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 .
Information Disclosure Statement
The information disclosure statements (IDS) submitted on February 22, 2024 and July 9, 2025 are in compliance with the provisions of 37 CFR 1.97. Accordingly, the information disclosure statements are being considered by the examiner.
Claim Rejections - 35 USC § 102
The following is a quotation of the appropriate paragraphs of 35 U.S.C. 102 that form the basis for the rejections under this section made in this Office action:
A person shall be entitled to a patent unless –
(a)(1) the claimed invention was patented, described in a printed publication, or in public use, on sale, or otherwise available to the public before the effective filing date of the claimed invention.
(a)(2) the claimed invention was described in a patent issued under section 151, or in an application for patent published or deemed published under section 122(b), in which the patent or application, as the case may be, names another inventor and was effectively filed before the effective filing date of the claimed invention.
Claims 1-6, 10-14, 23-29 are rejected under 35 U.S.C. 102(a)(1) as being anticipated by Rong et al.
(“NormAE: Deep Adversarial Learning Model to Remove Batch Effects in Liquid Chromatography Mass Spectrometry-Based Metabolomics Data”, 2020).
Regarding claim 1, Rong teaches a method of training a classifier (Page 5084, Figure 1C) to predict a disease status of a patient (Page 5086, Column 2, Paragraph 1, “the performance of the methods was mainly evaluated in two aspects… retention of biological information1”) to be executed by a processor2, the method comprising:
i) receiving a plurality of training data sets (Page 5083, Column 2, Paragraph 1, “We used two LC-MS untargeted metabolomics data sets to evaluate our method”) derived from a plurality of batches (Page 5083, Column 2, Paragraph 1, “one contained a total of 644 plasma samples in four batches with 85 QCs and the other contained 644 plasma samples in four batches with 81 QCs”) , each training data set comprising: a diagnostic data (Page 5084, Column 1, Paragraph 1, “The R package XCMS (version 3.2) was then used to preprocess these files with the mzXML format and generate a data matrix that consisted of the retention time, mass-to-charge ratio (m/z) values, and peak intensity. After XCMS data processing, the R package CAMERA was used to annotate the peaks.”) ; a batch ID (Page 5084, Figure 1C3) ; and a disease ID (Page 5086, Column 2, Paragraph 1, “the performance of the methods was mainly evaluated in two aspects… retention of biological information… The latter was evaluated using… predicted accuracy of CE versus CRC”4);
ii) using the received plurality of training data sets to generate a batch classifier (Page 5084, Figure 1C5); and
iii) using the received plurality of training data and the generated batch classifier to generate a disease status classifier (Page 5084, Figure 1C6)
Regarding claim 2, Rong teaches a method of training a classifier to predict a responsible variable of a subject, to be executed by a processor, the method comprising: i) receiving a plurality of training data sets derived from a plurality of batches, each training data set comprising: an (one or more) explanatory variable; a batch ID; and a target variable; ii) using the received plurality of training data sets to generate a batch classifier; and iii) using the received plurality of training data and the generated batch classifier to generate a target variable classifier (see claim 1 analysis7).
Regarding claim 3, Rong teaches the method of claim 1 wherein said generating the disease status classifier comprises using a neural network (Page 5084, Figure 1C).
Regarding claim 4, Rong teaches wherein the neural network comprises an input layer, at least one hidden layer and an output layer, and wherein [[step]] iii) comprises inputting an output of the batch classifier in one of the at least one hidden layers (Page 5084, Figure 1C).
Regarding claim 5, Rong teaches wherein the neural network comprises an input layer, a plurality of hidden layers and an output layer, and wherein [[step]] iii) comprises inputting an output of the batch classifier in one of the plurality of hidden layers (Page 5084, Figure 1C).
Regarding claim 6, Rong teaches wherein [[step]] iii) comprises inputting an output of the batch classifier in one of the second half from the middle of the plurality of hidden layers (Page 5084, Figure 1C8).
Regarding claim 10, Rong teaches wherein the neural network is selected from the group consisting of: Perceptron (P), Feed Forward (FF), Radial Basis Function Network (RBF), Deep Feed Forward (DFF), Recurrent Neural Network (RNN), Long/Short Term Memory (LSTM), Gated Recurrent Unit (GRU), Auto Encoder (AE), Variational AE (VAE), Denoising Auto Encoder (DAE), Sparse AE (SAE), Markov Chain (MC), Hopfield Network (HN), Boltzmann Machine (BM), Restricted BM (RBM), Deep Belief Network (DBN), Deep Convolutional Network (DCN), Deconvolutional Network (DN), Deep Convolutional Inverse Graphics Network (DCIGN), Generative Adversarial Network (GAN), Liquid State Machine (LSM), Extreme Learning Machine (ELM), Echo State Network (ESN), Deep Residual Network (DRN), Kohonen Network (KN), Support Vector Machine (SVM), and Neural Turing Machine (NTM) (Page 5084, Figure 1C9).
Regarding claim 11, Rong teaches wherein [[step]] ii) comprises training the batch classifier (Page 5084, Column 2, Paragraph 1, “To optimize the AE, an adversarial training procedure and an discriminator Fb that is trained to classify the batch labels based on the latent representations are required.”).
Regarding claim 12, Rong teaches wherein said training the batch classifier comprises using a regression model (Page 5086, Column 1, Equation 310).
Regarding claim 13, Rong teaches wherein the regression model is a linear regression model (Page 5084, Column 2, Equation 211).
Regarding claim 14, Rong teaches wherein the regression model is a logistic regression model (Page 5084, Column 1, Equation 312) .
Regarding claim 23, Rong teaches wherein the disease is a cancer (Page 5083, Column 2, Paragraph 2, “They consisted of… 571 colorectal cancer (CRC) patients”)
Regarding claim 24, Rong teaches wherein the cancer is selected from a group of: brain tumor, lung cancer, breast cancer, thyroid cancer, esophagus cancer, liver cancer, biliary tract cancer, gastric cancer, pancreas cancer, colorectal cancer, prostate cancer, renal cancer, bladder cancer, uterine cancer, cervical cancer, ovarian cancer, skin cancer, lymphoma, leukemia (see claim 23 analysis).
Regarding claim 25, the majority of limitations present are covered in the rejection of claim 1. New limitations introduced in this claim will be covered below.
Rong teaches providing a patient data comprising a diagnostic data related to the patient (Page 5083-5084, “analysis of all plasma samples… preprocess these files with the mzXML format and generate a data matrix that consisted of the retention time, mass-to-charge ratio (m/z) values, and peak intensity.”), and using the generated batch classifier to select a batch ID among the plurality of batches (Page 5084, Column 1, Paragraph 2, “A total of 644 subject samples were separated into four batches”), which the patient data is likely to match among the plurality of batches (Page 5084, Column 2, Paragraph 1, “…an discriminator Fb that is trained to classify the batch labels based on the latent representations”).
Regarding claim 26, all of limitations present are covered in the rejection of claim 1 and claim 25.
Regarding claim 27, the majority of limitations present are covered in the rejection of claim 1 and claim 25. New limitations introduced in this claim will be covered below.
Rong teaches providing a classifier to predict a disease status of a patient (Page 5084, Figure 1C).
Regarding claim 28, all of limitations present are covered in the rejection of claim 1, claim 4 and claim 25.
Regarding claim 29, Rong teaches a computer program, to be executed by a processor, comprising the method of claim 26 (Page 5086, Column 1, Paragraph 7, “The NormAE model was implemented using PyTorch modules (version 1.2.0). The source code is available at https://github.com/luyiyun/NormAE”)
Claim Rejections - 35 USC § 103
The following is a quotation of 35 U.S.C. 103 which forms the basis for all obviousness rejections set forth in this Office action:
A patent for a claimed invention may not be obtained, notwithstanding that the claimed invention is not identically disclosed as set forth in section 102, if the differences between the claimed invention and the prior art are such that the claimed invention as a whole would have been obvious before the effective filing date of the claimed invention to a person having ordinary skill in the art to which the claimed invention pertains. Patentability shall not be negated by the manner in which the invention was made.
Claims 7, 9 are rejected under 35 U.S.C. 103 as being unpatentable over Rong et al.
(“NormAE: Deep Adversarial Learning Model to Remove Batch Effects in Liquid Chromatography Mass Spectrometry-Based Metabolomics Data”, 2020), in view of Xu et al. (“I-Vector-Based Patient Adaptation of Deep Neural Networks for Automatic Heartbeat Classification”, 2019).
Regarding claim 7, Rong teaches the method of inputting an output of the batch classifier into one of the pluralities of hidden layers (see claim 5 analysis). Rong fails to teach wherein the hidden layer is one of the last third of the plurality of hidden layers. However, Xu teaches a system of auxiliary input into a hidden layer of a neural network, where the input occurs in the last third of the plurality of hidden layers (Page 724, Column 1, Paragraph 4, “The general classifier has three hidden layers”, Page 724, Table 313)
Rong and Xu are considered analogous to the invention because all are directed towards machine learning systems. Therefore, it would have been obvious to one of ordinary skill in the art
before the effective filing date of the invention to have modified Rong to incorporate the teachings of Xu, and input the output oof the batch classifier in one of the last third of the plurality of hidden layers. Doing so can significantly affect the ensuing classification of the main neural network model (see Page 724, Table 3 of Xu).
Regarding claim 9, Rong teaches the method of inputting an output of the batch classifier into one of the pluralities of hidden layers (see claim 5 analysis). Rong fails to teach wherein the hidden layer is the last hidden layer just before the output layer. However, Xu teaches a system of auxiliary input into a hidden layer of a neural network, where the input occurs in the last hidden layer (see claim 7 analysis).
Rong and Xu are considered analogous to the invention because all are directed towards machine learning systems. Therefore, it would have been obvious to one of ordinary skill in the art
before the effective filing date of the invention to have modified Rong to incorporate the teachings of Xu, and input the output oof the batch classifier in one of the last third of the plurality of hidden layers. Doing so can significantly affect the ensuing classification of the main neural network model (see Page 724, Table 3 of Xu).
Claim 8 is rejected under 35 U.S.C. 103 as being unpatentable over Rong et al.
(“NormAE: Deep Adversarial Learning Model to Remove Batch Effects in Liquid Chromatography Mass Spectrometry-Based Metabolomics Data”, 2020), in view of Xu et al. (“I-Vector-Based Patient Adaptation of Deep Neural Networks for Automatic Heartbeat Classification”, 2019) and in further view of Breier et al. (“POSTER: Practical Fault Attack on Deep Neural Networks”, 2018).
Regarding claim 8, Rong teaches the method of inputting an output of the batch classifier into one of the pluralities of hidden layers (see claim 5 analysis). Rong fails to teach wherein the hidden layer is one of the last quarter of the plurality of hidden layers. However, Xu teaches a system of auxiliary input into a hidden layer of a neural network, where the input occurs in the last hidden layer (see claim 7 analysis) and Breier teaches a system of fault injection into a hidden layer of a neural network, where the neural network has four hidden layers, and the fault injection occurs in the last hidden layer (Page 2206, Table 214, Page 2206, Column 2, Paragraph 2, “Figures 1 (a) and 1 (b) show a random fault model when attacking the last hidden layer”).
Rong, Xu and Breier are considered analogous to the invention because all are directed towards machine learning systems. Therefore, it would have been obvious to one of ordinary skill in the art
before the effective filing date of the invention to have modified Rong to incorporate the teachings of Xu and Breier, and input the output of the batch classifier in one of the last quarter of the plurality of hidden layers. Doing so can significantly affect the ensuing classification of the main neural network model (see Page 724, Table 3 of Xu), in a way distinct from alternating an input layer or the output layer of the neural network (see Bullets 1-3 of Page 2206, Column 1, Paragraph 2 of Breier).
Claims 15-22 are rejected under 35 U.S.C. 103 as being unpatentable over Rong et al.
(“NormAE: Deep Adversarial Learning Model to Remove Batch Effects in Liquid Chromatography Mass Spectrometry-Based Metabolomics Data”, 2020), in view of Varela-Martinez et al. (“Molecular Signature of Aluminum Hydroxide Adjuvant in Ovine PBMCs by Integrated mRNA and microRNA Transcriptome Sequencing”, 2018).
Regarding claim 15, Rong fails to teach the further limitations of the claim. However, Varela-Martinez teaches wherein the diagnostic data comprises gene expression levels (Page 3, Column 2, Paragraph 4, “Differential gene expression analysis was performed using three different R packages…”).
Rong and Varela-Martinez are considered analogous to the invention because all are directed towards disease analysis methods that correct for batch effect. Therefore, it would have been obvious to one of ordinary skill in the art before the effective filing date of the invention to have modified Rong to incorporate the teachings of Varela-Martinez and applied the system to gene expression levels. Doing so allows one to get accurate readings on diagnostic data that is most prone to batch effect.
Regarding claim 16, Rong fails to teach the further limitations of the claim. However, Varela-Martinez teaches wherein the diagnostic data comprises RNA expression levels (Page 4, Column 1, Paragraph 4, “The differential miRNA15 expression analysis was performed using the edgeR package”).
Rong and Varela-Martinez are considered analogous to the invention because all are directed towards disease analysis methods that correct for batch effect. Therefore, it would have been obvious to one of ordinary skill in the art before the effective filing date of the invention to have modified Rong to incorporate the teachings of Varela-Martinez and applied the system to RNA expression levels. Doing so allows one to get accurate readings on diagnostic data that is most prone to batch effect.
Regarding claim 17, Rong fails to teach the further limitations of the claim. However, Varela-Martinez teaches wherein the RNA expression levels are acquired by a microarray measurement or a sequencing method on RNAs (Page 2, Column 2, Paragraph 4, “Thus, the main objective of this study was to identify the molecular signature activated by vaccines and adjuvants… by combining the molecular information provided by RNA sequencing of both mRNA and miRNA in an invivo experiment.”).
Rong and Varela-Martinez are considered analogous to the invention because all are directed towards disease analysis methods that correct for batch effect. Therefore, it would have been obvious to one of ordinary skill in the art before the effective filing date of the invention to have modified Rong to incorporate the teachings of Varela-Martinez and applied the system to RNA expression levels acquired by a sequencing method. Doing so allows one to get accurate readings on diagnostic data that is most prone to batch effect.
Regarding claim 18, Rong fails to teach the further limitations of the claim. However, Varela-Martinez teaches wherein the RNA expression levels comprise a profile including a plurality of RNA expression levels (Page 4, Column 1, Paragraph 3, “The sRNAbech module was used to map the sequences to the Ovisaries reference genome Oar3.1, to profile the expression of small RNAs and to predict novel miRNAs”).
Rong and Varela-Martinez are considered analogous to the invention because all are directed towards disease analysis methods that correct for batch effect. Therefore, it would have been obvious to one of ordinary skill in the art before the effective filing date of the invention to have modified Rong to incorporate the teachings of Varela-Martinez and applied the system to a profile that included multiple RNA expression levels. Doing so allows one to get accurate readings on diagnostic data that is most prone to batch effect.
Regarding claim 19, Rong fails to teach the further limitations of the claim. However, Varela-Martinez teaches wherein the RNA may be cfRNA, RNA in cells, or RNAs included in extracellular vesicles (Page 3, Column 1, Paragraph 2, “Total RNA was extracted from PBMCs16…”).
Rong and Varela-Martinez are considered analogous to the invention because all are directed towards disease analysis methods that correct for batch effect. Therefore, it would have been obvious to one of ordinary skill in the art before the effective filing date of the invention to have modified Rong to incorporate the teachings of Varela-Martinez and applied the system to RNA collected from cells. Doing so allows one to get accurate readings on diagnostic data that is most prone to batch effect.
Regarding claim 20, Rong fails to teach the further limitations of the claim. However, Varela-Martinez teaches wherein the RNA is selected from the group consisting of miRNA, and mRNA (see claim 17 analysis).
Rong and Varela-Martinez are considered analogous to the invention because all are directed towards disease analysis methods that correct for batch effect. Therefore, it would have been obvious to one of ordinary skill in the art before the effective filing date of the invention to have modified Rong to incorporate the teachings of Varela-Martinez and applied the system to mRNA. Doing so allows one to get accurate readings on diagnostic data that is most prone to batch effect.
Regarding claim 21, Rong fails to teach the further limitations of the claim. However, Varela-Martinez teaches wherein the RNA is derived from a body fluid selected from the group consisting of: blood, serum, plasma, lymph fluid, tissue fluids, interstitial fluid, intercellular fluid, cavity fluid, serosal fluid, pleural fluid, ascites fluid, pericardial fluid, cerebrospinal fluid, joint fluid (synovial fluid), and aqueous humor of the eye (aqueous) (Page 3, Column 1, Paragraph 1, “For the isolation of ovine peripheral blood mononuclear cells (PBMCs), blood was collected from the jugular vein of 14 Rasa Aragonesa sheep.”).
Rong and Varela-Martinez are considered analogous to the invention because all are directed towards disease analysis methods that correct for batch effect. Therefore, it would have been obvious to one of ordinary skill in the art before the effective filing date of the invention to have modified Rong to incorporate the teachings of Varela-Martinez and applied the system to RNA derived from blood. Doing so allows one to get accurate readings on diagnostic data that is most prone to batch effect.
Regarding claim 22, Rong fails to teach the further limitations of the claim. However, Varela-Martinez teaches wherein the RNA is derived from a tissue obtained by a biopsy or during a surgical operation (see claim 21 analysis17).
Rong and Varela-Martinez are considered analogous to the invention because all are directed towards disease analysis methods that correct for batch effect. Therefore, it would have been obvious to one of ordinary skill in the art before the effective filing date of the invention to have modified Rong to incorporate the teachings of Varela-Martinez and applied the system to RNA derived from a biopsy. Doing so allows one to get accurate readings on diagnostic data that is most prone to batch effect.
Claims 30 is rejected under 35 U.S.C. 103 as being unpatentable over Rong et al. (“NormAE: Deep Adversarial Learning Model to Remove Batch Effects in Liquid Chromatography Mass Spectrometry-Based Metabolomics Data”, 2020), in view of Benjamin et al. (“US 20210174908”).
Regarding claim 30, Rong teaches the program comprising the method of claim one (see claim 1 and claim 29 analysis). Rong fails to teach a computer system (Paragraph 50, “ In some aspects, the method is implemented in a computer system…”) of classifying a disease status of a patient (Abstract, “The disclosure provides population and non-population-based classifiers to categorize patients and cancers”), comprising:
at least one processor (Paragraph 50, “…a computer system comprising at least one processor”)
a memory storing at least one program to be executed by the at least one processor (Paragraph 50, “…the at least one memory comprising instructions executed by the at least one processor”).
Rong and Benjamin are considered analogous to the invention because all are directed towards disease analysis methods that correct for batch effect. Therefore, it would have been obvious to one of ordinary skill in the art before the effective filing date of the invention to have modified Rong to incorporate the teachings of Benjamin and appl the system to a hardware setting. Doing so allows for easy storage and reproducibility of the system.
Conclusion
Any inquiry concerning this communication or earlier communications from the examiner should be directed to MATTHEWOS MESFIN whose telephone number is (571)270-0782. The examiner can normally be reached Monday-Friday 8am-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, Cesar Paula can be reached at (571) 272-4128. 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.
/MATTHEWOS MESFIN/Examiner, Art Unit 2145
/CESAR B PAULA/Supervisory Patent Examiner, Art Unit 2145
1 Biological information related to disease, “consisted of 73 chronic enteritis (CE) patients and 571 colorectal cancer (CRC) patients”, Page 5083, Column 2, Paragraph 2
2 “The Norm AE model was implemented using PyTorch modules”, Page 5086, Column 1, Paragraph 7
3 The batch label
4 CRC and CE are the two types of cancer patient data they received. In order to gauge the predictive capabilities of the system on the two, the data must include an identifier for which disease the sample came from, i.e. a disease ID
5 Fb is the batch classifier
6 The disease status classifier is the encoder/decoder that the batch classifier feeds into. It classifies disease status by reconstructing the input. The disease status classifying is evidence by the above references.
7 Claim 2 is a broader set of limitations that are all essentially covered by the rejection of claim 1
8 The batch classifier output goes into the third of four hidden layers.
9 The neural network is a form of an auto encoder.
10 The loss function can be considered a form of regression.
11 The loss function is a form of MAE (Mean Absolute Error) which is a linear loss function
12 This specific loss calculates likelihood which requires a logistic model
13 The table shows the i-vector being injected into each of the hidden layers, including bottom H, which is the last of the three.
14 Shows hidden layer 4
15 miRNA is a subcategory of RNA
16 The previous paragraph indicates that PBMC is an acronym for peripheral blood mononuclear cells
17 A form of liquid biopsy
Prosecution Insights