DETAILED ACTION
Applicant’s response, filed 23 April 2025, has been fully considered. The following rejections and/or objections are either reiterated or newly applied. They constitute the complete set presently being applied to the instant application.
Notice of Pre-AIA or AIA Status
The present application, filed on or after March 16, 2013, is being examined under the first inventor to file provisions of the AIA .
Continued Examination Under 37 CFR 1.114
A request for continued examination under 37 CFR 1.114, including the fee set forth in 37 CFR 1.17(e), was filed in this application after final rejection. Since this application is eligible for continued examination under 37 CFR 1.114, and the fee set forth in 37 CFR 1.17(e) has been timely paid, the finality of the previous Office action has been withdrawn pursuant to 37 CFR 1.114. Applicant's submission filed on 23 April 2025 has been entered.
Claim Status
Claims 1, 3-8, 10-18, 21-23, and 55-60 are pending and examined herein.
Claims 1, 3-8, 10-18, 21-23, and 55-60 are rejected.
Claims 1, 3, and 4 are objected to.
Priority
Claims 1, 3-8, 10-18, 21-23, and 55-60 are granted the claim to the benefit of priority to U.S. Provisional application 62/699135 filed 17 July 2018. Thus, the effective filling date of claims 1, 3-8, 10-18, 21-23, and 55-60 is 17 July 2018.
Drawings
The drawings received 14 December 2020 are objected to as noted below.
The drawings are objected to as failing to comply with 37 CFR 1.84(p)(5) because they include the following reference character(s) not mentioned in the description: 13 (in claim 3). Corrected drawing sheets in compliance with 37 CFR 1.121(d), or amendment to the specification to add the reference character(s) in the description in compliance with 37 CFR 1.121(b) are required in reply to the Office action to avoid abandonment of the application. Any amended replacement drawing sheet should include all of the figures appearing on the immediate prior version of the sheet, even if only one figure is being amended. Each drawing sheet submitted after the filing date of an application must be labeled in the top margin as either “Replacement Sheet” or “New Sheet” pursuant to 37 CFR 1.121(d). If the changes are not accepted by the examiner, the applicant will be notified and informed of any required corrective action in the next Office action. The objection to the drawings will not be held in abeyance.
The drawings are objected to as failing to comply with 37 CFR 1.84(p)(4) because reference character “1”, “15”, and “16” has been used to designate both cells from an embryo and a plasma sample from a pregnant mother (1 in figure 1), validation statistics (15 in figure 3) and input to a neural network (15 in figure 5), network weights (16 in figure 3) and layers of a neural network (16 in figure 5 and figure 6). Corrected drawing sheets in compliance with 37 CFR 1.121(d) are required in reply to the Office action to avoid abandonment of the application. Any amended replacement drawing sheet should include all of the figures appearing on the immediate prior version of the sheet, even if only one figure is being amended. Each drawing sheet submitted after the filing date of an application must be labeled in the top margin as either “Replacement Sheet” or “New Sheet” pursuant to 37 CFR 1.121(d). If the changes are not accepted by the examiner, the applicant will be notified and informed of any required corrective action in the next Office action. The objection to the drawings will not be held in abeyance.
Claim Objections
The objection of claim 55 in Office action mailed 06 November 2024 is withdrawn in view of the amendment received 23 April 2025.
Claims 1, 3, and 4 are objected to because of the following informalities: Claims 1, 3, and 4 recite “a plurality of SNV loci” but should read “a plurality of single nucleotide variant (SNV) loci”. Appropriate correction is required.
Claim Rejections - 35 USC § 112
The rejection on the ground of 112/a new matter of claim 5 in Office action mailed 06 November 2024 is withdrawn in view of the amendment of “wherein the neural network outputs a ploidy state value, and wherein the neural network comprises a plurality of layers and a plurality of weights” received 23 April 2025.
112/a
The following is a quotation of the first paragraph of 35 U.S.C. 112(a):
(a) IN GENERAL.—The specification shall contain a written description of the invention, and of the manner and process of making and using it, in such full, clear, concise, and exact terms as to enable any person skilled in the art to which it pertains, or with which it is most nearly connected, to make and use the same, and shall set forth the best mode contemplated by the inventor or joint inventor of carrying out the invention.
The following is a quotation of the first paragraph of pre-AIA 35 U.S.C. 112:
The specification shall contain a written description of the invention, and of the manner and process of making and using it, in such full, clear, concise, and exact terms as to enable any person skilled in the art to which it pertains, or with which it is most nearly connected, to make and use the same, and shall set forth the best mode contemplated by the inventor of carrying out his invention.
Claims 1, 3-8, 10-18, 21-23, and 55-60 are rejected under 35 U.S.C. 112(a) or 35 U.S.C. 112 (pre-AIA ), first paragraph, as failing to comply with the written description requirement. The claim(s) contains subject matter which was not described in the specification in such a way as to reasonably convey to one skilled in the relevant art that the inventor or a joint inventor, or for applications subject to pre-AIA 35 U.S.C. 112, the inventor(s), at the time the application was filed, had possession of the claimed invention.
The rejection below is newly recited.
Claims 1, 3, and 4 recite “generating an updated machine learning model… wherein at least one of generating the first training data or generating the second training data comprises augmenting the first training data or the second training data, respectively, with one or more synthetic cases that are based on one or more of the first plurality of cases or the second plurality of cases, respectively”. The MPEP states “The written description requirement for a claimed genus may be satisfied through sufficient description of a representative number of species. A "representative number of species" means that the species which are adequately described are representative of the entire genus. Thus, when there is substantial variation within the genus, one must describe a sufficient variety of species to reflect the variation within the genus” (MPEP 2163.05(I)(B)). There is not an adequate written description for the generically recited process of generating an updated machine learning model using generically recited training data which encompasses all possible training data which includes all possible synthetic data based on the generically recited training data to be used for training a machine learning model. The instant disclosure only provides that the training data is genetic data but does not provide an adequate written description on the genus of generic training data or a representative number of species of generic training data used in generating an updated model. Dependent claims 5-8, 10-18, 21-23, and 55-60 are rejected by virtue of their dependency on a rejected claim without alleviating the rejection.
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.
Claim 6, 8, 10-18, and 21-23 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 rejections below are newly recited.
Claims 6, 11, 13, 15, and 18 recites “the synthetic case” which renders the metes and bounds of the claim indefinite. The indefiniteness arises because it is unclear if “the synthetic case” is referring to one of the “one or more synthetic cases” in claim 1 (also which synthetic case in the one or more synthetic cases is this referring to in claim 1) or to the “synthetic case” in claim 6. Dependent claims 10, 12, 14, 16, 17, and 21-23 are rejected by virtue of their dependency on a rejected claim without alleviating the indefiniteness. For the sake of furthering examination, “the synthetic case” of these claims will be interpreted as referring to the “synthetic case” in claim 6.
Claim 8 recites the limitation "the amplification products are sequenced with a depth of…" in line 1 of the claim. There is insufficient antecedent basis for this limitation in the claim. The indefiniteness arises because the claim does not make clear what “the amplification products” are. For the sake of furthering examination, this limitation will be interpreted as wherein the sequencing data was sequenced with a depth of read of at least 200.
Claim 11 recites “the one or more of the plurality of cases” and claim 15 recites “one of the cases of the plurality of cases” which renders the metes and bounds of the claim indefinite. The indefiniteness arises because it is unclear if “the plurality of cases” is referring to the “first plurality of cases” or “second plurality of cases” in claim 1 or the “plurality of cases” in claim 6. For the sake of furthering examination, “the one or more of the plurality of cases” will be interpreted as referring to the “plurality of cases” in claim 6.
Claim 16 recites “the genetic sequencing data or genetic array data is obtained using a biological sample that comprises a plasma sample…” which renders metes and bounds of the claim indefinite. The indefiniteness arises because it is unclear if “the genetic sequencing data or genetic array data” is referring to the “genetic sequencing data or genetic array data” in claim 1 or the “genetic sequencing data or genetic array data” in claim 6. For the sake of furthering examination, “the genetic sequencing data or genetic array data” will be interpreted as referring to the “genetic sequencing data or genetic array data” in claim 6.
Claim 21 recites “wherein the modifying” which renders the metes and bounds of the claim indefinite. The indefiniteness arises because it is unclear if “the modifying” is referring to the process of “iteratively modifying the neural network” or the step of “modifying one or more of the plurality of weights” in claim 6. Dependent claim 22 is rejected by virtue of its dependency on a rejected claim without alleviating the indefiniteness. For the sake of furthering examination, “wherein the modifying” will be interpreted as referring to the process of “iteratively modifying the neural network” in claim 6.
Claim Rejections - 35 USC § 101
35 U.S.C. 101 reads as follows:
Whoever invents or discovers any new and useful process, machine, manufacture, or composition of matter, or any new and useful improvement thereof, may obtain a patent therefor, subject to the conditions and requirements of this title.
Claims 1, 3-8, 10-18, 21-23, and 55-60 are rejected under 35 U.S.C. 101 because the claimed invention is directed to a judicial exception (i.e., a law of nature, a natural phenomenon, or an abstract idea) without significantly more.
The rejection below was previously recited but has been modified to address claim amendments.
(Step 1)
Claims 1, 3-8, 10-18, 21-23 and 55-60 fall under the statutory category of a process.
(Step 2A Prong 1)
Under the BRI, the instant claims recite judicial exceptions that are an abstract idea of the type that is in the grouping of a “mental process”, such as procedures for evaluating, analyzing or organizing information, and forming judgement or an opinion. The instant claims further recite judicial exceptions that are an abstract idea of the type that is in the grouping of a “mathematical concept”, such as mathematical relationships and mathematical equations.
Independent claims 1, 3, and 4 recite mental processes of “generating first training data…”, “generate second training data…”, and “wherein at least one of generating the first training data or generating the second training data comprises augmenting…”.
Independent claims 1, 3, and 4 recite mathematical concepts of “training a first machine learning model using the generated first training data” and “updating the first machine learning model using the generated second training data to obtain the updated machine learning model”.
Independent claim 1 further recites a mental process and mathematical concept of providing genetic sequencing data or genetic array data of a plurality of SNV loci to obtain as output classification information indicative of a ploidy state of the fetal chromosome.
Independent claim 3 further recites a mental process and mathematical concept of providing genetic sequencing data or genetic array data of a plurality of SNV loci corresponding to tumor DNA to obtain as output classification information to identify amplified tumor DNA.
Independent claim 4 further recites a mental process and mathematical concept of providing genetic sequencing data or genetic array data of a plurality of SNV loci to obtain as output classification information to identify and quantify amplified transplant donor DNA.
Dependent claim 6 recites mathematical concepts of “determining, for a training sample, genetic sequencing data or genetic array data for a plurality of genetic positions, “determining true state values for a plurality of genetic segments…”,“determining a neural network comprising one or more layers for calling state values…” “iteratively modifying the neural network until an exit condition is satisfied…”, “determining a batch of data comprising a plurality of cases…”, “generating a synthetic case based on…”, “including the synthetic case in the batch…”, “augmenting the true state values…”, generate a output comprising one or more state values for each case, and “modifying one or more of the plurality of weights based on the network output”. Dependent claim 11 recites a mental process by “the synthetic case to include a segment that is a homolog…” and “generating the homolog”. Dependent claim 13 recites mathematical concepts of “using the unphased genotypes to generate statistics and using the statistics to generate the synthetic case”. Dependent claim 15 recites a mental process of “generating the synthetic case to comprise simulating a chromosomal microdeletion…”. Dependent claim 16 recites a mathematical concept of training the neural network to determine the ploidy state of a region with a microdeletion. Dependent claim 17 recites mental process of predict the occurrence of a specific microdeletion in the fetus of the pregnant mother using sequencing data. Dependent claim 18 recites a mental process of “generating a plurality of synthetic cases, including the synthetic case, by simulating a chromosomal microdeletion…”. Dependent claim 21 recites a mathematical concept of “perturbing the batch of data…”. Dependent claim 22 recites mathematical concept of “perturbing a plurality of array reads for single nucleotide polymorphisms by multiplying the array reads by respective scalars”. Dependent claim 55 recites mental processes of “selecting a first case from the training data…”, “selecting a segment within an aneuploidy region…”. Dependent claim 55 recites a mathematical concept of “the first case and the altered first case are used for training or updating of the first machine learning model”.
The claims recite a process of generating training data, augmenting training data with synthetic cases, analyzing sequencing data to assess a fetal ploidy state of a chromosome, analyze sequencing data to identify tumor DNA, analyze sequencing data to identify and quantify donor derived DNA. The human mind is capable of analyzing genetic data to provide classification information, generating training data, augmenting data with synthetic data. The claims further recite mathematical concepts as training and updating a machine learning model (the instant specification provides that training utilizes stochastic gradient descent-like optimization which is a mathematical process which is a series of mathematical calculations for updating neural network parameters). Dependent claims 7, 8, 10, 23, and 56-60 further limit the mental process/mathematical concept recited in the independent claim but do not change their nature as a mental process/mathematical concept.
(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). Integration into a practical application is evaluated by identifying whether there are any additional elements recited in the claim and evaluating those additional elements to determine whether they integrate the exception into a practical application.
The additional element in claims 1, 3, and 4 of a machine learning model comprising a neural network, the additional element in claim 5 of wherein the neural network comprises a plurality of layers and a plurality of weights, the additional element in claim 11 of a second neural network, the additional element in claim 12 of wherein the second neural network is a generative adversarial network, the additional element in claim 14 of wherein the second neural network includes an autoencoder network, the additional element in claim 17 of the updated machine learning model comprising the neural network does not integrate the judicial exceptions into a practical application because this simply applying the judicial exceptions to neural network technology (2106.05(h)). Further, the additional elements of the neural networks do not integrate the judicial exceptions into a practical application because they also amount to mere instructions to apply the exception. These additional elements amount to mere instructions to apply because the claims only recite the idea of an outcome of using the neural networks without reciting details of how a solution is accomplished (see MPEP 2106.05(f)).
Thus, the additional elements do not integrate the judicial exceptions into a practical application.
(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 additional element in claims in claims 1, 3, 4, 5, 11, 12, 14, and 17 of using a machine learning model comprising a neural network, multiple layers in a neural network, autoencoder neural networks and generative adversarial neural networks are conventional as shown by Schmidhuber et al. (Neural networks 61 (2015): 85-117; newly cited) which reviews neural network technology including neural networks with multiple layers (page 87 right col.) and autoencoders (pages 89, 93, and 102) and as shown by Wang et al. (IEEE/CAA Journal of Automatica Sinica, vol. 4, no. 4, pp. 588-598, 2017; newly cited) which reviews general adversarial neural networks .
Thus, the additional elements (alone or in combination) are not sufficient to amount to significantly more than the judicial exception because they are conventional.
Response to Arguments
Applicant's arguments filed 23 April 2025 have been fully considered but they are not persuasive.
Applicant point to the newly amended limitations in the independent claims and the instant disclosure which states use of the synthetic cases in training can provide for a neural network readily able to call a sub chromosomal aneuploidy, far more efficiently and accurately than some other techniques. Applicant argues the claims are not directed to any judicial exception for being integrated into a practical application at least because the claimed invention improves the function of a computer or improves another technology or technical field (Reply p. 11).
This argument has been fully considered but found to be not persuasive. The MPEP states at 2106.05(a) “It is important to note, the judicial exception alone cannot provide the improvement. The improvement can be provided by one or more additional elements… In addition, the improvement can be provided by the additional element(s) in combination with the recited judicial exception”. The determination of an improvement to technology has two steps, the identification of additional elements (which define the technology) and the evaluation of the additional elements to determine if the improvement is realized in the additional elements either by the additional elements themselves or the additional element in combination with the judicial exception (i.e. the interaction between the judicial exceptions and the additional elements). In the instant case the additional elements of neural networks in the claims do not interact with the judicial exceptions in a manner that provide an improvement in neural network technology. These recited additional elements amount to simply applying judicial exceptions to the technological environment of neural networks (see MPEP 2106.05(h)) and these additional elements further constitute as mere instructions to apply the exception because the claims only recite the idea of an outcome of using the neural networks without reciting details of how a solution is accomplished (see MPEP 2106.05(f)). Further, the claims do not provide an improvement to computer technology because the claims do not recite the use of a computer nor is there evidence that if a computer was recited that the computer would function in a different or improvement manner than being utilized as a tool to process judicial exceptions. The argued improvement of the ability to detect sub-chromosomal aneuploidies through analyzing genetic data falls under an improvement in the judicial exception of analyzing data which is not an improvement to technology because the improvement is not provided in neural network technology.
Claim Rejections - 35 USC § 103
The rejection on the ground of 103 of claims 1, 5-8, 10-18, 21-23, and 55-58 as being unpatentable over Zimmermann et al. (Prenatal Diagnosis, 32, 2012 pages 1233-1241; cited in IDS 06 August 2021) in view of Neocleous et al. (IEEE Journal of Biomedical and Health Informatics, vol. 20, no. 5, pp. 1427-1438, Sept. 2016; previously cited) in Office action mailed 06 November 2024 is withdrawn in view of the amendments of adding active steps for generating an updated machine learning model comprising a neural network and removing the physical steps of isolating DNA, preparing amplified DNA, and sequencing received 23 April 2025.
The rejection on the ground of 103 of claim 3 as being unpatentable over Xu et al. (Cancer Letters, Vol. 370, 2016, 324-331; cited in IDS 06 August 2021) in view of Oustimov et al. (Translational cancer research. 2014 Jun;3(3); previously cited) in Office action mailed 06 November 2024 is withdrawn in view of the amendments of adding active steps for generating an updated machine learning model comprising a neural network and removing the physical steps of isolating DNA, preparing amplified DNA, and sequencing received 23 April 2025.
The rejection on the ground of 103 of claim 4 as being unpatentable over Grskovic et al. (Journal of Molecular Diagnostics, Vol. 18, No. 6, Nov 2016, 890-902; cited in IDS 06 August 2021) in view of Min et al. (Brief Bioinform. 2017 Sep 1;18(5):851-869; previously cited) in Office action mailed 06 November 2024 is withdrawn in view of the amendments of adding active steps for generating an updated machine learning model comprising a neural network and removing the physical steps of isolating DNA, preparing amplified DNA, and sequencing received 23 April 2025.
The rejections below are newly recited necessitated by amendment.
Claims 1, 3, 5, 7, 8, 57 and 58 are rejected under 35 U.S.C. 103 as being unpatentable over Babiarz et al. (US 20170107576 A1; previously cited) in view of Tran et al. (Advances in neural information processing systems 30 (2017); newly cited).
Independent claim 1 is directed to a method for assessing a ploidy state of a fetal chromosome of a pregnant woman, comprising providing genetic sequencing data or genetic array data of a plurality of SNV loci as input to the updated machine learning model comprising the neural network to obtain, as output from the machine learning model, classification information indicative of a ploidy state of the fetal chromosome.
The claim recites genetic sequencing data and genetic array in the alternative form and the BRI of the claim only requires one of this data to be used as input to the model. Babiarz et al. shows utilizing genetic data at the set of polymorphic loci on a chromosome or chromosome segment in a mixed sample comprising fetal DNA and maternal DNA from the mother of the fetus by measuring the quantity of each allele at each locus (Babiarz et al. [0067]). Babiarz et al. shows a classification method utilizing a mathematical process to select a hypothesis which fits the new genetic data the best thereby determining the number of copies of the chromosome segment of interest in the genome of the fetus which is interpreted as classification information indicative of a ploidy state of the fetal chromosome (the hypothesis specifies the number of copies of the chromosome or chromosome segment of interest present in the genome of a fetus) (Babiarz et al. [0067]). Babiarz et al. shows the hypotheses for genetic analysis are represented as models and/or algorithms, the modeler may operate in accordance with models and/or algorithms trained by a machine learning unit and the machine learning unit may develop such models and/or algorithms by applying a classification algorithm to a training set database (Babiarz et al. [0737]).
Babiarz et al. does not show generating an updated machine learning model comprising a neural network, wherein generating the updated machine learning model comprises: generating first training data using a first plurality of cases, training a first machine learning model using the generated first training data; generating second training data using a second plurality of cases, updating the first machine learning model using the generated second training data to obtain the updated machine learning model, wherein at least one of generating the first training data or generating the second training data comprises augmenting the first training data or the second training data, respectively, with one or more synthetic cases that are based on one or more of the first plurality of cases or the second plurality of cases, respectively, or an updated machine learning model comprising the neural network.
Like Babiarz et al., Tran et al. shows a machine learning algorithm for classifying data. Tran et al. shows generating an updated machine learning model comprising a neural network by iteratively training a neural network by updating parameter values using training data (Tran et al. page 2 para. 3 and figure 1). Tran et al. further shows iteratively generating training data as a Bayesian data augmentation algorithm which jointly augments the training data and optimizes the network parameters (Tran et al. page 2 para. 3 and figure 1). Tran et al. shows this Bayesian data augmentation algorithm runs iteratively, where at each iteration new synthetic training points are sampled (Tran et al. page 2 para. 3). Tran et al. shows that the synthetic data is based on the observed training data during a given iteration of the algorithm (Tran et al. page 2 figure 1).
Independent claim 3 is directed to the generation of an updated machine learning model comprising a neural network of claim 1 and to use the updated machine learning model comprising the neural network to produce classification information to identify amplified tumor DNA.
Babiarz et al. shows determining ploidy of a chromosomal segment, can further include detecting a single nucleotide variant at a single nucleotide variant location in a set of single nucleotide variant locations which indicates the presence of circulating tumor nucleic acids in the sample (Babiarz et al. [0349]).
Claim 5 is directed to wherein the neural network outputs a ploidy state value, and wherein the neural network comprises a plurality of layers and a plurality of weights.
Babiarz et al. in view of Tran et al. shows a neural network model from producing a value of a ploidy state of a chromosome or chromosomal segment (Babiarz et al. [0067]). Babiarz et al. in view of Tran et al. shows neural network parameters which are updated during the training process (Tran et al. page 4 para. 1-2). Parameters of a neural network is defined by weights and biases which connect layers of the model.
Claim 7 is directed to wherein the plurality of SNV loci comprises at least 10 loci
Babiarz et al. shows at least 10 or more loci are analyzed for a chromosome or chromosome segment of interest (Babiarz et al. [0091]).
Claim 8 is directed to wherein the sequencing depth is at least 200.
Babiarz et al. shows a depth of read greater than 250 (Babiarz et al. [0441]).
Claim 57 is directed to wherein the updated machine learning model is configured to detect sub-chromosomal aneuploids. Claim 58 is directed to wherein the updated machine learning model is trained to detect sub-chromosomal segments of aneuploides selected from a group consisting of deletion segments, duplications, and/or trisomy segments.
Babiarz et al. shows a hypothesis (which is modeled using allele counts at loci and learned by the updated model) refers to a possible state such as a deletion of a chromosome segment (Babiarz et al. [0166]).
An invention would have been obvious to one or ordinary skill in the art if some motivation in the prior art would have led that person to modify reference teachings to arrive at the claimed invention. It would have been obvious to one of ordinary skill in the art before the effective filling date of the invention to have modified classification method for determining classification information indicative of a ploidy state of a fetal chromosome utilizing machine learning of Babiarz et al. to utilize the machine learning model comprising a neural network classifier and the training algorithm of Tran et al. because this would allow for a method which utilizes a neural network model which intakes numerical data such as allele counts in a mixed sample of a mother and fetus at a plurality of genomic loci to determine the association between allele counts and information indicative of ploidy of a fetal chromosome by iteratively training/updating the neural network model by augmenting a training set with synthetic examples based on the observed training data which leads to better classification results of machine learning models (Tran et al. page 1 abstract). One would have a reasonable expectation of success for this modification because Babiarz et al. shows a classification method which predicts classification information of fetal ploidy utilizing numerical genetic data such as allele counts in a mixed sample by building probability distributions while the neural network of Tran et al. learns the probabilities of data being associated with a given class by updating network parameters.
Claim 4 is rejected under 35 U.S.C. 103 as being unpatentable over Sharon et al. (PLOS Computational Biology 13(8): e1005629; newly cited) in view of Tran et al. (Advances in neural information processing systems 30 (2017); newly cited).
Claim 4 is directed to a method for assessing transplant donor DNA, comprising providing genetic sequencing data or genetic array data of a plurality of SNV loci as input to the updated machine learning model comprising a neural network to obtain, as output form the machine learning model, classification information to identify and quantify amplified transplant donor DNA
The claim recites genetic sequencing data and genetic array in the alternative form and the BRI of the claim only requires one of this data to be used as input to the model. Sharon et al. shows a model that identifies and quantifies transplant donor derived cfDNA fragments using allele abundance at loci in cfDNA sequences derived from sequencing data (Sharon et al. page 3-4 and figure 1).
Sharon et al. does not show generating an updated machine learning model comprising a neural network, wherein generating the updated machine learning model comprises: generating first training data using a first plurality of cases, training a first machine learning model using the generated first training data; generating second training data using a second plurality of cases, updating the first machine learning model using the generated second training data to obtain the updated machine learning model; wherein at least one of generating the first training data or generating the second training data comprises augmenting the first training data or the second training data, respectively, with one or more synthetic cases that are based on one or more of the first plurality of cases or the second plurality of cases, respectively
Tran et al. shows generating an updated machine learning model comprising a neural network by iteratively training a neural network by updating parameter values using training data (Tran et al. page 2 para. 3 and figure 1). Tran et al. further shows generating training data as a Bayesian data augmentation algorithm which jointly augments the training data and optimizes the network parameters (Tran et al. page 2 para. 3 and figure 1). Tran et al. shows this Bayesian data augmentation algorithm runs iteratively, where at each iteration new synthetic training points are sampled (Tran et al. page 2 para. 3). Tran et al. shows that the synthetic data is based on the observed training data during a given iteration of the algorithm (Tran et al. page 2 figure 1).
An invention would have been obvious to one or ordinary skill in the art if some motivation in the prior art would have led that person to modify reference teachings to arrive at the claimed invention. It would have been obvious to one of ordinary skill in the art before the effective filling date of the invention to have modified model for that identifies and quantifies transplant donor derived cfDNA fragments using allele abundance at loci in cfDNA sequences derived from sequencing data of Sharon et al. to utilize the machine learning model comprising a neural network classifier and the training algorithm of Tran et al. because this would allow for a method which utilizes a neural network model which intakes numerical data such as allele abundance from genetic loci derived from sequencing data in a mixed sample to determine the association between allele abundance at genomic locations and information indicative of identifying and quantifying transplant donor DNA by iteratively training/updating the neural network model by augmenting a training set with synthetic examples based on the observed training data which leads to better classification results of machine learning models (Tran et al. page 1 abstract). One would have a reasonable expectation of success for this modification because Sharon et al. shows a model which identifies and quantifies transplant donor DNA by utilizing numerical genetic data such as allele abundance in a mixed sample while the neural network of Tran et al. learns probabilities of data being associated with an output by updating network parameters.
Response to Arguments
Applicant's arguments filed 23 April 2025 have been fully considered but they are not persuasive.
Applicant arguments are rendered moot because they are directed to references that are no longer relied upon for the rejection. There is a new ground of rejection necessitated by amendment.
Conclusion
No claims are allowed.
Claims 6, 10-18, 21-23, 55, 56, 59, and 60 are free of the prior art of record. The prior art of record does not show or render obvious the specific training process in claims 6 and 55, the specific synthetic data in claims 6, 55, and 56, or the specific input and data used for training in claims 59 and 60. Therefore, these claims are free of the prior art of record.
This Office action is a Non-Final action. A shortened statutory period for reply to this action is set to expire THREE MONTHS from the mailing date of this action.
Any inquiry concerning this communication or earlier communications from the examiner should be directed to JONATHAN EDWARD HAYES whose telephone number is (571)272-6165. The examiner can normally be reached M-F 9am-5pm.
Examiner interviews are available via telephone, in-person, and video conferencing using a USPTO supplied web-based collaboration tool. To schedule an interview, applicant is encouraged to use the USPTO Automated Interview Request (AIR) at http://www.uspto.gov/interviewpractice.
If attempts to reach the examiner by telephone are unsuccessful, the examiner’s supervisor, Olivia Wise can be reached at 571-272-2249. The fax phone number for the organization where this application or proceeding is assigned is 571-273-8300.
Information regarding the status of published or unpublished applications may be obtained from Patent Center. Unpublished application information in Patent Center is available to registered users. To file and manage patent submissions in Patent Center, visit: https://patentcenter.uspto.gov. Visit https://www.uspto.gov/patents/apply/patent-center for more information about Patent Center and https://www.uspto.gov/patents/docx for information about filing in DOCX format. For additional questions, contact the Electronic Business Center (EBC) at 866-217-9197 (toll-free). If you would like assistance from a USPTO Customer Service Representative, call 800-786-9199 (IN USA OR CANADA) or 571-272-1000.
/J.E.H./Examiner, Art Unit 1685
/KAITLYN L MINCHELLA/Primary Examiner, Art Unit 1685