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 .
Claim Status
Claims 1-22 are pending and are examined on the merit.
Priority
As recorded on the 06/22/2023 filing receipt, the effective filing date of the claimed invention is 05/11/2023. At this point in examination, all claims have been interpreted as being accorded on this priority date. In future actions, the effective filing date of one or more claims may change, due to amendments to the claims, or further analysis of the disclosure(s) of the priority application(s).
Information Disclosure Statement
The information disclosure statement (IDS) submitted on 11/06/2023 is in compliance with the provisions of 37 CFR 1.97. Accordingly, the list of cited references was considered in full by the examiner. A signed copy of the corresponding 1449 form has been included with this Office action.
Drawings
The black and white drawings filed 05/11/2023 are accepted.
Objection to the drawings
The supplemental drawings filed 05/11/2023 are objected to. The file wrapper contains color drawings under Supplemental Content, and the color appears to be integral to the disclosure. Color drawings are not accepted in utility applications unless a petition filed under 37 CFR 1.84(a)(2) is granted. 37 CFR 1.4(c) and the USPTO "Requirements of a Petition" webpage (https://www.uspto.gov/patents-application-process/petitions/01-requirements-petition) provide further information on filing a petition. It is particularly important that such a petition must clearly explain why color drawings are necessary (37 CFR 1.84(a)(2)). Any such petition must be accompanied by the appropriate fee set forth in 37 CFR 1.17(h) and, unless already present, an amendment to include the following language as the first paragraph of the brief description of the drawings section of the specification:
The patent or application file contains at least one drawing executed in color. Copies of this patent or patent application publication with color drawing(s) will be provided by the Office upon request and payment of the necessary fee.
Specification
The specification filed 05/11/2023 is accepted.
Claim rejection - 35 USC§ 112(b)
The following is a quotation of 35 U.S.C. 112(b):
(b) CONCLUSION. —The specification shall conclude with one or more claims particularly pointing out and distinctly claiming the subject matter which the inventor or a joint inventor regards as the invention.
Claim 22 is 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.
Claim 22 is to a 101 machine or manufacture, i.e. a "a system" in this instance, limited according to its claimed physical structure, but it is not clear what is the structure associated with the recited "[determining step] ..." and similar steps. The recited "determining step" is not interpreted as requiring structure clearly linking the "system" to the recited steps in a structural sense appropriate to a claim to a machine or manufacture, as such, rendering the claims indefinite. Relatedly, MPEP 2173.05(p). II pertains regarding a claim directed to both product and process.
This rejection might be overcome by, for example, reciting a data storage device, comprised by the "system" and instructions stored therein and configured according to the recited elements and steps, as supported at instant specification pg. 8 last paragraph.
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 1-22 are rejected under 35 U.S.C. 101 because the claimed invention is directed to an abstract idea without significantly more.
Matter belonging to no statutory category (claim 22)
Regarding claim 22, the claimed invention is directed to a non-statutory subject matter.
The claim does not fall within at least one of the four categories of patent eligible subject matter because the claim is not recited as a process, and the claim is not limited to any particular structure as a 101 machine or manufacture. The claim reads on transitory propagating signals which are not proper patentable subject matter because it does not fit within any of the four statutory categories of invention Un re Nuijten, Federal. Circuit, 2006).
For the purposes of compact prosecution, the claim will be analyzed as if it were a statutory category of invention, however, the claim must be amended.
Judicial exceptions (JE) to 101 patentability (claims 1-22)
The Supreme Court has established a two-step framework for this analysis, wherein a claim does not satisfy § 101 if (1) it is “directed to” a patent-ineligible concept, i.e., a law of nature, natural phenomenon, or abstract idea, and (2), if so, the particular elements of the claim, considered “both individually and as an ordered combination,” do not add enough to “transform the nature of the claim into a patent-eligible application.” Elec. Power Grp., LLC v. Alstom S.A., 830 F.3d 1350, 1353 (Fed. Cir. 2016) (quoting Alice, 134 S. Ct. at 2355). Applicant is also directed to MPEP 2106.
Step 1: The instantly claimed invention (claim(s) 1-21 being representative) is directed to methods. Therefore, the instantly claimed invention falls into one of the four statutory categories. [Step 1: YES]
The instantly claimed invention (claim(s) 22 being representative) is directed to a system without any recitation of physical structure, therefore, does not fall within any statutory category [Step 1: NO]. Thus, instant claim 22 is non-statutory, warranting a rejection for failure to claim statutory subject matter.
Step 2A: First it is determined in Prong One whether a claim recites a judicial exception, and if so, then it is determined in in Prong Two if the recited judicial exception is integrated into a practical application of that exception.
Step 2A, Prong 1: Under the MPEP § 2106.04, the Step 2A (Prong 1) analysis requires determining whether a claim recites an abstract idea, law of nature, or natural phenomenon.
Claim(s) 1-22 recite the following steps which fall under the mathematical concepts, mental processes, and/or certain methods of organizing human activity groupings of abstract ideas:
Claims 1, 21, and 22 recite computing a tree that includes representations computed from genetic sequences of a plurality of microbes within a microbiome sample of a subject; the limitation computing a tree is considered mathematical calculation, see specification pg. 9, last para., and as such, falls into mathematical concepts groupings of abstract ideas.
Claims 1, 21, and 22 further recite mapping the tree to a data structure that defines the taxonomy hierarchy and defines positions between neighboring representations of the plurality of microbes; the limitation mapping a tree to a data structure is considered a mathematical calculation, as disclosed in instant specification pg. 25, para. 3-10, and as such, falls into mathematical concepts groupings of abstract ideas. Also, mental process of mapping a tree to a data structure.
Claims 1, 21, and 22 further recite feeding the data structure into a machine learning model that processes neighboring representations according to position; the limitation feeding data structure into a model is considered a mathematical calculation, see specification pg. 3, para. 6, and as such, falls into mathematical concepts groupings of abstract ideas.
Claim 21 further recites creating a training dataset of a plurality of records (mental process of creating a dataset) and training the machine learning model on the training dataset (under a broadest reasonable interpretation, training a machine learning model includes mathematical concepts).
Claim 22 further recites the machine learning model is trained by: computing the tree that includes representations computed from genetic sequences of the plurality of sample microbiome samples (mathematical calculation/mathematical concepts); mapping the tree to the data structure (mathematical calculation/mathematical concepts); creating a training dataset of a plurality of records mental process of creating a dataset).
Claims 2 and 3 recite ordering elements of the data structure each denoting a type of microbe according to at least one similarity feature between different types of microbes at a same taxonomy level of the taxonomy hierarchy while preserving the structure of the tree; the limitation “ordering”, given the plain meaning of ordering, encompasses observation, evaluation, judgment, and opinion (See MPEP 2106.04(a)(2), subsection III.) performable by human mind (mental process), since human mind is capable of ordering data.
Claim 4 recites that the ordering is computed recursively per taxonomic category level (mathematical calculation/mathematical concepts).
Claim 5 recites that the ordering is done for positioning taxa with similar frequencies relatively closer together and taxa with less similar frequencies further apart (mathematical calculation/mathematical concepts).
Claim 6 recites that at least one similarity feature includes similaity of frequency of microbes (mathematical calculation/mathematical concepts).
Claim 7 recites that at least one similarity feature is computed based on Euclidean distances (mathematical calculation/mathematical concepts).
Claim 8 recites that at least one similarity feature is computed by building a dendrogram based on the Euclidean distances computed for a hierarchical clustering of the representations of the microbes according to frequency (mathematical calculation/mathematical concepts).
Claim 9 recites computing the tree that includes representations computed from genetic sequences of the plurality of sample microbiome samples (mathematical calculation/mathematical concepts); mapping the tree to the data structure mathematical calculation/mathematical concepts); creating a training dataset of a plurality of records (mental process of creating a dataset);
Claim 9 further recites training the machine learning model on the training dataset; the limitation training a machine learning model, is considered a mathematical calculation, since the model is trained by computing a tree and mapping (mathematical processes). As such, said limitation falls into mathematical concepts groupings of abstract ideas.
Claim 10 recites including each observed taxa of microbes of the microbiome sample in a leaf at a respective taxonomic level of the taxonomy hierarchy (mental process of including data) and adding to each leaf a log-normalized frequency of the observed taxa of microbes ( mathematical process of adding an average; also, mental process of adding data), and each internal node includes an average of direct descendants of the internal node located at lower levels.
Claims 14 and 15 recite applying an explainable Al platform to the machine learning model for obtaining an estimate of portions of the data structure used by the machine learning model for obtaining the outcome (mathematical calculations/ to get results (mathematical concepts)) and projecting the portions of the data structure to the tree (mathematical calculation/mathematical concepts).
Claim 15 further recites iterating the applying (mental process of repeating).
Claim 17 recites pre-processing the genetic sequences by clustering the genetic sequences to create Amplicon Sequence Variants (ASVs) and creating a vector of the ASVs (mental process clustering data and creating a vector/matrix/table).
Claim 18 recites computing a log-normalization of the ASV frequencies of the vector to obtain a log-normalized vector of ASV frequencies (mathematical calculation/mathematical concepts).
Claim 19 recites recursively ordering the log-normalized ASV frequencies according to at least one similarity feature computed based on Euclidean distances between log-normalized ASV frequencies (mental process of repeatedly reordering data).
Claims 11-13, 16, and 20 provide additional information.
Additionally, claims 1-22 recite a correlation between microbiome sample of subject and state of the medical condition of the subject, and as such, falls into judicial exception of Laws of nature and natural phenomena. See MPEP 2106(b) I.
The identified claims recite a law of nature, a natural phenomenon (product of nature) and/or fall into one of the groups of abstract ideas of mathematical concepts, mental processes, and/or certain methods of organizing human activity for the reasons set forth above. See MPEP 2106.04 (a)(2) III and MPEP 2106.04 (b) I. Therefore, claims are directed to one or more judicial exception(s) and require further analysis in Prong Two. [Step 2A, Prong 1: YES]
Step 2A: Prong 2: Under the MPEP § 2106.04, the Step 2A, Prong 2 analysis requires identifying whether there are any additional elements recited in the claim beyond the judicial exception(s), and evaluating those additional elements to determine whether they integrate the exception into a practical application of the exception. This judicial exception is not integrated into a practical application for the following reasons.
The additional elements of claim(s) 1-22 include the following.
Claims 1, 21, and 22 recite obtaining the state of the medical condition of the subject as an outcome of the machine learning model (outputting).
Claims 9 and 21-22 recite obtaining a plurality of sample microbiome samples from a plurality of subjects.
Claim 14 recites obtaining the outcome.
The additional elements of obtaining a plurality of sample microbiome amount to nothing more than gathering the data necessary to perform the abstract idea. Obtaining a biological sample is performed in order to gather data for the menta/mathematical analysis step and is a necessary precursor for the recited exception.
Furthermore, the additional elements of obtaining the state of the medical condition /outputting amount to necessary data gathering and outputting. The courts have found the limitations that amount to necessary data gathering and outputting are insignificant extra-solution activity that do not integrate a recited judicial exception into a practical application in Mayo, 566 U.S. at 79, 101 USPQ2d at 1968 and O/P Techs., Inc. v. Amazon.com, Inc., 788 F.3d 1359, 1363, 115 USPQ2d 1090, 1092-93 (Fed. Cir. 2015) (see MPEP 2106.05(g)).
Therefore, the additionally recited elements amount to insignificant extra-solution activity and, as such, the claims as a whole do not integrate the abstract idea into practical application.
MPEP 2106.04(d). I lists the following example considerations for evaluating whether a judicial exception is integrated into a practical application:
An improvement in the functioning of a computer or an improvement to other technology or another technical field, as discussed in MPEP §§ 2106.04(d)(1) and 2106.05(a).
Applying or using a judicial exception to effect a particular treatment or prophylaxis for a disease or medical condition, as discussed in MPEP § 2106.04(d)(2);
Implementing a judicial exception with, or using a judicial exception in conjunction with, a particular machine or manufacture that is integral to the claim, as discussed in MPEP § 2106.05(b).
Effecting a transformation or reduction of a particular article to a different state or thing, as discussed in MPEP § 2106.05(c); and
Applying or using the judicial exception in some other meaningful way beyond generally linking the use of the judicial exception to a particular technological environment, such that the claim as a whole is more than a drafting effort designed to monopolize the exception, as discussed in MPEP § 2106.05(e).
In Step 2A, Prong 1 above, claim steps and/or elements were identified as part of one or more judicial exceptions (JEs).
In Step 2B below, any remaining steps and/or elements are therefore in addition to the identified JE(s). Any such additional steps and additional elements are further discussed in Step 2B.
Here in Step 2A, Prong 2, no additional step or element clearly demonstrates integration of the JE(s) into a practical application.
At this point in examination, it is not yet the case that any of the Step 2A, Prong 2 considerations enumerated above clearly demonstrates integration of the identified JE(s) into a practical application. Referring to the considerations above, none of 1. an improvement, 2. treatment, 3. a particular machine or 4. a transformation is clear in the record.
In conclusion regarding Prong 2, claims 1-22 are directed to an abstract idea. [Step 2A, Prong 2: NO]
Step 2B: In the second step it is determined whether the claimed subject matter includes additional elements that amount to significantly more than the judicial exception. An inventive concept cannot be furnished by an abstract idea itself. See MPEP § 2106.05.
The additional elements of claim(s) 1-22 include the following.
Claims 1, 21, and 22 recite obtaining the state of the medical condition of the subject as an outcome of the machine learning model (outputting).
Claims 9 and 21-22 recite obtaining a plurality of sample microbiome samples from a plurality of subjects.
Claim 14 recites obtaining the outcome.
The additional elements of obtaining a plurality of sample microbiome amount to nothing more than gathering the data necessary to perform the abstract idea. Obtaining a biological sample is performed in order to gather data for the mental/mathematical analysis step and is a necessary precursor for the recited exception.
Therefore, the additional element is not sufficient to amount to significantly more than the judicial exception.
Taken alone, the additional elements do not amount to significantly more than the above-identified judicial exception(s). Even when viewed as a combination, the additional elements fail to transform the exception into a patent-eligible application of that exception. Thus, the claims as a whole do not amount to significantly more than the exception itself. [Step 2B: NO]
Therefore, the instantly rejected claims are not drawn to eligible subject matter as they are directed to an abstract idea without significantly more.
Claim Rejections - 35 USC § 103
In the event the determination of the status of the application as subject to AIA 35 U.S.C. 102 and 103 (or as subject to pre-AIA 35 U.S.C. 102 and 103) is incorrect, any correction of the statutory basis (i.e., changing from AIA to pre-AIA ) for the rejection will not be considered a new ground of rejection if the prior art relied upon, and the rationale supporting the rejection, would be the same under either status.
The following is a quotation of 35 U.S.C. 103 which forms the basis for all obviousness rejections set forth in this Office action:
A patent for a claimed invention may not be obtained, notwithstanding that the claimed invention is not identically disclosed as set forth in section 102, if the differences between the claimed invention and the prior art are such that the claimed invention as a whole would have been obvious before the effective filing date of the claimed invention to a person having ordinary skill in the art to which the claimed invention pertains. Patentability shall not be negated by the manner in which the invention was made.
The factual inquiries for establishing a background for determining obviousness under 35 U.S.C. 103 are summarized as follows:
1. Determining the scope and contents of the prior art.
2. Ascertaining the differences between the prior art and the claims at issue.
3. Resolving the level of ordinary skill in the pertinent art.
4. Considering objective evidence present in the application indicating obviousness or nonobviousness.
Claims 1-3, 5-6, 9-15, 17-19, and 21-22 are rejected under 35 U.S.C. 103 as being unpatentable over Reiman ( PopPhy-CNN: A Phylogenetic Tree Embedded Architecture for Convolutional Neural Networks to Predict Host Phenotype From Metagenomic Data, Published in: IEEE Journal of Biomedical and Health Informatics ( Volume: 24, Issue: 10, October 2020), Page(s): 2993 – 3001; as cited in the IDS form dated 11/06/2023), in view of Mills (US20250179576A1; as cited in the attached 892 form).
Regarding claims 1, 21, and 22, the recited computer implemented method of determining the state of a medical condition of a subject, is taught as, deep learning framework for the prediction of host phenotype with the ability of facilitating the retrieval of predictive microbial taxa (inherently disclosing computer-implemented) (Reiman: abstract).
The recited computing a tree that includes representations computed from genetic sequences of a plurality of microbes within a microbiome sample of a subject, wherein the tree includes the representations of the plurality of microbes arranged according to a taxonomy hierarchy, is taught as, constructing a phylogenetic tree to preserve the relationship among the microbial taxa in the profile; the phylogenetic tree populated with the relative abundance of microbial taxa in a metagenomic sample (Reiman: pg. 2994, col. 1, last para. – col. 2, first para.; abstract).
The recited mapping the tree to a data structure that defines the taxonomy hierarchy and defines positions between neighboring representations of the plurality of microbes, is taught as, the tree is populated with the relative abundance of microbial taxa in each individual profile and represented in a 2D matrix; The constructed matrices provide spatial and quantitative information in the metagenomic data, which are more suitable to CNN compared to the input vectors of relative microbial taxa abundances in an arbitrary order; empowering CNNs to explore the spatial relationship of the taxonomic annotations on the tree and their quantitative characteristics in metagenomic data (Reiman: pg. 2994, col. 2, para. 1; abstract; Appendix: Algorithm S1 and S2).
The recites feeding the data structure into a machine learning model that processes neighboring representations according to position, is taught as, inputting the 2D matrix into a convolutional neural network (Reiman: abstract).
The recited obtaining the state of the medical condition of the subject as an outcome of the machine learning model, is taught as, a practical deep learning framework for the prediction of host phenotype with the ability of facilitating the retrieval of predictive microbial taxa (Reiman: abstract).
Further regarding claims 21 and 22, the recited creating a training dataset of a plurality of records, wherein a record comprises the data structure and a ground truth indication of the state of a medical condition of a subject from the plurality of subjects, is taught as, used nine publicly available datasets to evaluate PopPhy CNN. Three datasets are contained within the MetAML package: cirrhosis, type 2 diabetes (T2D), and obesity; the six other datasets were taken from a study on inflammatory bowel disease (IBD) investigating the differences between the microbial community during disease remission and flares (Reiman: pg. 2996, col. 1, last para. – col. 2, first para.; Table I-VI).
Further regarding claims 21 and 22, the recited training the machine learning model on the training dataset, wherein the machine learning model processes neighboring representations according to position, wherein the machine learning model generates the state of the medical condition of the subject in response to an input of a data structure computed from a microbiome sample obtained from the subject; is taught as, the taxa and count table are used to create and populate a phylogenetic tree, which is represented as a matrix and used to train a CNN; Features are extracted from the trained model (Reiman: Fig. 1; pg. 2996, col. 2, subsection C. Extraction of important features).
Further regarding claims 1, 21, and 22, Reiman does not teach that the data structure defines positions between neighboring representations of the plurality of microbes. Mills teaches this limitation.
Mills teaches a method of determining an abundance of one or more microbial features in a sample obtained from a subject to predict and/or diagnose an disease status of the subject including 16S processing pipeline, taxonomic classification, computing beta diversity distances unweighted UniFrac and weighted UniFrac distance matrices (for example, ordering based on similarity) (Mills: claim 36, [0174]). Mills further teaches hierarchical clustering of studies and pathways was performed using Euclidean distance (Mills: [0019] and [0020]).
Regarding claim 2, the recited ordering elements of the data structure each denoting a type of microbe according to at least one similarity feature between different types of microbes at a same taxonomy level of the taxonomy hierarchy while preserving the structure of the tree, is taught as, a prototype of our algorithm to transform the microbial taxonomic abundance profiles into a structured data by using a phylogenetic tree, where a phylogenetic tree that captures similarity information among OTUs can be constructed by comparing the microbial genomes based on multiple sequence alignment and organizing similar taxa into clades (Reiman: pg. 2994, col. 2, last para. – pg. 2995 col.1; Appendix).
Regarding claim 3, the recited ordering is done per taxonomic category level, for positioning the representations of microbes that are more similar closer together and representation of microbes that are less similar are positioned further apart while preserving the structure of the tree, is taught as, the similarity between taxa is represented by their closeness in the tree; PhyloT was used to create the phylogenetic tree, and a constant distance of one between nodes in the tree is assumed; the phylogenetic tree is structured using ancestral nodes from both taxonomic groups and subgroups with no defined distances between nodes (Reiman: pg. 2994, col. 2, last para. – pg. 2995 col.1; Appendix)
Regarding claim 5, the recited ordering is done for positioning taxa with similar frequencies relatively closer together and taxa with less similar frequencies further apart, is taught as, the value of each OTU from a sample is assigned to its respective node in the tree. The tree is then populated such that an abundance value for each internal node is equal to the sum of its children's abundance values (Reiman: pg. 2995, col. 1, para. 2, Appendix: Algorithm S1).
Regarding claim 6, the recited similarity feature includes similaity of frequency of microbes, is taught as, annotating the tree with abundance values (Reiman: pg. 2995, col. 1, para. 2).
Regarding claim 9, the recited machine learning model is trained by: obtaining a plurality of sample microbiome samples from a plurality of subjects, computing the tree that includes representations computed from genetic sequences of the plurality of sample microbiome samples; mapping the tree to the data structure; creating a training dataset of a plurality of records, wherein a record comprises the data structure and a ground truth indication of the state of the medical condition of a subject from the plurality of subjects; and training the machine learning model on the training dataset, is taught as, using datasets of cirrhosis, type 2 diabetes (T2D), inflammatory bowel disease (IBD), and obesity, as well as multiclass datasets taken from diseased patients and healthy subjects and evaluating the algorithm using said datasets (Reiman: pg. 2996, subsection D. datasets used in evaluation).
Regarding claim 10, Mills teaches a method of determining an abundance of one or more microbial features in a sample obtained from a subject to predict and/or diagnose an disease status of the subject including 16S processing pipeline, taxonomic classification, computing beta diversity distances unweighted UniFrac and weighted UniFrac distance matrices (for example, ordering based on similarity) (Mills: claim 36, [0174]). Mills further teaches hierarchical clustering of studies and pathways was performed using Euclidean distance (Mills: [0019] and [0020]).
Further regarding the recited the tree is created by including each observed taxa of microbes of the microbiome sample in a leaf at a respective taxonomic level of the taxonomy hierarchy, and adding to each leaf a log-normalized frequency of the observed taxa of microbes, and each internal node includes an average of direct descendants of the internal node located at lower levels, is taught as, samples were first normalized to relative abundances by dividing ASV read counts by the total reads per sample. Relative abundances were summed at each taxonomic level using Pandas version 0.25.3 (https://pandas.pydata.org/). At the per-study level, log 2 ratios of ASD and control subject means were calculated using SciPy version 1.5.2 and log transformed in NumPy version 1.19.1 (Mills: [0177]).
Regarding claim 11, the recited data structure is implemented as a graph that includes values of the tree and an adjacent matrix, that are fed into a layer of a graph convolutional neural network implementation of the machine learning model, wherein an identity matrix is added with a learned coefficient, and output of the layer is fed into a fully connected layer that generates the outcome, is taught as, adjacency matrix containing the tree representations; once the tree has been annotated with abundance values, it is transformed into a matrix format (Reiman: Appendix: Algorithm S2).
Regarding claim 12, the recited data structure is represented as an image created by projecting the tree to a two dimensional matrix with a plurality of rows matching a number of different taxonomy category levels of the taxonomy hierarchy represented by the tree, and a column for each leaf, wherein the machine learning model is implemented as a convolutional neural network, is taught as, adjacency matrix containing the tree representations; once the tree has been annotated with abundance values, it is transformed into a matrix; for a given row, the children of the nodes from that node is selected and their abundance is placed in the subsequent row in order that their parents appear, starting with the left most column, where the columns equal to the largest number of nodes in any layer, where the model is a convolutional neural network (Reiman: Appendix: para. 1; Algorithm S2).
Regarding claim 13, the recited at each respective level of the different taxonomy category levels, a value of the two dimensional matrix at the respective level is set to the value of leaves below the level or a value at a higher level is set to have an average of values of a level below, and below a leaf values are set to zero, and wherein a level below includes a plurality of different values, a plurality of positions of the two dimensional matrix of the current level are set to an average of the plurality of different values of the level below, is taught as, The tree is used as a template to construct a populated tree for each sample in the dataset. The value of each OUT from sample is assigned to its respective node in the tree. The tree is then populated such that an abundance value for each internal node is equal to the sum of its children’s abundance values. Once the tree has been annotated with abundance values, it is transformed into a matrix by placing the root’s abundance in the top left corner of a matrix. Then for a given row, the children of the nodes from that row are selected and their abundances are placed in the subsequent row in the order that their parents appear, starting with the leftmost column. (Reiman: pg. 2995, col. 1, para. 2; Appendix: para. 1; Algorithm S2).
Regarding claim 14, the recited applying an explainable Al platform to the machine learning model for obtaining an estimate of portions of the data structure used by the machine learning model for obtaining the outcome, and projecting the portions of the data structure to the tree for obtaining an indication of which microorganisms of the microbiome most contributed to the outcome of the machine learning model, is taught as, applying he Shapley additive explanations (SHAP) package to the Gradient Boosting Classifier model trained on the aggregate ASV dataset to determine the importance of each feature (Mills: [0181]; FIG. 5C).
Regarding claim 15, the recited iterating the applying the explainable Al platform for each of a plurality of data structures of sequentially obtained microbiome samples, for creating a plurality of heatmaps each at a different color channel and combining the plurality of heatmaps into a single multi-channel image, and projecting the multi-channel image on the tree, is taught as, the relative abundance and SHAP value for each sample of the top 25 features plotted in decreasing feature importance from the gradient boosting classifier model trained on the aggregate data at the ASV level. FIG. 5D is a cluster heatmap generated and colored by the SHAP values of the top 25 features from the gradient boosting classifier model trained and tested on the aggregate data and each cohort independently (Mills: [0020]; FIG. 5C-D).
Regarding claim 17, the recited pre-processing the genetic sequences by clustering the genetic sequences to create Amplicon Sequence Variants (ASVs), and creating a vector of the ASVs, wherein each entry of the vector represents a microbe at a certain taxonomy level of the taxonomy hierarchy and comprises the representation, wherein the tree is created from the vector by placing respective values of the vector at corresponding taxonomic category levels of the taxonomy hierarchy of the tree, is taught as, using amplicon sequencing variants (ASVs) for taxonomic resolution analysis; utilizing the GBC model on read count data either uncollapsed or aggregated at seven different taxonomic levels (Kingdom, Phylum, Class, Order, Family, Genus, Species, ASV) were performed; (Mills: [0149])
Regarding claim 18, the recited computing a log-normalization of the ASV frequencies of the vector to obtain a log-normalized vector of ASV frequencies, is taught as, samples were first normalized to relative abundances by dividing ASV read counts by the total reads per sample. Relative abundances were summed at each taxonomic level using Pandas; The log 10 transformed relative abundance in each sample for the significant taxa that were highlighted was performed in Pandas version 0.25.3. To avoid instances of infinity due to log 10 transformation, the minimum non-zero normalized relative abundance value was added to all samples utilizing NumPy version 1.19.1. Next, each sample's normalized relative abundance was log 10 transformed with NumPy version 1.19.1 before plotting (Mills: [0177]).
Rationale for combining Reiman and Mills:
In KSR Int 'l v. Teleflex, the Supreme Court, in rejecting the rigid application of the teaching, suggestion, and motivation test by the Federal Circuit, indicated that “The principles underlying [earlier] cases are instructive when the question is whether a patent claiming the combination of elements of prior art is obvious. When a work is available in one field of endeavor, design incentives and other market forces can prompt variations of it, either in the same field or a different one. If a person of ordinary skill can implement a predictable variation, § 103 likely bars its patentability.” KSR Int'l v. Teleflex lnc., 127 S. Ct. 1727, 1740 (2007).
Applying the KSR standard to Reiman and Mills, Examiner concludes that the combination of Reiman and Mills represents the use of known techniques to improve similar methods. Both Reiman and Mills are directed to microbiome classification from metagenomic data. Reiman only disclosed computing a phylogenetic tree, mapping the tree to a data structure assuming constant distance between nodes, and feeding the data structure into a machine learning model. In the same field of research, Mills provided microbiome classification calculating the biological distance (or dissimilarity) between two different microbial communities, for the purpose of predicting state of the medical condition. Combining the distance/dissimilarity calculation of Mills with taxonomy hierarchy representation of Reiman would have allowed reordering taxa before convolution. One ordinary skilled in the art before he effective filing data of the claimed invention would have had a reasonable expectation of success at combining Reiman’s phylogenetic tree embedding approach with Mills classification framework. This combination would have been expected to have provided more accurate results. Therefore, the invention would have been prima facie obvious to one of skill in the art before the effective filing date of the claimed invention, absent evidence to the contrary.
Claims 7 and 8 are rejected under 35 U.S.C. 103 as being unpatentable over Reiman in view of Mills, as applied to claims 1-3, 5-6, 9-15, 17-19, and 21-22 above, and further in view of Fioravanti (Phylogenetic convolutional neural networks in metagenomics, BMC Bioinformatics 2018, 19 (Suppl 2):49, pages: 1-54; as cited in the attached 892 form).
Claims 7 and 8 depend on claims 1, 2, and 6. Limitations of said claims are taught in the above rejections.
Regarding claim 7, Reiman and Mills do not teach that the similarity feature is computed based on Euclidean distances. Fioravanti teaches this limitation.
Fioravanti teaches a novel deep learning architecture for the classification of metagenomics data based on the Convolutional Neural Networks, with the patristic distance defined on the phylogenetic tree being used as the proximity measure. The patristic distance between variables is used together with a sparsified version of Multidimensional scaling to embed the phylogenetic tree in a Euclidean space (Fioravanti: Abstract).
Fioravanti further teaches map the discrete space of the set of leaves into an Euclidean space of a priori chosen dimension, by associating each leaf to a point P i in the Euclidean space with variable Euclidean coordinates preserving the tree distance (Fioravanti: pg. 3, col. 1, para. 1).
Regarding claim 8, the recited building a dendrogram based on the Euclidean distances computed for a hierarchical clustering of the representations of the microbes according to frequency, is taught as, computing the patristic distance between two leaves in a tree (Fioravanti: pg. 3, col. 1, last para.; Fig. 1).
Rationale for combining Reiman, Mills, and Fioravanti:
Applying the KSR standard to Reiman, Mills, and Fioravanti, Examiner concludes that the combination of Reiman, Mills, and Fioravanti represents applying a known technique to a known method. Reiman, Mills, and Fioravanti are directed to microbiome classification from metagenomic data. Reiman and Mills disclosed computing a taxonomic hierarchy-aware phylogenetic tree, mapping the tree to a data structure, and feeding the data structure into a machine learning model. In the same field of research, Fioravanti provided the known technique of Euclidean distance-mapping, to treat evolutionary close bacteria as spatial neighbors. Combining the multiclass disease classification of Reiman and Mills with the known Euclidean distance-mapping of Fioravanti would have allowed a convolutional neural network to operate on taxonomic data. One ordinary skilled in the art before he effective filing data of the claimed invention would have had a reasonable expectation of success at combining Reiman and Mills’s classification approach with Fioravanti’s known mapping technique. This combination would have been expected to have provided an improved deep learning model to diagnose multiple overlapping conditions from complex microbiome data. Therefore, the invention would have been prima facie obvious to one of skill in the art before the effective filing date of the claimed invention, absent evidence to the contrary.
Claim 16 is rejected under 35 U.S.C. 103 as being unpatentable over Reiman in view of Mills, as applied to claims 1-3, 5-6, 9-15, 17-19, and 21-22 above, and further in view of Zhang (BCM3D 2.0: accurate segmentation of single bacterial cells in dense biofilms using computationally generated intermediate image representations, biofilms and microbiome, 18 December 2022, pages 1-13; as cited in the attached 892 form).
Claim 16 depends on claim 1. Limitations of claim 1 are taught in the above rejections.
Regarding claim 16, Khan and Reiman do not teach combining image representation of plurality of trees into a 3D image depicting temporal data, and the 3D image is fed into a 3D-CNN implementation of the machine learning model. Zhang teaches an integrated image analysis pipeline that combines deep learning with conventional image analysis to detect and segment single biofilm-dwelling cells in 3D fluorescence images; training CNNs to translate 3D fluorescence images into intermediate 3D image representations (Ahang: abstract).
Rationale for combining Reiman, Mills, and Zhang:
Applying the KSR standard to Reiman, Mills, and Zhang, Examiner concludes that the combination of Reiman, Mills, and Zhang represents applying a known technique to a known method. Reiman, Mills, and Zhang are directed to microbial classification. Reiman and Mills disclosed computing a taxonomic hierarchy-aware phylogenetic tree, mapping the tree to a data structure, and feeding the data structure into a machine learning model. In the same field of research, Zhang provided the known technique of employing a 3D-CNN to 3D image into an intermediate image representation to measure microbial cells. Combining the multiclass disease classification of Reiman and Mills with the known 3D- CNN technique of Zhang would have allowed a convolutional neural network to process multi-level data and preserve evolutionary tree while fixing data sparsity. One ordinary skilled in the art before he effective filing data of the claimed invention would have had a reasonable expectation of success at combining Reiman and Mills’s classification approach with Zhang’s known 3D-CNN. This combination would have been expected to have provided an improved deep learning model to diagnose conditions from complex microbiome data. Therefore, the invention would have been prima facie obvious to one of skill in the art before the effective filing date of the claimed invention, absent evidence to the contrary.
Claim 19 is rejected under 35 U.S.C. 103 as being unpatentable over Reiman in view of Mills, as applied to claims 1-3, 5-6, 9-15, 17-18, and 21-22 above, and further in view of Ditzler (Multi-Layer and Recursive Neural Networks for Metagenomic Classification, IEEE Transactions on Nanobaioscience, Vol. 14, No. 6, September 2015; As cited in the IDS form 11/06/2023).
Claim 20 depends on claim 1. Limitations of claim 1 are taught in the above rejections.
Regarding claim 19, Mills teaches plotting the relative abundance and SHAP value for each sample of the top 25 features plotted in decreasing feature importance from the gradient boosting classifier model trained on the aggregate data at the ASV level. generating a cluster heatmap colored by the SHAP values of the top 25 features from the gradient boosting classifier model trained and tested on the aggregate data and each cohort independently. Clustering of both cohorts and features was performed using hierarchical clustering of Euclidean distances (Mills: [0019-0020]).
Reiman and Mills do not teach recursively ordering the log-normalized ASV frequencies according to at least one similarity feature computed based on Euclidean distances. This limitation is taught by Ditzler.
Ditzler teaches a multi-layer and recursive neural networks for metagenomic classification. Ditzler further teaches a deep learning approach for parsing natural scenes and language using recursive neural networks, which allow for structure prediction; applying PCoA with the Hellinger distance to the inner nodes of the trees generated with the recursive neural network (Ditzler: abstract, pg. 612, Subsection B. and C.; Figure. 4 and 5).
Rationale for combining Reiman, Mills, and Ditzler:
Applying the KSR standard to Reiman, Mills, and Ditzler, Examiner concludes that the combination of Reiman, Mills, and Ditzler represents applying a known technique to a known method. Reiman, Mills, and Ditzler are directed to microbial classification. Reiman and Mills disclosed computing a taxonomic hierarchy-aware phylogenetic tree, mapping the tree to a data structure, and feeding the data structure into a machine learning model. In the same field of research, Ditzler provided the known technique of employing a recursive neural network to classify metagenomic sample phenotypes. Combining the multiclass disease classification of Reiman and Mills with the known recursive neural network of Ditzler would have allowed a convolutional neural network to produce a hierarchical relationship of samples that can be visualized as a tree. One ordinary skilled in the art before he effective filing data of the claimed invention would have had a reasonable expectation of success at combining Reiman and Mills’s classification approach with Ditzler known recursive neural network. This combination would have been expected to have provided an improved deep learning model to diagnose conditions from complex microbiome data. Therefore, the invention would have been prima facie obvious to one of skill in the art before the effective filing date of the claimed invention, absent evidence to the contrary.
Claim 20 is rejected under 35 U.S.C. 103 as being unpatentable over Reiman in view of Mills, as applied to claims 1-3, 5-6, 9-15, 17-19, and 21-22 above, and further in view of Pinto (Gestational diabetes is driven by microbiota-induced inflammation months before diagnosis, Gut Microbiota, 10 January 2023, pages: 918-928; as cited in the IDS form dated 11/06/2023).
Claim 16 depends on claim 1. Limitations of claim 1 are taught in the above rejections.
Regarding claim 20, Khan teaches that the disease classification from microbial metagenome where the medical state of the subject is type II diabetes or impaired glucose tolerance. Khan and Reiman do not teach that the medical state of the subject is an indication of a complication that arises during pregnancy that is linked to lifelong health risks, including at least one of: gestational diabetes, preeclampsia, preterm birth, and postpartum depression. Pinto teaches that gestational diabetes is driven by microbiota-induced inflammation. Pinto further teaches developing a model to predict gestational diabetes using a machine learning model (Pinto: abstract).
Rationale for combining Reiman, Mills, and Pinto:
Applying the KSR standard to Reiman, Mills, and Zhang, Examiner concludes that the combination of Reiman, Mills, and Pinto represents simple substitution of one known element for another. Reiman, Mills, and Pinto are directed to microbiota-induced conditions. Reiman and Mills disclosed computing a taxonomic hierarchy-aware phylogenetic tree, mapping the tree to a data structure, and feeding the data structure into a machine learning model. Pinto stated that conditions such as gestational diabetes are driven by microbiota-induced inflammation. It would have been prima facie obvious to ne ordinary skilled in the art to use the gestational diabetes related microbiome to predict the state of the disease since early diagnosis is effective in reducing incidence and associated short-term and long-term morbidities, as stated by Pinto (abstract).
Conclusion
No claims are allowed.
Any inquiry concerning this communication or earlier communications from the examiner should be directed to GHAZAL SABOUR whose telephone number is (703)756-1289. The examiner can normally be reached M-F 7:30-5:00.
Examiner interviews are available via telephone, in-person, and video conferencing using a USPTO supplied web-based collaboration tool. To schedule an interview, applicant is encouraged to use the USPTO Automated Interview Request (AIR) at http://www.uspto.gov/interviewpractice.
If attempts to reach the examiner by telephone are unsuccessful, the examiner’s supervisor, Larry D. Riggs can be reached at (571) 270-3062. 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.
/G.S./Examiner, Art Unit 1686
/LARRY D RIGGS II/Supervisory Patent Examiner, Art Unit 1686
Prosecution Insights