to "methods of organizing human activity" (analyzing and comparing data) and "mental processes" (evaluating variant types). Specifically, it describes analyzing gene sequence data to calculate a hazard score, which is a mathematical calculation or mental process, falling under MPEP 2106.04(a).
Claim 8 and 18 recites “when the null variant analysis, … are performed to evaluate the probability of loss of function intolerance, …. is greater than a predetermined threshold, …. are loss of function”. (Mathematical concept or human organizing activity) (Mental process) Evaluating the probability" of loss of function intolerance. These steps can be performed mentally or with basic pen-and-paper.
Claim 9 and 19 recites “the step of producing the clinical score …. comprises judging whether … then performing a dominant-recessive analysis, ….. combination thereof. (Mathematical calculation and Mental process and human organizing activity)
Claim 10 recites “the step of producing the functional score …. prediction tool based ….. comprises judging whether the mutation …. analysis with … to produce the functional score; …. performed to produce the functional score.” (Mathematical calculation and mental process) A method of analyzing genetic data using standard, well-known algorithmic components (SIFT, polymorphism analysis) and applying mathematical judgment (if X, then Y) based on variant types, constitutes an abstract idea.
Claim 10 additionally recites “the functional variant hazard prediction tool comprises a scale-invariant feature transform unit, a polymorphism phenotype analysis unit, and a site hazard prediction unit”(Mathematical concept, based on algorithm)
As such claims 1-20 recite an abstract idea (Step 2A, Prong 1: YES).
Step 2A, Prong 2
Claims found to recite a judicial exception under Step 2A, Prong 1 are then further analyzed to determine if the claims as a whole integrate the recited judicial exception into a practical application or not (Step 2A, Prong 2). Specifically, the claims recite the following additional elements:
Claim 1 recites
Receiving related information, …. person information and variant analysis. (Organizing human activity)
“Wherein the clinical database includes ClinVar database”.
Claim 4 recites “the population database comprises a genome aggregation database and a 1,000 genomes project database”.
Claim 7 recites “the classification method of claim 6, wherein the gene loss-of-function index is probability of loss of function intolerance”.
Claim 11 recites “comprising a computer processor and a memory.”
The additional elements recited in claims 1 and 4, namely, the use or incorporation of the ClinVar, genomeAD, and 1000 Genomes databases to received related information to determine a pathogenic score fail to provide a practical application. The core of the claimed invention remains the algorithmic determination of a pathogenic score. Because bioinformaticians and geneticists routinely rely on publicly available genomic databases (e.g., ClinVar, GenomeAD, and 1000 Genomes), utilizing these databases to filter benign variants from deleterious variants constitutes a routine, well-understood, and conventional process in the art.
For claim 7, The limitation “gene loss-of-function index is probability of loss of function intolerance” The claim does not show that the classification method provides a “specific improvement” to a technical process or that the formula is applied to a practical application. The claim simply uses this index to classify without providing a new, tangible, and practical use, so, it is "directed to" the abstract idea/natural phenomenon.
Regarding Claim 10, the additional elements recite various computational tools used to calculate a pathogenic score. However, these tools represent conventional bioinformatics pipelines operating on abstract mental or mathematical processes. Examples of such routine extra-solution activity include:
SIFT Analysis: A method for predicting whether an amino acid substitution affects protein function by comparing sequences to homologous proteins.
Polymorphism Phenotype Analysis: An assessment of the structural and functional consequences of a missense variant.
Site Hazard Prediction: The identification of splicing variants—mutations affecting how pre-mRNA is processed into mature mRNA.
These limitations describe mere data collection and the application of abstract mathematical and mental processes, and lack the necessary integrative steps to transform them into a practical application per MPEP 2106.
With respect to claim 11, the additional elements merely recite a conventional computer processor and a memory used to execute the claimed instructions. Under Alice Corp. v. CLS Bank Int'l, simply implementing or analyzing data using a generic, general-purpose computer does not transform an unpatentable abstract idea into a patent-eligible application.
Under the MPEP 2106.05(g) guidelines regarding insignificant extra-solution activity, the mere act of crunching data on a conventional computing system is well-understood, routine, and conventional in the art of bioinformatics pipelines. The recited limitations serve solely as data-gathering or analyzing activities. Because these additional elements do not reflect any specific improvement to computer functioning or physical technology, the claim fails to integrate the judicial exception into a practical application, and instead amounts to insignificant, routine post-solution activity.
There are no limitations that indicate that the pathogenic classifier requires anything other than a conventional computer to execute the instructions (a series of steps). As such, these limitations equate to mere instructions to implement the abstract idea on a generic computer that the courts have stated does not render an abstract idea eligible in Alice Corp., 573 U.S. at 223, 110 USPQ2d at 1983. See also 573 U.S. at 224, 110 USPQ2d at 1984.
The above recited additional elements do not provide a practical application of the recited judicial exception. As such, claims 1-20 are directed to an abstract idea (Step 2A, Prong 2: NO).
Step 2B
Claims found to be directed to a judicial exception are then further evaluated to determine if the claims recite an inventive concept that provides significantly more than the judicial exception itself (Step 2B). The claims do not include additional elements that are sufficient to amount to significantly more than the judicial exception because the claims recite additional elements that equate to mere instructions to apply the recited exception in a generic computing environment or well-understood, routine and conventional activity.
As discussed above, there are no additional limitations to indicate that the claimed classifier requires anything other than generic computer components in order to carry out the recited abstract idea in the claims. Claims that amount to nothing more than an instruction to apply the abstract idea using a generic computer do not render an abstract idea eligible. Alice Corp., 573 U.S. at 223, 110 USPQ2d at 1983. See also 573 U.S. at 224, 110 USPQ2d at 1984.
Furthermore, the additional elements recited in the claims amount to well-understood, routine and conventional activity. As such, the combination of additional elements recited in the claims is well-understood, routine and conventional.
The additional elements do not comprise an inventive concept when considered individually or as an ordered combination that transforms the claimed judicial exception into a patent-eligible application of the judicial exception. Therefore, the claims do not amount to significantly more than the judicial exception itself (Step 2B: No). As such, claims 1-20 are not patent eligible.
Claim Rejections - 35 USC § 103
In the event the determination of the status of the application as subject to AIA 35 U.S.C. 102 and 103 (or as subject to pre-AIA 35 U.S.C. 102 and 103) is incorrect, any correction of the statutory basis (i.e., changing from AIA to pre-AIA ) for the rejection will not be considered a new ground of rejection if the prior art relied upon, and the rationale supporting the rejection, would be the same under either status.
The following is a quotation of 35 U.S.C. 103 which forms the basis for all obviousness rejections set forth in this Office action:
A patent for a claimed invention may not be obtained, notwithstanding that the claimed invention is not identically disclosed as set forth in section 102, if the differences between the claimed invention and the prior art are such that the claimed invention as a whole would have been obvious before the effective filing date of the claimed invention to a person having ordinary skill in the art to which the claimed invention pertains. Patentability shall not be negated by the manner in which the invention was made.
The factual inquiries for establishing a background for determining obviousness under 35 U.S.C. 103 are summarized as follows:
1. Determining the scope and contents of the prior art.
2. Ascertaining the differences between the prior art and the claims at issue.
3. Resolving the level of ordinary skill in the pertinent art.
4. Considering objective evidence present in the application indicating obviousness or nonobviousness.
This application currently names joint inventors. In considering patentability of the claims the examiner presumes that the subject matter of the various claims was commonly owned as of the effective filing date of the claimed invention(s) absent any evidence to the contrary. Applicant is advised of the obligation under 37 CFR 1.56 to point out the inventor and effective filing dates of each claim that was not commonly owned as of the effective filing date of the later invention in order for the examiner to
Claim 1, 5, and 9, 11, 15, and 19 are rejected under 35 U.S.C. 103 as being unpatentable over Richards et al. (Genetics in medicine, Volume 17 | Number 5 | May 2015) in view of Laundrum et al. (Nucleic Acids Research, 2016, Vol. 44), Loanidis et al. (The American Journal of Human Genetics 99, 877–885, October 6, 2016), Kircher et al. (Nature Genetics, VOLUME 46 | NUMBER 3 | MARCH 2014), and further in view of Favalli et al. (The American Journal of Human Genetics 108, 682–695, April 1, 2021)
Regarding claim 1 and 11:
Richard et al. teaches collecting and evaluating variant information from:
Patient phenotype data (Figure 1; pg. 11, c1, middle; pg. 18, c1, bottom)
Familial segregation data (pg. 9, middle; Figure 1; pg. 11, c2, bottom; pg. 1, c2, bottom; pg. 8, bottom; pg. 9top; pg.13, c2 middle)
Population observations, population database (pg. 4, c2, top; Figure 1)
Unrelated affected individuals (Table 1, c2, middle; pg. 8 middle; pg. 9, c2, middle; pg. 13, c2, bottom)
Computational analysis. (Figure 1; pg. 5, c2, top; pg. 9, c1, bottom)
Population allele frequency databases as pathogenicity evidence under ACMG criteria BA1, BS1 and PM2 (Table 1; Table 6; pg. 9, top; pg. 10, c1, bottom)
Combining multiple weighted evidence categories to classify variants according to pathogenicity criteria. (pg. 8, bottom; pg. 9, middle)
Assigning pathogenic/likely pathogenic classifications corresponding to likelihoods of disease causation. (Abstract; pg. 3. c1, middle; pg. 7, c2, middle)
All of above teaching by Richard et al. reads to claim limitation “receiving related information ………. unrelated ……variant analysis” and “aggregating multiple pathogenicity evidence scores into a combined pathogenicity determination.” His teaching suggests integrating multiple categories of evidence for pathogenicity determination.
However, Richards et al. does not teach explicitly the use of population frequency information for pathogenicity estimation. He also does not teach use of ClinVar database as a repository of clinical interpretations of variants and summing the population score, the variant type score, the clinical score, and the functional score to produce a pathogenic score.
Kircher et al. further teaches generating pathogenicity metrics using population frequency information from reference databases (Abstract; pg. 1, c2, middle; pg. 4, c2, middle) and CADD functional deleteriousness scoring (Abstract) suggesting the limitation of “producing a variant ……. prediction tool”.
Richard et al. teaches an approach to variant classification using phenotype, familial segregation, and public databases. However, because the analysis is limited in throughput, a PHOSITA would look to computational tools to generate quantitative, scalable metrics of pathogenicity.
However, Kercher et al. does not teach explicitly the use ClinVar database as a repository of clinical interpretations of variants.
Landrum et al. teaches automated use of ClinVar database as a repository of clinical interpretations of variants and associated pathogenicity evidence (pg. 1, c2, bottom; Figure 1)
Landrum et al. does not explicitly teach calculation of the functional score, adding up all the scores generated and determining disease probability.
Favalli et al. teaches automated use of ClinVar classifications and associated evidence to generate pathogenicity likelihood scores. (Abstract; pg. 2, top, middle) Favalli et al. teaches assigning pathogenicity likelihood scores and reclassification probabilities based on combined evidence models. (Abstract; pg. 2, c1, middle; pg. 7, c2, middle) Favalli et al. additionally teaches automated ingestion of publicly available variant evidence datasets and patient associated variant information for automated pathogenicity determination. (Abstract; pg. 2, c1, top, middle)
Thus, the above art taught Favalli reads the claim limitation of “Producing a clinical score….. ClinVar database” and “determining probability ….. pathogenic score”
Favalli et al. teaches combining multiple evidence classes into a unified pathogenicity likelihood score using machine learning aggregation methods (Abstract; pg. 2, c1, top, middle)
Under BRI, “summing” reasonably encompasses mathematical aggregation, weighted combination, or algorithmic integration of multiple scores.
Ioannidis et al. teaches combining multiple functional tools into a unified pathogenicity framework. (Abstract; pg. 3, c1, top)
Thus, the above art by Favalli and Ioannidis reads the claim limitation of “Producing a functional score ….. hazard prediction tools” and “summing the population score, ……… a pathogenic score.”
One of ordinary skill in the art (PHOSITA) at the effective filing date would have been motivated to combine the teachings of Richards, Landrum, Ioannidis, Kircher, and Favalli for the following reasons:
Richards provides the foundational framework establishing that accurate pathogenicity assessment requires the synthesis of multiple, distinct evidence categories (e.g., computational, population, functional).
Landrum (ClinVar) teaches a centralized, standardized repository of clinical evidence. A PHOSITA would instantly recognize this as the ideal infrastructure to automate the gathering of the evidence categories required by Richards because the Richards (ACMG/AMP) framework requires combining dozens of distinct criteria (e.g., population frequency, functional data, and de novo observations). A PHOSITA understands that manually searching literature for these isolated data points across thousands of variants is impossible at scale. ClinVar naturally automates this process by storing these variables as structured, quarriable data.
Ioannidis and Kircher teach computational methods and algorithms for scoring variant pathogenicity. Kircher et al. introduced CADD (Combined Annotation Dependent Depletion), a widely adopted computational framework used in clinical pipelines to score and estimate the relative deleteriousness and pathogenicity of genetic variants. Ioannidis et al. developed REVEL (Rare Exome Variant Ensemble Learner). REVEL is a leading ensemble meta-predictor that integrates multiple individual computational algorithms (like CADD) to output a single score, providing an objective, automated line of evidence to classify disease-causing variants.
In clinical assessment, integrating these computational methods addresses the massive bottleneck of evaluating Variants of Uncertain Significance (VUS), as recognized by clinical guidelines (such as the ACMG/AMP criteria) which routinely incorporate computational evidence into variant classification.
Combining CADD scores from Kircher with Richard's ACMG classification scheme equips the system with an automated, standardized, and empirical filter for prioritizing variants.
Favalli utilizes machine learning models to automate the aggregation of clinical and computational evidence, generating a final pathogenicity likelihood score This continuous score streamlines the classification process, helping to accurately estimate the probability of a genetic mutation being a pathogenic variant.
Combining these known prior art sources and computational predictors to improve automated pathogenicity classification represents merely the predictable use of prior art elements according to their established functions. A PHOSITA facing the need to scale variant interpretation workflows would see an obvious benefit in combining Landrum’s data with the computational (Ioannidis/Kircher) and machine-learning (Favalli) automation techniques.
This combination yields entirely predictable results and is consistent with the rationales outlined in MPEP 2143.01 (combining known elements to yield predictable results) and KSR International Co. v. Teleflex Inc., 550 U.S. 398 (2007). Furthermore, the explicit use of ClinVar to automate pathogenicity workflows was an established, common-sense practice in the field of genomics before the effective filing date.
To make the filtering and prediction system fully effective, a skilled artisan would be motivated to cross-reference their automated scores against a known, public clearinghouse of clinical interpretations (ClinVar). Landrum provides the empirical validation and historical clinical context to complement the computational predictions of Kircher.
Regarding claim 5 and 15:
Richards et al. teaches that a variant (benign criterion BS2) is observed in a healthy adult individual. For recessive disorders, observing the variant in the homozygous (or compound heterozygous) state in a healthy adult provides supportive benign evidence. (pg. 13, top) He also teaches pathogenic criterion PM3 applies to recessive disorders and involves observing the variant in trans (on the opposite chromosome) with a pathogenic variant. However, observing a variant in the homozygous state (two copies of the exact same variant) inherently fulfills this concept and counts as moderate pathogenic evidence. (pg.10, bottom; pg. 14, top)
Thus, above reference suggesting the claim limitation of “continue to use the population database to perform the homozygous observational analysis.”
Regarding claim 9 and 19:
Richards et al. teaches interpretation according to dominant and recessive inheritance models, including: autosomal dominance inheritance. (table 5, table 6) and autosomal recessive inheritance. (table 5, table 6)
Thus, the Richards et al. teach dominant-recessive analysis. Maps to “performing a dominant … analysis”.
Claim 2 and 12 are rejected under 35 U.S.C. 103 as being unpatentable over Richards et al. in view of Laundrum et al., Ioannidis et al., Kircher et al., Favalli et al. as applied to claims 1, 5 and 9 above, and further in view of TaYoun et al. (Human Mutation. 2018; 39:1517–1524).
Richards et al. in view of Laundrum et al., Ioannidis et al., Kircher et al., and Favalli et al. as applied to claims 1, 5 and 9.
In addition to limitations taught by Richards et al. for claims 1, 5 and 9, Richards et al. also teaches incorporation of functional evidence into pathogenicity classification under ACMG criteria PS3 and BS3, including functional studies demonstrating damaging or benign effects (pg. 8, middle; pg. 9, c2 bottom; pg. 10, c1 top; pg. 9 top). He also teaches use of “functional studies” demonstrating deleterious effects of variants. (pg. 8, middle; pg. 9, c2, bottom; pg. 10, c1, top; pg. 9, top) suggesting the limitation of “related information further comprises loss-of-function test data, …. combination thereof.”
Richards et al. in view of Laundrum et al., Ioannidis et al., Kircher et al., and Favalli et al. do not explicitly teach incorporation of functional data such as “loss-of-function test data, protease kinetic …. combination thereof.”
Abou Tayoun et al. teaches interpretation of predicted and experimentally verified loss-of-function (LoF) variants, including null variants, nonsense variants, frameshift variants, splice variants, and functional evidence relevant to LoF determination (Abstract) suggesting the claim limitation “related information further comprises loss-of-function test data”.
Abou Tayoun et al. further teaches gene-specific application of loss of function interpretation rules depending on gene-disease relationships. (Abstract; pg. 12, bottom; pg. 13, top)
Which suggests the limitation “special target disease or gene selection”
It would have been obvious to a person having ordinary skill in the art (PHOSITA) as of the effective filing date to combine the teachings of Richard et al., and Abou Tayoun et al.
This combination is supported by the following rationales:
The American College of Medical Genetics and Genomics (ACMG) guidelines (Richard et al.) alongside with Abou Tayoun et al. inherently prescribed integrating functional and disease-specific data into variant interpretation frameworks. Adapting and incorporating loss-of-function data and biochemical/kinetic assays into automated systems is simply the predictable application of known prior-art elements according to their established functions.
Because Abou Tayoun et al. publications discuss ACMG workflows to include gene- and disease-specific interpretation logic, standardizing these rules within automated systems reflects routine optimization. Combining these biologically relevant evidence types predictably improved variant classification reliability and accuracy, representing design incentives well within the capability of a PHOSITA.
Such integration represents the predictable use of known prior-art elements for their established purpose, yielding nothing more than predictable results consistent with KSR Int’l Co. v. Teleflex Inc., 550 U.S. 398 (2007).
Each of the elements in claim was well known in the prior art, and their integration into automated pathogenicity classification systems existing prior to the effective filing date would have been obvious with a reasonable expectation of success.
Claim 3 and 13 are rejected under 35 U.S.C. 103 as being unpatentable over Richards et al. in view of Laundrum et al., Ioannidis et al., Kircher et al., Favalli et al. as applied to claims 1, 5 and 9 above, and further in view of Amendola et al. (The American Journal of Human Genetics 98, 1067–1076, June 2, 2016)
Richards et al. in view of Laundrum et al., Ioannidis et al., Kircher et al., and Favalli et al. as applied to claims 1, 5 and 9.
In addition to limitations taught by Richards et al. for claims 1, 5, 9, and 2, he also teaches evaluating variant allele frequency in population databases including Exome sequencing project, 1000 Genome s, and gnomeAD for pathogenicity assessment under ACMG criteria BA1. BS1, and PM2. (pg. 8, middle; pg. 9, top). He also teaches ACMG criterion BS2, which evaluates whether a variant is observed in healthy individuals in homozygous or heterozygous state. (pg. 9, top, pg. 10, c2, bottom)
and combining multiple ACMG evidence criteria into an overall pathogenicity determination. (Table 5)
Thus, the above cited art teaches performing homozygous observational analysis using population database to derive pathogenicity evidence, which under BRI reasonably constitutes a “second population score” which reads to limitation of “performing a homozygous ….. a second population score”
Richards et al. in view of Laundrum et al., Ioannidis et al., Kircher et al., Favalli et al. does not teach explicitly “performing a frequency variation analysis on the variant sequence information using the population database to generate a first population score.”
Amendola et al. teaches use of minor allele frequency thresholds to classify pathogenic versus benign variants. (pg. 2, c1, middle) Under BRI, determining allele frequency and assigning corresponding ACMG evidence weights reasonably constitutes generating a “first population score” which reads to the limitation of “performing a frequency variation ……………… a first population score”.
It would have been obvious to a person having ordinary skill in the art (PHOSITA) at the effective filing date to combine the teachings of Richards et al., Laundrum et al., Ioannidis et al., Kircher et al., Favalli et al and Amendola et al. for the following reasons:
The prevailing ACMG/AMP guidelines established an industry-wide mandate to integrate multiple categories of population evidence into variant pathogenicity assessments.
Using population allele frequency analysis and homozygous occurrence analysis concurrently was an art-recognized, standard approach to distinguish benign from pathogenic variants. A PHOSITA would routinely combine these metrics for a more comprehensive assessment.
Automated variant interpretation systems already existed and were designed to seamlessly aggregate multiple ACMG criteria (including frequency and homozygosity) into unified, standardized scoring systems.
Combining multiple population-derived metrics yields a predictable and expected improvement in classification accuracy, consistency, and reliability.
Pursuant to the principles set forth in KSR International Co. v. Teleflex Inc., 550 U.S. 398 (2007) and outlined in MPEP 2143.01, this combination represents nothing more than the predictable use of known prior-art elements according to their established functions to achieve expected results.
Therefore, the cited references, viewed collectively, teach or render obvious, specifically:
Performing allele frequency analysis using population databases;
Performing homozygous occurrence analysis using population databases; and
Combining those individual analyses into overall population evidence score.
Claim 4 and 14 are rejected as being unpatentable under 35 U.S.C 103 over Richards et al. in view of Laundrum et al., Ioannidis et al., Kircher et al., Favalli et al., and Amendola et al. as applied to claims 1, 5, 9, and 3 above, in view of Gudmundsson et al. (Human Mutation. 2022; 43:1012–1030)
Richards et al. in view of Laundrum et al., Ioannidis et al., Kircher et al., Favalli et al. and Amendola et al. as applied to claims 1, 5, 9 and 3.
In addition to limitations taught by Richards et al. for claims 1, 5, and 9, and 3, Richards et al. also teaches use of multiple population database including: Exome sequencing project, Exome Aggregation consortium (ExAC) and 1000 Genome project for pathogenicity analysis under ACMG criteria BA1, BS1, and PM2. (pg. 10, c1, middle; Table 1, 3, 4). He also discloses applying pathogenicity evidence rules based on allele frequency thresholds including BA1, BS1, and PM2. (pg.10, c1, middle, c2, bottom, Table 6) Thus, the Richard et al. teach the claim limitation of “use of both genomeAD and 1000 Genome project database.”
Richards et al. in view of Laundrum et al., Ioannidis et al., Kircher et al., Favalli et al. and Amendola et al. does not teach “performing a frequency variation analysis on the variant sequence information using the population database to generate a first population score.”
Gudmundsson et al. teaches the genomeAD as a large-scale aggregated population database used for variant interpretation workflows (Abstract, Figure 3) and using genomeAD alleles counts and allele frequencies as principal evidence when sufficient population are available, including use of variant occurrence counts and constrains metrics. (Abstract, Figure 3) He also teaches that different population database may provide complementary information and that variant interpretation workflows routinely use multiple databases to compensate for insufficient observations in any source. (pg. 6, c2, bottom, pg.8, c2, bottom)
Under BRI, using a primary database when allele observations exceed a threshold reasonably reads the claimed limitation “continue to use the genome aggregation database.”
Also selecting a secondary database when observation in a primary database is insufficient reasonably reads to the claimed fallback use of 1000 Genome database. “when the mutation ….. 1000 Genome Project database”.
A PHOSITA would have been motivated to use multiple population databases to evaluate genetic variants based on the following rationales:
Selecting among databases based on observation count and frequency sufficiency represents predictable data-quality management and routine optimization.
It was well known to prioritize large aggregated databases (e.g., ExAC, gnomAD) for superior statistical power when observations were sufficient, and to routinely use secondary databases (e.g., 1000 Genomes) to supplement or confirm rarity analysis when observations were sparse.
Threshold-based evidence application was fundamental to ACMG criteria (e.g., BA1, BS1, PM2), requiring comprehensive allele-frequency analysis across databases to improve reliability.
This combination represents the predictable use of known prior-art elements according to their established functions, aligning with the teachings of KSR Int’l Co. v. Teleflex Inc., 550 U.S. 398 (2007). Therefore, the cited references collectively teach or render obvious the limitations of claim 4, specifically:
The use of genome aggregation databases and the 1000 Genomes project database;
Prioritizing one database when sufficient allele observations exist; and
Selecting/switching to another database when allele observations are insufficient.
Claim 6 and 16 are rejected as being obvious over Richards et al. in view of Laundrum et al., Ioannidis et al., Kircher et al., Favalli et al., Abou Tayoun et al. as applied to claims 1, 5, 9 and 2 above, and further in view of Kai Wang et al. (Nucleic Acids Research, 2010, Vol. 38, No. 16) and William McLaren et al. (Genome Biology (2016) 17:122)
Richards et al. in view of Laundrum et al., Ioannidis et al., Kircher et al., Favalli et al. and Abou Tayoun et al. as applied to claims 1, 5, 9 and 2.
Richards et al., Laundrum et al., Ioannidis et al., Kircher et al., Favalli et al. and Abou Tayoun et al. does not teach to create a more efficient and robust variant hazard evaluation system using various types of genetic analysis.
In addition to limitations taught by Kircher eta al., Ioannidis et al. for claims 1, 5, and 9 rejections, Kircher et al. also teaches CADD, a computational framework generating deleteriousness scores and pathogenicity predictions for genomic variants. (Abstract; pg. 1, c2, bottom)
Ioannidis et al. further teaches REVEL, an ensemble variant pathogenicity predictor generating hazard/pathogenicity scores for sequence variants. (Abstract)
Kai Wang et al. (ANNOVAR) and William McLaren et al. (Ensembl VEP) both teach computational tools that generate functional consequence annotations and pathogenicity-related information from genomic variants. (Abstracts)
Under BRI, these above references reasonably map the claim limitation “variation pattern prediction tools” producing “gene sequence variation hazard information”
Kai Wang et al. also teaches annotation of the following: Nonsense variants, Splice variants, missense variants, synonymous variants, intronic variants, UTR variants, frameshift variants, nonframeshift indels, CNVs. (pg. 3, c1, middle; pg. 3, c2, top; Figure 1)
In addition to limitations taught by Abou Tayoun et al. for claim 2 rejections, he also teaches LoF interpretation frameworks including PVS1 analysis for: Nonsense variants, canonical splice variants, initiation codon variants, exon deletions, frameshift variants. (pg. 2, c1, c2)
Under BRI, ACMG PVS1 strength assignment and LoF annotation reasonably maps the claim limitation “gene loss of function index”.
Richard et al. and Abou Tayoun et al. also teach analysis of null variants including nonsense and frameshift variants under ACMG PVS1 criteria (pg. 3, c2, bottom and pg. 2, c1, bottom respectively) and Abou Tayoun et al. further teaches splice site LoF evaluation under ACMG PVS1. (pg. 2, middle) suggesting the limitations of “performing a null variant analysis.” and “splice variant analysis” respectively.
Ioannidis et al. teaches missense pathogenicity prediction using REVEL. (Abstract) Which suggests the limitation of “missense variant analysis”.
Kai Wang et al. classify non-frameshift insertion/deletion variants (Figure 1) suggesting the limitation of “in-frame indels variant analysis”.
Abou Tayoun et al. further discuses initiation codon variants and start loss interpretation within LoF analysis frameworks (pg. 5, c1, top) which suggests the limitation of “start loss variant analysis”.
Kai Wang et al. annotate synonymous/silent variants (Figure 1) suggests the limitation of “silent variant analysis”. Kai Wang et al. teach annotation and classification of intronic variants (pg. 3, c1, middle) which suggests limitation of “intronic variant analysis”.
Kircher et al. teaches pathogenicity scoring for noncoding variants including promoter and regulatory-regions variants. (pg. 3, c1, top)
Also, Kai Wang et al. teaches annotation of UTR and regulatory-region variants. Thus, above cited art reads to limitation of “non-coding variant analysis in UTR or promoter”. Kai Wang et al. also teach computational pathogenicity interpretation of copy number variants (pg. 1, c2, bottom) suggesting the limitation of “copy number variation analysis”.
A person of ordinary skill in the art (PHOSITA) at the time of the effective filing date would have found the claimed combination of Richards et al., Kai Wang et al., William McLaren et al., Kircher et al., Ioannidis et al., and Abou Tayoun et al. obvious. This combination falls squarely under the KSR rationale of combining familiar elements according to known methods to yield predictable results.
The obviousness is supported by the following findings:
The cited references represent established, well-known bioinformatics practices. Annotation tools (such as ANNOVAR and VEP) were already known to categorize variants, while computational predictors (such as those taught by Kircher et al. and Ioannidis et al.) independently generated deleteriousness and Loss-of-Function (LoF) metrics. Combining these known functions into a unified scoring framework yields predictable results without any unexpected outcomes.
Integrating standardized evaluation rules (e.g., ACMG variant interpretation guidelines and the evaluation of diverse variant classes as taught by Richards et al., Kai Wang et al. and Abou Tayoun et al. into a comprehensive clinical genomics pipeline constitutes a basic design choice. A PHOSITA applying common sense would view this as a straightforward step to create a more efficient and robust variant evaluation system.
The combined elements perform only the functions they were known to perform in the prior art. Because each component retains its established function, the resulting framework provides anticipated and predictable outputs to a PHOSITA.
Claim 7, 8, 17, and 18 are rejected as being unpatentable under 35 U.S.C. 103 over Richards et al. in view of Laundrum et al., Ioannidis et al., Kircher et al., Favalli et al., Abou Tayoun et al., Kai Wang et al. and William McLaren et al. as applied to claims 1, 5, 9, 2 and 6 above, and further in view of LEK et al. (2 8 8 | N ATU R E | VO L 5 3 6 | 1 8 Au g u s t 2 0 1 6).
Regarding claim 7 and 17:
Richards et al. in view of Laundrum et al., Ioannidis et al., Kircher et al., Favalli et al., Abou Tayoun et al., Kai Wang et al. and William McLaren et al. as applied to claims 1, 5, 9, 2 and 6.
Richard et al., Abou Tayoun et al., Kai Wang et al., and William McLaren et al. does not teach "the gene loss-of-function index is the probability of loss of function intolerance."
Lek et al. teaches the “probability of being loss of -function intolerant (pLI)…” metric developed using ExAC population data. (Abstract; Figure 3; pg.4, c2) Lek et al. further teaches that pLI is a gene-level intolerance metric used to identify genes intolerant to truncating variation and useful for clinical interpretation. (pg. 2, c1, middle; pg. 4, c2, bottom; pg. 6, c1, bottom)
Thus, Lek at al. also discloses a “probability of loss of function intolerance” metric corresponding directly to the claimed limitation “gene loss of function index”
Accordingly, the cited references (for claim 1, 5, 9, 2, and 6 and Lek et al.) collectively teach incorporation of pLI metrics into automated pathogenicity systems and suggests the claim limitation of “use of pLI in variant interpretation”.
A person of ordinary skill in the art (PHOSITA) would have been motivated to combine these teachings prior to the effective filing date. The proposed use of the probability of loss-of-function intolerance (pLI) in a gene loss-of-function index constitutes obviousness under 35 U.S.C. 103, as outlined in MPEP 2143.01 and KSR Int'l Co. v. Teleflex Inc.
The rejection is based on the following rationales:
Incorporating the pLI metric into a gene loss-of-function index represents the simple swapping of one known LoF intolerance metric for another within an established framework. It is well-settled that replacing one familiar, known element with another to obtain predictable results does not render an invention patentable.
A PHOSITA would have been motivated to make this substitution based on the known, established utility of pLI for successfully prioritizing pathogenic LoF variants. Because pLI, like other constraint metrics, was already utilized in scoring pipelines, combining them yields highly predictable outputs consistent with conventional knowledge.
The claimed system merely reflects the predictable use of prior art elements according to their known functions. The cited references collectively render the combination obvious, as the elements act together in the same way they operate separately.
Regarding claim 8 and 18:
Tayoun et al. also teaches evaluation of Nonsense/null variants, canonical splice variants, initiation codon/start-loss variants for determining applicability of ACMG PVS1 loss-of-function evidence. (pg. 2, c2, bottom; pg.5, c1, middle; pg. 5, c2, top) The reference explains that these variant classes are evaluated in connection with whether the affected gene is intolerant to loss-of-function variation and suggesting the limitation of “when the null variant …. Loss of function intolerance”
Lek et al. teaches pLI as a metric quantifying whether genes are intolerant to protein truncating/loss-of-function variants. (Abstract; Figure 3; pg.4, c2)
Thus, the cited references collectively teach “evaluating null, splice, and start-loss variants to determine lOF intolerance probability.”
Lek et al. also teaches threshold based pLI interpretation, including LoF intolerant (pLI greater than or equal to 0.9) genes. (Figure 3; pg. 4, c2) Lek et al. further teaches that genes with high pLI scores are highly intolerant to LoF variants and are enriched for severe disease associated genes. (pg. 4, c2)
Under BRI, assigning “high risk” when pLI exceeds a threshold reasonably encompasses automated pathogenicity classification using known pLI cutoff values. And it maps to the claim limitation of “if the probability of loss ….. are loss of function”. In addition, Lek et al. teaches low-pLI genes as LoF tolerant genes: “LoF tolerant (pLI is lesser than or equal to 0.1).”
Under BRI, automatically classifying variants in LoF-tolerant genes as lower risk reasonably encompasses the claimed limitation of “if the probability ….. that a risk is low.”
Claim 10 and 20 are rejected as being unpatentable under 35 U.S.C. 103 over Richards et al. in view of Laundrum et al., Ioannidis et al., Kircher et al., Favalli et al.., Kai Wang et al., and William McLaren et al., as applied to claims 1, 5, 9, and 6 and further in view of Henikoff et al. (Nucleic Acids Research, 2012, Vol. 40), Adzhubei et al. (Curr Protoc Hum Genet. 2013 January; 0 7: Unit7.20), Jaganathan et al. (2019, Cell 176, 535–548) and Xiaoming Liu et al. (Genome Medicine (2020) 12:103)
Richards et al. in view of Laundrum et al., Ioannidis et al., Kircher et al., Favalli et al.., Kai Wang et al., and William McLaren et al., as applied to claims 1, 5, 9, and 6.
As describe before, Kai Wang et al. (ANNOVAR) teaches classifying variants by consequence type, including: missense variants, splice-site variants, synonymous variants, frameshift variants and Intronic variants. Richards et al. similarly teaches evaluating different ACMG-evidence criteria depending upon mutation class.
Richards et al. in view of Laundrum et al., Ioannidis et al., Kircher et al., Favalli et al.., Kai Wang et al., and William McLaren et al., does not teach “functional variant hazard prediction tool comprises a scale-invariant feature transform unit …. hazard prediction unit.” And generation of functional score of variants based on the judgment of whether the mutant is missense or splice site variant.
Henikoff et al. disclose the SIFT algorithm (“Sorting intolerant from tolerant”) for predicting whether amino acid substitutions affect protein function. (Abstract; pg. 1, c2, middle) The acronym “SIFT” directly corresponds to “scale invariant feature transform” terminology recited in the claim under BRI. SIFT produces deleteriousness scores for missense variants and widely incorporated into clinical variant interpretation pipelines. Accordingly, SIFT teaches the claimed “functional variant hazard ….. scale-invariant feature transform unit.”
Adzhubei et al. expressly disclose Polyphen-2(“Polymorphism Phenotyping v2”), a computational tool for predicting damaging effects of missense mutations. Polyphen directly corresponds to the claimed “polymorphism phenotype analysis unit.” Polyphen-2 produces pathogenicity probabilities and functional impact predictions for missense variants. (Abstract) Thus, Polyphen-2 teaches the claimed limitation “a polymorphism phenotype analysis unit.”
Jaganathan et al. disclose SpliceAI, a deep learning-based splice sites prediction tool for evaluating splice-altering variants. SpliceAI predicts splice donor/acceptor disruption probabilities and pathogenic splice consequences. (Abstract) Under BRI, splice-site pathogenicity prediction systems reasonably maps to the limitation “site hazard prediction unit.” Jaganathan et al. also teaches using SpliceAI to evaluate splice variants and generate splice disruption probabilities. (Abstract) Richards et al. teach applying splice specific ACMG evidence criteria for splice variants. Thus, the cited references teach using splice-specific hazard prediction tools when splice variants are detected. And maps to the claim limitation of “when the mutation ….. is performed.”
Xiaoming Liu et al. (dbNSFP) additionally aggregates multiple missense and splice prediction annotations. (academic.oup.com) Thus, the prior art teaches “determining whether a mutation is missense or splice related.” Xiaoming Liu et al. (dbNSFP) aggregates outputs from both: SIFT and Polyphen-2 for missense pathogenicity evaluation. (pg. 4, c1) Kercher et al. further teach combining multiple functional prediction tools into integrated deleteriousness scoring systems. (Abstract) Accordingly, the cited references teach both SIFT and Polyphen-2 together for missense variant scoring which maps to the claim limitation “when the mutation sites ….. performed.”
SIFT, Polyphen-2, SplicAI, and CADD each generate numerical deleteriousness/pathogenicity scores. Under BRI, these outputs reasonably constitute “functional score”
One of ordinary skill in the art (PHOSITA) before the effective filing date (EFD) would have been motivated to combine the teachings of Henikoff et al. (SIFT), Adzhubei et al. (Polyphen-2), Jaganathan et al. (SplicAI), Richards et al., Kercher et al. (CADD), Kai Wang et al. (ANNOVAR), and Xiaoming Liu et al. (dbSNP) to arrive at the claimed invention of claim 10. The proposed combination relies on multiple accepted obviousness rationales outlined in MPEP 2143:
SIFT and Polyphen-2 were known, specialized tools routinely utilized together in the art to evaluate missense variants. SplicAI was a standard, known tool used to evaluate splice-site hazards. Incorporating these established tools into an integrated pipeline represented the predictable use of prior-art elements, with each tool performing its standard, intended function. (MPEP 2143.01 I)
Variant annotation frameworks such as ANNOVAR taught the classification of variants by consequence type prior to processing. Because integrated pipelines commonly routed missense variants to missense predictors and splice variants to splice predictors, combining these distinct, specialized tools was a predictable optimization of automated pathogenicity classification systems. (MPEP 2143.01 V)
A PHOSITA would have had a reasonable expectation of success in combining these tools, as each reference functions as designed without interfering with the predictive capabilities of the others. Thus, the combination constitutes nothing more than predictable variations of prior art elements, rendering the system obvious under KSR Int'l Co. v. Teleflex Inc., 550 U.S. 398, 416 (2007).
Conclusion
No claims are allowed.
Any inquiry concerning this communication or earlier communications from the examiner should be directed to ARSHAD KHAN whose telephone number is (571)272-9812. The examiner can normally be reached Mon-Fri-7:30-5:00 PM.
Examiner interviews are available via telephone, in-person, and video conferencing using a USPTO supplied web-based collaboration tool. To schedule an interview, applicant is encouraged to use the USPTO Automated Interview Request (AIR) at http://www.uspto.gov/interviewpractice.
If attempts to reach the examiner by telephone are unsuccessful, the examiner’s supervisor, Larry Riggs can be reached at 5712703062. 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.
/A.H.K./Examiner, Art Unit 1686
/LARRY D RIGGS II/Supervisory Patent Examiner, Art Unit 1686
Prosecution Insights