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 .
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and the fee set forth in 37 CFR 1.17(e) has been timely paid, the finality of the previous Office action has been withdrawn pursuant to 37 CFR 1.114.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 (i.e., amendment to the claims) filed on November 21, 2025 has been entered. Claims 1, 8, 9, and 14 are currently pending and have been examined as discussed below.
Status of the Claims
The office action is in response to the claims filed on November 21, 2025, for the application filed on November 24, 2023, which is a continuation of International Application No. PCT/KR2022/007419 filed on May 25, 2022, which claims priority to Korean Application No. KR10-2021-0066616 filed on May 25, 2021. Claims 1, 8, 9, and 14 are currently pending and have been examined as discussed below.
Priority
The certified copy of the foreign priority application was filed on December 19, 2023, and it is a non-English language KR application (KR10-2021-0066616). When a claim to priority and the certified copy of the foreign application are received while the application is pending before the examiner, the examiner should review the certified copy to see that it contains no obvious formal defects and that it corresponds in number, date and country to the application identified in the application data sheet for an application filed on or after September 16, 2012, or oath or declaration or application data sheet for an application filed prior to September 16, 2012. See MPEP 215(I).
Applicant cannot rely upon the certified copy of the foreign priority application to overcome the rejections because a translation of said application has not been made of record in accordance with 37 CFR 1.55. When an English language translation of a non-English language application is required, the translation must be that of the certified copy (of the foreign application as filed) submitted together with a statement that the translation of the certified copy is accurate. See MPEP 215 and 216.
Accordingly, the Office requires that English language translations of the associated non-English language foreign application and non-English language PCT application be filed, with the translations being that of the certified copy of the foreign application as filed and submitted together with a statement that the translation of the certified copy is accurate.
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.
The factual inquiries set forth in Graham v. John Deere Co., 383 U.S. 1, 148 USPQ 459 (1966), that are applied for establishing a background for determining obviousness under 35 U.S.C. 103 are summarized as follows:
Determining the scope and contents of the prior art.
Ascertaining the differences between the prior art and the claims at issue.
Resolving the level of ordinary skill in the pertinent art.
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 consider the applicability of 35 U.S.C. 102(b)(2)(C) for any potential 35 U.S.C. 102(a)(2) prior art against the later invention.
Claims 1, 8 – 9, and 14 are rejected under 35 U.S.C. 103(a) as being unpatentable over Kubinski (U.S. Pub. No. 2022/0328132 A1) in view of Park (U.S. Pub. No. 2021/0063407 A1), NPL Wingfield, NPL Townes, Wade (U.S. Pub. No. 2017/0199189 A1), Kashyap (U.S. Pub. No. 2021/0128644 A1), NPL Ma, and Bajaj (U.S. Pub. No. 2014/0179726 A1).
Regarding independent claims 1 and 9, Kubinski teaches the limitations of representative claim 9 identified in bold as:
A diagnostic apparatus for diagnosing an enteric disorder by using a machine learning model (Abstract of Kubinski, Methods, devices, and systems for detecting inflammatory bowel disease (IBD) are described herein… The method further includes calculating a likelihood of inflammatory bowel disease with a machine learning model using the preprocessed data as inputs. In the instant application, the broadest reasonable interpretation of “a diagnostic apparatus for diagnosing an enteric disorder by using a machine learning model” reads on the device in Kubinski (Abstract) for detecting inflammatory bowel disease (IBD) and calculating a likelihood of inflammatory bowel disease with a machine learning model.), comprising:
a processor; and a memory for storing instructions which when executed by the processor cause the processor to (Paragraphs [0121] and [0122] of Kubinski. In the instant application, the broadest reasonable interpretation of “a processor; and a memory for storing instructions which when executed by the processor cause the processor to” reads on the processor and the non-transitory computer readable medium in Kubinski (Paragraphs [0121] and [0122]) in which the computer program is stored for implementing all the modules (for e.g., modules 103, 105, 107, 203, 205, 209, 211, 223, 225, 227, 307, 407, 409, 410, 411, 419, 421, 423, 425, 427, 503, 505, 507, 509, etc.).):
extract multiple microbial data based on an analysis result of a mixture of a gut-derived substance collected from a subject and a gut environment-like composition (Paragraph [0118] of Kubinski, DNA can then be extracted using a validated DNA extraction method, such as the protocol and instruments used in the Qiagen PowerFecal DNA extraction kit. Extracted DNA can then have its 16S rRNA gene V4 region amplified using 515F-806R primers as specified in the Earth Microbiome Project. Paragraph [0119] of Kubinski, DNA can then be extracted using a validated DNA extraction method, such as the protocol and instruments used in the Qiagen PowerFecal DNA extraction kit. Extracted DNA can then have its 16S rRNA gene V4 region amplified using 515F-806R primers as specified in the Earth Microbiome Project… Raw sequencing data can then be processed into taxonomic features as outlined in the present subject matter. For a given stool sample, the use of genus level data is proposed, paired with ILR normalization, and zero-centered batch reduction prior to prediction generation using a Random Forest model. Predictive values generated by the Random Forest model can then be used as a diagnostic aid. In the instant application, the broadest reasonable interpretation of “extract multiple microbial data based on an analysis result of a mixture of a gut-derived substance collected from a subject and a gut environment-like composition” reads on the activity in Kubinski (Paragraphs [0118] and [0119]) of extracting DNA and amplifying the 16S rRNA gene V4 region of the extracted DNA using 515F-806R primers as specified in the Earth Microbiome Project, and processing raw sequencing data into taxonomic features, e.g., for a given stool sample, the use of genus level data is proposed, paired with ILR normalization, and zero-centered batch reduction prior to prediction generation.), including culturing the mixture under anaerobic conditions for 18 to 24 hours, centrifuging the cultured mixture to separate a supernatant and a precipitate, and analyzing the supernatant and the precipitate to extract data on at least one of a content, a concentration or a kind of one or more of endotoxins, hydrogen sulfides, short-chain fatty acids (SCFAs) or microbiota-derived metabolites contained in the cultured mixture, and a change in kind, a concentration, a content or a diversity of bacteria included in the microbiota;
select multiple microbe-related features to be used for the machine learning model from the multiple microbial data based on a predetermined feature selection algorithm (Paragraph [0046] of Kubinski, The 16S rRNA gene may then be sequenced. Raw sequencing data from the 16S rRNA gene may be processed via a bioinformatic pipeline into taxonomic features such as operational taxonomic units (OTUs), bacterial genera or bacterial species. These features may then be filtered and normalized and batch effects can be removed. Paragraph [0051] of Kubinski, Filtering the data removes outliers that could have a negative impact on the machine learning model's performance. In the instant application, the broadest reasonable interpretation of “select multiple microbe-related features to be used for the machine learning model from the multiple microbial data based on a predetermined feature selection algorithm” reads on the activity in Kubinski (Paragraphs [0046] and [0051]) of filtering and normalizing taxonomic features of processed raw sequencing data from the 16S rRNA gene, such as operational taxonomic units (OTUs), bacterial genera or bacterial species, removing batch effects, and removing outliers.), wherein the selecting of the multiple microbe-related features comprises:
performing a primary selection from features through a Boruta algorithm to select a subset of features;
checking an error value depending on a number of features through a binomial deviance plot to determine an optimal range of the number of features, wherein the optimal range is 1 to 23;
determining an optimal number of features within the optimal range based on accuracy, wherein the optimal number is 14; and
selecting the 14 multiple microbe-related features through an extreme gradient boosting (XGB) model based on importance of the features;
train the machine learning model by using the multiple microbe-related features (Paragraph [0110] of Kubinski, At 467, the batch-effect-free, filtered, normalized and engineered features are provided to a training module 427. The training module 427 can be configured to train and test different machine learning modules using a leave-one-dataset-out approach to determine the best combination of data preprocessing and machine learning model. At 469, the training module 427 outputs a trained and tested machine learning model 429. In the instant application, the broadest reasonable interpretation of “a training unit configured to train the machine learning model by using the multiple microbe-related features” reads on the activity in Kubinski (Paragraph [0110]) of training and testing different machine learning modules using the batch-effect-free, filtered, normalized and engineered features of processed raw sequencing data from the 16S rRNA gene.), and to perform supervised learning based on labeling of whether the enteric disorder is present for each of the microbial data and contents of microbes included in the selected multiple microbe-related features (Paragraph [0095] of Kubinski, At 467, [T]he training unit 120 may train machine learning model to predict whether constipation is present for each of microbial data by performing supervised learning based on labeling of whether constipation is present for each of the microbial data (learning data) and the amount of microbes related to the selected feature. In the instant application, the broadest reasonable interpretation of “perform supervised learning based on labeling of whether the enteric disorder is present for each of the microbial data and contents of microbes included in the selected multiple microbe-related features” reads on the activity in Kubinski (Paragraph [0095]) of performing supervised learning based on labeling of whether constipation is present for each of the microbial data (learning data) and the amount of microbes related to the selected feature.); and
diagnose the enteric disorder by inputting, into the trained machine learning model, the microbial data collected from the subject to be tested, wherein diagnosing of the enteric disorder is based on an output value of the trained machine learning model (Paragraph [0068] of Kubinski, At 113, the batch reduced, normalized, filtered and engineered features are then fed into a machine learning module 105. The machine learning module 105 is configured to apply a pre-trained machine learning model to the features to generate a classification value at 115 of IBD status for the sample. The classification 115 can then be used to generate IBD prediction 107. In the instant application, the broadest reasonable interpretation of “a diagnostic unit that diagnoses the enteric disorder by inputting, into the trained machine learning model, the microbial data collected from the subject to be tested, wherein diagnosing of the enteric disorder is based on an output value of the trained machine learning model” reads on the activity in Kubinski (Paragraph [0068]) of generating, by feeding the batch reduced, normalized, filtered and engineered features into the machine learning module 105, the classification value of IBD status for the sample and the IBD prediction 107.),
wherein each of the microbes is selected from genera belonging to comprising the family Tannerellaceae, the family Bifidobacteriaceae, the family Ruminococcaceae, the family Clostridaceae, the family Lachnospiraceae, the family Bacteroidaceae, the family Erysipelatoclostridiaceae, the family Veilonellaceae, the family Bacteroidaceae, the family Ruminococcaceae, the family Lachnospiraceae, and the family Anaerovoracaceae.
Kubinski does not appear to explicitly recite, but Park teaches the limitations in bold as “extract multiple microbial data based on an analysis result of a mixture of a gut-derived substance collected from a subject and a gut environment-like composition, including culturing the mixture under anaerobic conditions for 18 to 24 hours (Paragraph [0059] of Park, [T]he culturing in the process (b) may be performed for 12 hours to 48 hours and specifically for 18 hours to 24 hours, but may not be limited thereto. In the instant application, the broadest reasonable interpretation of “culturing the mixture under anaerobic conditions for 18 to 24 hours” reads on the activity in Park (Paragraph [0059]) performing the culturing in the process (b) for 18 to 24 hours.), centrifuging the cultured mixture to separate a supernatant and a precipitate, and analyzing the supernatant and the precipitate (Paragraph [0090] of Park, Each of the cultured test groups was centrifuged to separate the supernatant and the pellet. Then, metabolites, short-chain fatty acids, toxic substances and the like from the supernatant were analyzed and microbiota from the pellet were analyzed. In the instant application, the broadest reasonable interpretation of “centrifuging the cultured mixture to separate a supernatant and a precipitate and analyzing the supernatant and the precipitate” reads on the activity in Park (Paragraph [0090]) of centrifuging the cultured test groups to separate the supernatant and the pellet and analyzing the supernatant (e.g., its metabolites, short-chain fatty acids, toxic substances and the like) and the pellet (e.g., its microbiota).) to extract data on at least one of a content, a concentration or a kind of one or more of endotoxins, hydrogen sulfides, short-chain fatty acids (SCFAs) or microbiota-derived metabolites contained in the cultured mixture, and a change in kind, a concentration, a content or a diversity of bacteria included in the microbiota” (Paragraph [0062] of Park, [T]he analyzing of the culture in the process (c) is to analyze the kind, content and/or concentration of one or more of an endotoxin, hydrogen sulfide as a product of abnormal intestinal fermentation, short-chain fatty acids (SCFAs) and microbiota-derived metabolites contained in the culture and to analyze a change in kind, content and/or concentration when the sample is treated with the candidate material. In the instant application, the broadest reasonable interpretation of “extract data on at least one of a content, a concentration or a kind of one or more of endotoxins, hydrogen sulfides, short-chain fatty acids (SCFAs) or microbiota-derived metabolites contained in the cultured mixture, and a change in kind, a concentration, a content or a diversity of bacteria included in the microbiota” reads on the activity in Park (Paragraph [0062]) of analyzing the kind, content and/or concentration of one or more of an endotoxin, hydrogen sulfide as a product of abnormal intestinal fermentation, short-chain fatty acids (SCFAs) and microbiota-derived metabolites contained in the culture.).
Kubinski does not appear to explicitly disclose, but NPL Wingfield teaches the limitation identified in bold as “select multiple microbe-related features to be used for the machine learning model from the multiple microbial data based on a predetermined feature selection algorithm, wherein the selecting of the multiple microbe-related features comprises performing a primary selection from features through a Boruta algorithm to select a subset of features” (Third Paragraph in First Column on Page 1085 of NPL Wingfield, Boruta - an all-relevant feature selection algorithm based on a random forest wrapper and widely used in metagenomics [17] - was chosen to assess which features were relevant for each stage of the hybrid classifier (see Table I). In the instant application, the broadest reasonable interpretation of “the selecting of the multiple microbe-related features comprises performing a primary selection from features through a Boruta algorithm to select a subset of features” reads on Boruta in NPL Wingfield (Page 1085) being the all-relevant feature selection algorithm based on a random forest wrapper and widely used in metagenomics, with Boruta being chosen to assess which features were relevant for each stage of the hybrid classifier.).
Kubinski does not appear to explicitly disclose, but NPL Townes teaches the limitation identified in bold as “select multiple microbe-related features to be used for the machine learning model from the multiple microbial data based on a predetermined feature selection algorithm, wherein the selecting of the multiple microbe-related features comprises … checking an error value depending on a number of features through a binomial deviance plot to determine an optimal range of the number of features, wherein the optimal range is 1 to 23” (First Paragraph to Second Paragraph in First Column on Page 14 and FIG.1(d) of NPL Townes, Feature selection using deviance. Genes with constant expression across cells are not informative. Such genes may be described by the multinomial null model where
π
i
j
=
π
j
. Goodness of fit to a multinomial distribution can be quantified using deviance, which is twice the difference in log-likelihoods comparing a saturated model to a fitted model. The multinomial deviance is a joint deviance across all genes,and for this reason is not helpful for screening informative genes. Instead, one may use the binomial deviance as an approximation:
PNG
media_image1.png
95
560
media_image1.png
Greyscale
A large deviance value indicates the model in question provides a poor fit. Those genes with biological variation across cells will be poorly fit by the null model and will have the largest deviances. By ranking genes according to their deviances, one may thus obtain highly deviant genes as an alternative to highly variable or highly expressed genes.. In the instant application, the broadest reasonable interpretation of “the selecting of the multiple microbe-related features comprises … checking an error value depending on a number of features through a binomial deviance plot” reads on the activity in NPL Townes (First Paragraph to Second Paragraph in First Column on Page 14 and FIG.1(d)) of using binomial deviance to rank genes according to their deviances and select genes based on these rankings. The Office has determined that it was well known in the art of medical data mining and computer-aided diagnosis at the time of filing that binomial deviance measures model error (i.e., checking an error value) because it quantifies the discrepancy between a fitted logistic regression model (using binomial errors) and a perfect model (saturated model), with a lower deviance indicating a better fit (i.e., acting as a goodness-of-fit test).).
Kubinski does not appear to explicitly disclose, but Kashyap teaches the limitation identified in bold as “select multiple microbe-related features to be used for the machine learning model from the multiple microbial data based on a predetermined feature selection algorithm, wherein the selecting of the multiple microbe-related features comprises … checking an error value depending on a number of features through a binomial deviance plot to determine an optimal range of the number of features, wherein the optimal range is 1 to 23” (Paragraph [0037] of Kashyap, In some cases, a biomarker that can be used [to] indicate gut microbiota dysbiosis (e.g., to predict susceptibility to C. difficile infection) in a mammal can be one or more (e.g., at least one, at least two, at least three, at least four, or more) bacteria. The broadest reasonable interpretation of “determine an optimal range of the number of features, wherein the optimal range is 1 to 23” reads on the biomarker of Kashyap (Paragraph [0037]) that can be used [to] indicate gut microbiota dysbiosis and can be one or more (e.g., at least one, at least two, at least three, at least four, or more) bacteria. In the case where the claimed ranges "overlap or lie inside ranges disclosed by the prior art" a prima facie case of obviousness exists. See MPEP 2144.05.).
Kubinski does not appear to explicitly disclose, but Wade teaches the limitation identified in bold as “select multiple microbe-related features to be used for the machine learning model from the multiple microbial data based on a predetermined feature selection algorithm, wherein the selecting of the multiple microbe-related features comprises … determining an optimal number of features within the optimal range based on accuracy, wherein the optimal number is 14” (Paragraph [0139] of Wade, For example, the method can determine the prognostic outcome for particular subject x represented as input vector x, wherein input vector x comprises a number of variable features in relation to a scenario of interest for which there is a global dataset D of samples also having the same variable features relating to the scenario as input vector x, and for which an outcome is known, the method comprising: (A) optimizing the transductive model by: a) determining what number and a subset Vx of variable features of input vector x will be used in assessing an outcome for the input vector x ... h) until the accuracy is maximized, wherein a number and a subset Vx of variable features of input vector x, and a number Kx of samples from within the global data set D that form a neighborhood about input vector x are determined anew each time elements a) and b) are repeated while applying an optimization procedure to optimize Vx and/or Kx. In the instant application, the broadest reasonable interpretation of “the selecting of the multiple microbe-related features comprises … determining an optimal number of features within the optimal range based on accuracy” reads on the activity in Wade (Paragraph [0139]) of determining what number and a subset Vx of variable features of input vector x will be used in assessing an outcome for the input vector x.).
Kubinski does not appear to explicitly disclose, but Kashyap teaches the limitation identified in bold as “select multiple microbe-related features to be used for the machine learning model from the multiple microbial data based on a predetermined feature selection algorithm, wherein the selecting of the multiple microbe-related features comprises … determining an optimal number of features within the optimal range based on accuracy, wherein the optimal number is 14” (Paragraph [0037] of Kashyap, In some cases, a biomarker that can be used [to] indicate gut microbiota dysbiosis (e.g., to predict susceptibility to C. difficile infection) in a mammal can be one or more (e.g., at least one, at least two, at least three, at least four, or more) bacteria. The broadest reasonable interpretation of “the optimal number is 14” reads on the biomarker of Kashyap (Paragraph [0037]) that can be used [to] indicate gut microbiota dysbiosis … can be one or more. In the case where the claimed ranges "overlap or lie inside ranges disclosed by the prior art" a prima facie case of obviousness exists. See MPEP 2144.05.).
Kubinski does not appear to explicitly disclose, but NPL Ma teaches the limitation identified in bold as “select multiple microbe-related features to be used for the machine learning model from the multiple microbial data based on a predetermined feature selection algorithm, wherein the selecting of the multiple microbe-related features comprises … selecting the 14 multiple microbe-related features through an extreme gradient boosting (XGB) model based on importance of the features” (Third Paragraph in First Column to Second Paragraph in Second Column on Page 3 of NPL Ma, The core symptoms were selected by using Lasso analysis. The eXtreme Gradient boosting (XGBoost) algorithm was used in model establishment. The parameters of XGBoost were optimized for optimal model. After that, the performance of the model was evaluated by the accuracy. The XGBoost refers to a machine learning element algorithm, which usually sets a weak learner as a base classifer, and then a strong learner is constructed [21, 22]. In this study, we adopted the XGBoost based on a decision tree to achieve extreme gradient improvement by using the XGBoost and the caret packages. The five-fold cross-validation training was used for optimization of parameters. Parameters include nrounds, colsample_bytree, min_child_weight, eta, gamma, subsample, and max_depth. The optimal parameter combination was screened according to the accuracy, and after that, the optimal model was established. The optimal probability threshold for minimizing the error could be found through the information package. Figure 2 on Page 6 of NPL Ma, Core symptoms selection using the Lasso regression model. (a) Lasso coefficient profiles of the 37 texture features. A coefficient profile plot was produced against the log l sequence. (b) Tuning parameter
l
selection in the LASSO model used 10-fold cross validation via minimum criteria. Binomial deviance was plotted versus log
l
. Dotted vertical lines were drawn at the optimal values by using the minimum criteria, and the 1 standard error of the minimum criteria (the 1-SE criteria). A
l
value of 0.01389958 was chosen (1-SE criteria) according to 10-fold cross validation. In the instant application, the broadest reasonable interpretation of “the selecting of the multiple microbe-related features comprises … selecting the 14 multiple microbe-related features through an extreme gradient boosting (XGB) model based on importance of the features” reads on the activity in NPL Ma (Third Paragraph in First Column to Second Paragraph in Second Column on Page 3; and Figure 2 on Page 6) of performing a selection of symptoms by using a Lasso analysis, determining optimal values via the binomial deviance plotted versus log l, and optimizing parameters of an eXtreme Gradient boosting (XGBoost) algorithm.).
Kubinski does not appear to explicitly recite, but Bajaj teaches the limitation in bold as “each of the microbes is selected from genera comprising the family Tannerellaceae, the family Bifidobacteriaceae, the family Ruminococcaceae, the family Clostridaceae, the family Lachnospiraceae, the family Bacteroidaceae, the family Erysipelatoclostridiaceae, the family Veilonellaceae, the family Bacteroidaceae, the family Ruminococcaceae, the family Lachnospiraceae, and the family Anaerovoracaceae” (Paragraph [0155] of Bajaj, This accords with studies showing that … and Lachnospiraceae spp are associated with reduced intestinal inflammation in Crohn's disease.).
Therefore, it would have been obvious to one of ordinary skill in the art of medical data mining and computer-aided diagnosis at the time of filing to modify the method and system of Kubinski to: include the activities of culturing the mixture under anaerobic conditions for 18 to 24 hours, centrifuging the cultured mixture to separate a supernatant and a precipitate, and analyzing the supernatant and the precipitate to extract data on at least one of a content, a concentration or a kind of one or more of endotoxins, hydrogen sulfides, short-chain fatty acids (SCFAs) or microbiota-derived metabolites contained in the cultured mixture, and a change in kind, a concentration, a content or a diversity of bacteria included in the microbiota, as taught by Park (Paragraph [0059], [0062], and [0090]) in order to provide information for diagnosing a disease caused by an intestinal disorder (Paragraph [0075] of Park); implement the activity of the selecting of the multiple microbe-related features comprising performing a primary selection from features through a Boruta algorithm to select a subset of features, as taught by NPL Wingfield (Third Paragraph in First Column on Page 1085) in order to provide every relevant feature in a classification problem (Third Paragraph in First Column on Page 1085 of NPL Wingfield); implement the activity of the selecting of the multiple microbe-related features comprising checking an error value depending on a number of features through a binomial deviance plot, as taught by NPL Townes (First Paragraph to Second Paragraph in First Column on Page 14 and FIG.1(d)) in order to provide feature selection that excludes uninformative genes such as those which exhibit no meaningful biological variation across samples (Last Paragraph in Second Column on Page 1 to First Paragraph in First Column on Page 2 of NPL Townes); implement the activity of determining an optimal range of the number of features, wherein the optimal range is 1 to 23 and implement the optimal number being 14, as taught by Kashyap (Paragraph [0037]) in order to determine a clinical risk factor profile can help identify patients with diarrhea and dysbiosis who may be at higher risk of CDI (e.g., patients having irritable bowel disease, patients who are immunosuppressed, and/or hospitalized patients) (Paragraph [0035] of Kashyap); determining an optimal number of features within the optimal range based on accuracy, as taught by Wade (Paragraph [0139]) in order to improve diagnostic tests, and prevent or minimize some of the adverse outcomes of disease (Paragraph [0006] of Wade); implement the activity of selecting the 14 multiple microbe-related features through an extreme gradient boosting (XGB) model based on importance of the features, as taught by NPL Ma (Third Paragraph in First Column to Second Paragraph in Second Column on Page 3 and Figure 2 on Page 6) in order to establish a new diagnostic model based on clinical symptoms with reference to the Transparent Reporting of a multivariable prediction model for Individual Prognosis Or Diagnosis (First Column, Fourth Paragraph on Page 2 of NPL Ma); and implement each of the microbes being selected from genera comprising the family Tannerellaceae, the family Bifidobacteriaceae, the family Ruminococcaceae, the family Clostridaceae, the family Lachnospiraceae, the family Bacteroidaceae, the family Erysipelatoclostridiaceae, the family Veilonellaceae, the family Bacteroidaceae, the family Ruminococcaceae, the family Lachnospiraceae, and the family Anaerovoracaceae, as taught by Bajaj (Paragraph [0155]) in order to correlate the presence of particular microbes with particular diseases and conditions, and/or the risk of developing the same (Paragraph [0008] of Bajaj).
Regarding claims 8 and 14, Kubinski as modified by as modified by Park, NPL Wingfield, NPL Townes, Wade, Kashyap, NPL Ma, and Bajaj teaches the limitations in bold as “each of the microbes is selected from species belonging to comprising the genus Parabacteroides, the genus Bifidobacterium, the genus Subdoligranulum, the genus Clostridium, the genus Ruminococcus, the genus Bacteroides, the genus Erysipelatoclostridium, the genus RF39, the genus Veillonella, the genus Bacteroides, the genus Eubacterium, the genus GCA.900066575, and the genus UCG.010” (Paragraph [0037] of Kashyap, In some cases, bacteria can be decreased in gut microbiota dysbiosis. Examples of bacteria that can be decreased in gut microbiota dysbiosis include, without limitation, bacteria belonging to the genera … Bacteroides... In some cases, bacteria can be increased in gut microbiota dysbiosis. Examples of biomarkers that can be increased in gut microbiota dysbiosis include, without limitation, bacteria belonging to the genera… Bacteroides. In the instant application, the broadest reasonable interpretation of “each of the microbes is selected from species comprising the genus Parabacteroides, the genus Bifidobacterium, the genus Subdoligranulum, the genus Clostridium, the genus Ruminococcus, the genus Bacteroides, the genus Erysipelatoclostridium, the genus RF39, the genus Veillonella, the genus Bacteroides, the genus Eubacterium, the genus GCA.900066575, and the genus UCG.010” reads on the bacteria of Kashyap (Paragraph [0037] belonging to the genera Bacteroides).
Response to Amendment
Applicant’s amendment and arguments (Third Paragraph on Page 6 to First Paragraph on Page 8 of the Amendment filed November 21, 2025) regarding the rejection of claims 1, 6, 9, and 12 under 35 U.S.C. § 112 have been fully considered and are persuasive. Therefore, the rejection claims 1, 6, 9, and 12 under 35 U.S.C. § 112 has been withdrawn.
Applicant's amendment and arguments (Fifth Paragraph on Page 7 to Third Paragraph on Page 11 of the Amendment filed November 21, 2025) regarding the rejection of claims 1, 8 – 9, and 14 under 35 U.S.C. § 103 have been fully considered but are moot in view of the new grounds of rejection necessitated by the amendment.
In the Amendment (Seventh Paragraph to Eighth Paragraph on Page 9), Applicant argues:
The Claimed Invention Adds Two Novel Steps Absent from NPL Ma
Step (a) - Boruta Primary Selection: The claimed invention starts with a subset of features and uses Boruta algorithm for primary selection. NPL Ma uses the Delphi method (expert consultation) for preliminary screening of 37 symptoms, not any algorithmic method. NPL Ma does not mention or suggest the Boruta algorithm.
Step (c) - Accuracy-Based Optimization: After determining the range of 1 to 23, the claimed invention evaluates accuracy to select the optimal number "14". NPL Ma contains no such optimization step-the 19 features are simply the output of Lasso with the selected X, not the result of accuracy-based evaluation within a predetermined range.
NPL Ma has not been cited for disclosing these limitations. Thus, the Office has deemed the argument to be unpersuasive.
Conclusion
Any inquiry concerning this communication or earlier communications from the examiner should be directed to VINCENT CAESAR ILAGAN whose telephone number is (703) 756-1639. The examiner can normally be reached Monday - Friday 8:30 am - 6:00pm.
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, Jason B. Dunham, can be reached on (571) 272-8109. The fax phone number for the organization where this application or proceeding is assigned is 571-272-8300.
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.
/V.C.I./Examiner, Art Unit 3686
/DEVIN C HEIN/Examiner, Art Unit 3686