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 .
Remarks
In response to communications sent January 27, 2023, claim(s) 1-20 are pending in this application; of these claims 1, 8, and 19 are in independent form.
Drawings
Figure 1 should be designated by a legend such as --Prior Art-- because only that which is old is illustrated. See MPEP § 608.02(g). Corrected drawings in compliance with 37 CFR 1.121(d) are required in reply to the Office action to avoid abandonment of the application. The replacement sheet(s) should be labeled “Replacement Sheet” in the page header (as per 37 CFR 1.84(c)) so as not to obstruct any portion of the drawing figures. If the changes are not accepted by the examiner, the applicant will be notified and informed of any required corrective action in the next Office action. The objection to the drawings will not be held in abeyance.
Information Disclosure Statement
The Information Disclosure Statement(s) is/are acknowledged and the references contained therein have been considered by the Examiner. This includes the Information Disclosure Statements(s) filed on: February 9, 2023 and August 3, 2023.
Claim Analysis - 35 USC § 101
The independent claims contain an abstract idea of determining pathogenicity. However, the additional elements are not routine and conventional. The additional elements involve a structural rendition of a protein and a plurality of images of the structural rendition as the basis of a computation. The examiner found only two recent references that involve a plurality of images (e.g., vantage points) of a protein as the basis of a computation. These are CN 112330825 A and CN 111414802 A. Normally, the structural rendition itself, such as 3D coordinates available in the rendition, would be the basis of a computation so that structural information is not lost. The instant set of claims differs from the conventional practice by using a plurality of images (e.g., vantage points) because it is suited to algorithms involving convolutional neural networks. While there are many references involving analyzing vantage points (e.g., furniture in a room), this is not typically done for protein structures.
Claim Rejections - 35 USC § 103
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.
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.
Claim(s) 1- 20 is/are rejected under 35 U.S.C. 103 as being unpatentable over CN111414802A (“Wei”) in view of “Won”, which refers to: Won, Dhong-Gun, et al. "3Cnet: pathogenicity prediction of human variants using multitask learning with evolutionary constraints." Bioinformatics 37.24 (2021): 4626-4634.
Note: machine-translations of the Wei reference are provided with the Office Action. The machine translation used in the rejection is the one by Google Translate rather than the one by Clarivate, which is also in the file wrapper.
The Won reference was cited on Applicant’s information disclosure statement.
As to claim 1, Wei teaches a system comprising:
at least one processor (Wei Para [0006]-[0007]: computerization); and
a non-transitory computer readable medium comprising instructions that, when executed by the at least one processor, cause the system (Wei Para [0006]-[0007]: computerization) to:
access a structural rendition of amino acids of a protein (Wei Para [0007]: “Preprocessing an original three-dimensional model of a protein—including data type conversion and data size standardization—to obtain a preprocessed three-dimensional model” (machine-translated));
capture a plurality of images of those parts of the structural rendition that contain a target amino acid from the amino acids (Wei Para [0007]: “Acquiring multiple two-dimensional views of the preprocessed three-dimensional model” (machine-translated)); and
based at least in part on the plurality of images, determine (Wei Para [0007]: “extracting an image feature matrix for each of said two-dimensional views, and fusing all the image feature matrices to obtain a two-dimensional feature matrix of the protein” (machine-translated))” …
However, Wei does not teach:
based at least in part on the plurality of images, determine pathogenicity of a nucleotide variant that mutates the target amino acid into an alternate amino acid.
Nevertheless, Won teaches:
based at least in part on the plurality of images determine pathogenicity of a nucleotide variant that mutates the target amino acid into an alternate amino acid (Won p. 4628 section 3.3 “structure-based” predictions about proteins and amino-acid substitutions of single-nucleotide variants; Figure 3 illustrates using features to determine pathogenicity).
Wei and Won are in the same field of bioinformatics. Furthermore, it would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to modify the teachings of Wei to include the teachings of Won because using protein structure context improves the sensitivity of variant prediction (See Won’s abstract). There would be a reasonable expectation of success because no particular accuracy of the variant determination is claimed; thus, inputting the feature information outputted by Wei into the neural network of Won would be predictive to some extent, thus falling within the scope of the claim.
As to claim 2, Wei in view of Won teaches the system of claim 1, wherein images in the plurality of images are captured from multiple perspectives (Wei Para [0029]: “During the processing stage, multiple two-dimensional views of the protein are generated by a rendering engine; this process is analogous to using a virtual camera to capture images of the preprocessed three-dimensional model while rotating it. Based on the specific viewing angles—for instance, by extracting one view at every 30-degree interval—a total of 12 two-dimensional views are captured.” (machine-translated)).
As to claim 3, Wei in view of Won teaches the system of claim 2, wherein the images are captured from one or more of multiple points of view (Wei Para [0029]: “During the processing stage, multiple two-dimensional views of the protein are generated by a rendering engine” (machine-translated)), multiple orientations, multiple positions, or multiple zoom levels (this elements are claimed in the alternative and do not need to be mapped, although some are overlapping in scope).
As to claim 4, Wei in view of Won teaches the system of claim 1, further comprising instructions that, when executed by the at least one processor, cause the system to:
activate a feature configuration of the structural rendition prior to capturing the plurality of images (Wei Para [0007]: “extracting an image feature matrix for each of said two-dimensional views, and fusing all the image feature matrices to obtain a two-dimensional feature matrix of the protein” (machine-translated)); and
capture the plurality of images of the parts of the structural rendition with the activated feature configuration (Wei Para [0007]: “Fusing and computing the two-dimensional feature matrix and the three-dimensional feature matrix of the protein to obtain a protein data feature matrix” (machine-translated)).
As to claim 5, Wei in view of Won teaches the system of claim 4, wherein the feature configuration comprises displaying one or more of a residue detail of the amino acids (Wei Para [0029]: “multiple two-dimensional views of the protein are generated by a rendering engine” (machine-translated)), an atomic detail of the amino acids, a polymer detail of the amino acids, a ligand detail of the amino acids, a chain detail of the amino acids, surface effects of the amino acids (Wei Para [0028]: “The two-dimensional views obtained in Step S2 cover at least the entire outer surface of the preprocessed three-dimensional model” (machine-translated)), density maps of the amino acids, supramolecular assemblies of the amino acids, sequence alignments of the amino acids, docking results of the amino acids, trajectories of the amino acids, conformational ensembles of the amino acids, secondary structures of the amino acids, tertiary structures of the amino acids, or quaternary structures of the amino acids (these elements are claimed in the alternative and do not all need to be mapped).
As to claim 6, Wei in view of Won teaches the system of claim 5, wherein the surface effects include transparency (this element is a limitation to an element claimed in the alternative in claim 5, and therefore does not need to be mapped), and color coding for electrostatic and hydrophobic values (this element is a limitation to an element claimed in the alternative in claim 5, and therefore does not need to be mapped; nevertheless see Wei Para [0030]: “a rendering engine—centered around the Phong shading model—is then utilized to generate the rendered views of the protein from this polygonal mesh”).
As to claim 7, Wei in view of Won teaches the system of claim 4, wherein the feature configuration comprises:
setting one or more of a zoom factor for displaying the structural rendition (Wei Para [0027]: “converting the original three-dimensional model into a file format suitable for visualization and reading, and scaling the data size of the original three-dimensional model—either by reduction or enlargement—to convert it to a predetermined standard value” (machine-translated)), a lighting condition for displaying the structural rendition (Wei Para [0030]: “a rendering engine—centered around the Phong shading model—is then utilized to generate the rendered views of the protein from this polygonal mesh” (machine-translated)), or a visibility range for displaying the structural rendition; or
color coding the structural rendition by one or more of the amino acids, by evolutionary conservations of the amino acids (this element is claimed in the alternative), or by structural qualities of the amino acids (Wei Para [0030]: “a rendering engine—centered around the Phong shading model—is then utilized to generate the rendered views of the protein from this polygonal mesh” (machine-translated); the Examiner interprets that the shading determined by the Phong model is based on structural qualities such as atom positions).
As to claim 8, Wei teaches teaches a computer-implemented method (Wei Para [0006]-[0007]: computerization) of determining pathogenicity of variants (the preamble is not interpreted as a step of the method), including:
accessing a structural rendition of amino acids of a protein (Wei Para [0007]: “Preprocessing an original three-dimensional model of a protein—including data type conversion and data size standardization—to obtain a preprocessed three-dimensional model” (machine-translated);
capturing a plurality of images of those parts of the structural rendition that contain a target amino acid from the amino acids (Wei Para [0007]: “Acquiring multiple two-dimensional views of the preprocessed three-dimensional model” (machine-translated)); and
based at least in part on the plurality of images, determining (Wei Para [0007]: “extracting an image feature matrix for each of said two-dimensional views, and fusing all the image feature matrices to obtain a two-dimensional feature matrix of the protein” (machine-translated))” …
However, Wei does not teach:
based at least in part on the plurality of images, determining pathogenicity of a nucleotide variant that mutates the target amino acid into an alternate amino acid.
Nevertheless, Won teaches:
based at least in part on the plurality of images, determining pathogenicity of a nucleotide variant that mutates the target amino acid into an alternate amino acid (Won p. 4628 section 3.3 “structure-based” predictions about proteins and amino-acid substitutions of single-nucleotide variants; Figure 3 illustrates using features to determine pathogenicity).
Wei and Won are in the same field of bioinformatics. Furthermore, it would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to modify the teachings of Wei to include the teachings of Won because using protein structure context improves the sensitivity of variant prediction (See Won’s abstract). There would be a reasonable expectation of success because no particular accuracy of the variant determination is claimed; thus, inputting the feature information outputted by Wei into the neural network of Won would be predictive to some extent, thus falling within the scope of the claim.
As to claim 9, Wei in view of Won teaches the computer-implemented method of claim 8, wherein a pathogenicity determiner determines the pathogenicity of the nucleotide variant by processing, as input, the plurality of images, and generating, as output, a pathogenicity score for the alternate amino acid (Won p. 4628 section 3.3 “structure-based” predictions about proteins and amino-acid substitutions of single-nucleotide variants; Figure 3 illustrates using features to determine pathogenicity; Won teaches scores for predictions in Table 1 on page 4633; Wei teaches the nature of the “structure-based” predictions in this combination of teachings).
As to claim 10, Wei in view of Won teaches the computer-implemented method of claim 9, wherein the pathogenicity determiner is a convolutional neural network or another neural network (Won Abstract: 3Cnet, the pathogenicity detector based on features, uses a neural network; the features are taught in Wei Para [0032], which uses a convolutional neural network to determine the features).
As to claim 11, Wei in view of Won teaches the computer-implemented method of claim 10, wherein the pathogenicity determiner processes respective images in the plurality of images through respective convolutional neural networks to generate respective feature maps for the respective images (Wei Para [0031]: “The method for extracting the image feature matrices of the respective 2D views in Step S2 is as follows: a Convolutional Neural Network is utilized to extract the image feature matrices of the 2D views individually. After extracting multiple 2D views of the original 3D model, a Convolutional Neural Network (CNN) is employed to extract the features of the aforementioned 12 2D views separately.” (machine-translated)).
As to claim 12, Wei in view of Won teaches the computer-implemented method of claim 11, wherein the respective convolutional neural networks are respective instances of a same convolutional neural network (Won Abstract: 3Cnet, the pathogenicity detector based on features, uses a neural network; the features are taught in Wei Para [0032], which uses a convolutional neural network to determine the features).
As to claim 13, Wei in view of Won teaches the computer-implemented method of claim 12, wherein the pathogenicity determiner combines the respective feature maps into a pooled feature map, and processes the pooled feature map through a final convolutional neural network to generate the pathogenicity score for the alternate amino acid (Wei Para [0032]: “By extracting image features from each 2D view, a unified CNN (Convolutional Neural Network) architecture—incorporating a view-pooling layer—is employed to combine information derived from multiple 2D views, thereby generating a single, compact feature descriptor that characterizes the protein.” (machine-translated)).
As to claim 14, Wei in view of Won teaches the computer-implemented method of claim 13, wherein the respective images in the plurality of images are combined into a combined representation “By extracting image features from each 2D view, a unified CNN (Convolutional Neural Network) architecture—incorporating a view-pooling layer—is employed to combine information derived from multiple 2D views, thereby generating a single, compact feature descriptor that characterizes the protein.” (machine-translated)) for processing by the pathogenicity determiner (Won p. 4628 section 3.3 “structure-based” predictions about proteins and amino-acid substitutions of single-nucleotide variants; Figure 3 illustrates using features to determine pathogenicity).
As to claim 15, Wei in view of Won teaches the computer-implemented method of claim 14, wherein the respective images are arranged as respective color channels in the combined representation (Wei Para [0034]: “To facilitate the fusion of multi-view feature information derived from the 3D model, a "view-pooling layer" is introduced within this method; this layer serves to aggregate all the individual image feature matrices, thereby combining the information contained across the multiple 2D views of the 3D model into a single, compact object feature representation.” (machine-translated)).
As to claim 16, Wei in view of Won teaches the computer-implemented method of claim 15, wherein intensity values in the respective images are pixel-wise summed into pixels of the combined representation (Wei Para [0035]: “The specific method for fusing all the image feature matrices in Step S2 is as follows: the values located at corresponding positions across all the individual image feature matrices are compared; the maximum value identified at each specific position is then selected as the resultant value for that position, thereby yielding a single 2D matrix that serves as the 2D feature matrix for the protein.” (machine-translated); the Examiner argues that “summing” is a simple substitution that replaces “finding a maximum value” when combining data in a pointwise fashion).
As to claim 17, Wei in view of Won teaches the computer-implemented method of claim 8, wherein the parts of the structural rendition contain the target amino acid and some additional amino acids adjacent to the target amino acid acids (Wei Para [0007]: “Acquiring multiple two-dimensional views of the preprocessed three-dimensional model” (machine-translated), which implies that adjacent amino acids would be adjacent in a structural rendition derived from 3D locations).
As to claim 18, Wei in view of Won teaches the computer-implemented method of claim 8, wherein the structural rendition is a three-dimensional (3D) structural rendition (Wei Para [0007]: “Fusing and computing the two-dimensional feature matrix and the three-dimensional feature matrix of the protein to obtain a protein data feature matrix” (machine-translated)).
As to claim 19, Wei teaches a non-transitory computer-readable medium storing instructions that, when executed by at least one processor, cause a computing device to:
access a structural rendition of amino acids of a protein (Wei Para [0007]: “Preprocessing an original three-dimensional model of a protein—including data type conversion and data size standardization—to obtain a preprocessed three-dimensional model” (machine-translated));
capture a plurality of images of those parts of the structural rendition that contain a target amino acid from the amino acids (Wei Para [0007]: “Acquiring multiple two-dimensional views of the preprocessed three-dimensional model” (machine-translated)); and
based at least in part on the plurality of images, determine (Wei Para [0007]: “extracting an image feature matrix for each of said two-dimensional views, and fusing all the image feature matrices to obtain a two-dimensional feature matrix of the protein” (machine-translated))” …
However, Wei does not teach:
based at least in part on the plurality of images, determine pathogenicity of a nucleotide variant that mutates the target amino acid into an alternate amino acid.
Nevertheless, Won teaches:
based at least in part on the plurality of images, determine pathogenicity of a nucleotide variant that mutates the target amino acid into an alternate amino acid (Won p. 4628 section 3.3 “structure-based” predictions about proteins and amino-acid substitutions of single-nucleotide variants; Figure 3 illustrates using features to determine pathogenicity).
Wei and Won are in the same field of bioinformatics. Furthermore, it would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to modify the teachings of Wei to include the teachings of Won because using protein structure context improves the sensitivity of variant prediction (See Won’s abstract). There would be a reasonable expectation of success because no particular accuracy of the variant determination is claimed; thus, inputting the feature information outputted by Wei into the neural network of Won would be predictive to some extent, thus falling within the scope of the claim.
As to claim 20, Wei in view of Won teaches the non-transitory computer-readable medium of claim 19, further storing instructions that, when executed by the at least one processor, cause the computing device to determine the pathogenicity of the nucleotide variant by utilizing a neural-network-based pathogenicity determiner to process, as input, the plurality of images, and generate, as output, a pathogenicity score for the alternate amino acid (Won p. 4628 section 3.3 “structure-based” predictions about proteins and amino-acid substitutions of single-nucleotide variants; Figure 3 illustrates using features to determine pathogenicity; Won teaches scores for predictions in Table 1 on page 4633; Wei teaches the nature of the “structure-based” predictions in this combination of teachings).
Conclusion
The prior art made of record and not relied upon is considered pertinent to applicant's disclosure.
US-20200098450-A1 pertinence: plurality of angles and predicting activity of compound, but doesn't use neural network
US-20180365372-A1 pertinence: interpretation of variants using deep learning
US-20020015038-A1 pertinence: finding pockets in a protein model
CN 112330825 A title: “A Three-dimensional Model Searching Method Based On Two-dimensional Image Information”
Any inquiry concerning this communication or earlier communications from the examiner should be directed to Jesse P Frumkin whose telephone number is (571)270-1849. The examiner can normally be reached Monday - Friday, 10-5 ET.
Examiner interviews are available via telephone, in-person, and video conferencing using a USPTO supplied web-based collaboration tool. To schedule an interview, applicant is encouraged to use the USPTO Automated Interview Request (AIR) at http://www.uspto.gov/interviewpractice.
If attempts to reach the examiner by telephone are unsuccessful, the examiner’s supervisor, Olivia Wise can be reached at (571) 272-2249. The fax phone number for the organization where this application or proceeding is assigned is 571-273-8300.
Information regarding the status of published or unpublished applications may be obtained from Patent Center. Unpublished application information in Patent Center is available to registered users. To file and manage patent submissions in Patent Center, visit: https://patentcenter.uspto.gov. Visit https://www.uspto.gov/patents/apply/patent-center for more information about Patent Center and https://www.uspto.gov/patents/docx for information about filing in DOCX format. For additional questions, contact the Electronic Business Center (EBC) at 866-217-9197 (toll-free). If you would like assistance from a USPTO Customer Service Representative, call 800-786-9199 (IN USA OR CANADA) or 571-272-1000.
/JESSE P FRUMKIN/ Primary Examiner, Art Unit 1685 June 8, 2026
Prosecution Insights