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 .
Claims Status
Claims 1-20 are pending.
Claims 1-20 are examined.
Priority
The instant application is a continuation of PCT/CN2022/116095, filed 08/31/2022, which claims priority to Chinese Application No. 202111213797.X, filed 10/19/2021. Therefore, the Effective Filing Date (EFD) assigned to each of the claims 1-20 is the filing date of Chinese Application No. 202111213797.X, filed 10/19/2021.
Information Disclosure Statement
The Information Disclosure Statements filed 02/128/2024 is in compliance with the provisions of 37 CFR 1.97 and has therefore been considered. A signed copy of the IDS document is included with this Office Action.
Drawings
The drawings filed 05/10/2023 are accepted.
Specification
The disclosure is objected to because of the following informalities:
In paragraph [0012], “schematic diagram of another molecules of the method” should read “schematic diagram of another molecule
In paragraph [0070], “cavity of an sample alternative atom” should read “cavity of a
Appropriate correction is required.
Claim Objections
Claim 9 is objected to because of the following informalities:
In claim 9, “according to claim 1, , further comprising” should read “according to claim 1,
Appropriate correction is required.
Claim Rejections - 35 USC § 112
The following is a quotation of 35 U.S.C. 112(b):
(b) CONCLUSION.—The specification shall conclude with one or more claims particularly pointing out and distinctly claiming the subject matter which the inventor or a joint inventor regards as the invention.
The following is a quotation of 35 U.S.C. 112 (pre-AIA ), second paragraph:
The specification shall conclude with one or more claims particularly pointing out and distinctly claiming the subject matter which the applicant regards as his invention.
Claims 2-5, 8-10, 12-15, 18, and 19 are rejected under 35 U.S.C. 112(b) or 35 U.S.C. 112 (pre-AIA ), second paragraph, as being indefinite for failing to particularly point out and distinctly claim the subject matter which the inventor or a joint inventor (or for applications subject to pre-AIA 35 U.S.C. 112, the applicant), regards as the invention.
With respect to claims 2 and 12, the claims recite the limitation of “obtaining training samples, the training sample comprising”. The claims are indefinite because there is no antecedent basis for “the training sample”. It is unclear if this refers to each individual sample in the training samples or instead referring to the training samples as a whole.
With respect to claims 4 and 14, the claims recite the limitation of “a binding distance between the sample amino acid molecule and the sample alternative atom”. The claims are indefinite because there is no antecedent basis for “the sample alternative atom” and thus it is unclear what is being predicted.
With respect to claims 8 and 18, the claims recite the limitation of “obtaining training samples, the training sample comprising”. The claims are indefinite because there is no antecedent basis for “the training sample”. It is unclear if this refers to each individual sample in the training samples or instead referring to the training samples as a whole.
With respect to claims 9 and 19, the claims recite the limitation of “adjusting model parameters of the molecular binding model during determining that the training loss of the molecular binding model does not meet the training target”. The claims are indefinite because there is no antecedent basis for a step of “determining that the training loss of the molecular binding model does not meet the training target”. Thus, it is unclear when the model parameters are adjusted.
With respect to claim 10, the claim recites the limitation of “when the training losses comprise…”. The claim is indefinite because there is no antecedent basis for “the training losses”. This ambiguity would be corrected if claim 9, on which claim 10 is dependent, was dependent on claim 6, or if claim 10 is made dependent upon claim 6.
The remaining claims are rejected due to being dependent upon indefinite claims without remedying the indefiniteness.
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 1-20 are rejected under 35 U.S.C. 101 because the claimed inventions are directed to an abstract idea of mental steps, mathematic concepts, or a natural law without significantly more.
The MPEP at MPEP 2106.03 sets forth steps for identifying eligible subject matter:
(1) Are the claims directed to a process, machine, manufacture or composition of
matter?
(2A)(1) Are the claims directed to a judicially recognized exception, i.e. a law of nature,
a natural phenomenon, or an abstract idea?
(2A)(2) If the claims are directed to a judicial exception under Prong One, then is the
judicial exception integrated into a practical application?
(2B) If the claims are directed to a judicial exception and do not integrate the judicial
exception, do the claims provide an inventive concept?
With respect to step (1): Yes, the claims are directed to a method, a device, and a non-transitory computer readable storage medium.
With respect to step (2A)(1): The claims are directed to abstract ideas of mathematical concepts and mental processes.
“Claims directed to nothing more than abstract ideas (such as a mathematical formula or equation), natural phenomena, and laws of nature are not eligible for patent protection” (MPEP 2106.04). Abstract ideas include mathematical concepts (mathematical formulas or equations, mathematical relationships and mathematical calculations), certain methods of organizing human activity, and mental processes (procedures for observing, evaluating, analyzing/judging and organizing information (MPEP 2106.04(a)(2)). Laws of nature or natural phenomena include naturally occurring principles/relations that are naturally occurring or that do not have markedly different characteristics compared to what occurs in nature (MPEP 2106(b)).
Mathematical concepts recited in claims 1, 11, and 20:
inputting the protein feature information and the molecular feature information into a molecular binding model, and using the molecular binding model to determine binding activity feature information, embedding feature information and eutectic feature information between the sample protein molecule and the sample alternative molecule, the binding activity feature information characterizing activities of the sample protein molecule and the sample alternative molecule after virtual binding, the embedding feature information characterizing a degree of binding between the sample protein molecule and the sample alternative molecule, and the eutectic feature information characterizing whether a eutectic structure exists between the sample protein molecule and the sample alternative molecule
Mental processes recited in claims 1, 11, and 20:
determining a training loss of the molecule binding model based on the binding activity feature information, the embedding feature information and the eutectic feature information
outputting the molecular binding model as a trained molecular binding model when the training loss of the molecular binding model meets a training target
Dependent claims 2-10 and 12-19 recite additional steps that either are directed to abstract ideas or further limit the judicial exceptions in independent claims 1 and 11, and as such, are further directed to abstract ideas. Hence, the claims explicitly recite numerous elements that individually and in combination constitute abstract ideas. The relevant recitations are:
Claims 2 and 12 “using a feature extraction model to perform feature extraction processing on a training sample to obtain the protein feature information of the sample protein molecule and the molecular feature information of the sample alternative molecule”
Claim 3 and 13: “determining an adjacency matrix of the sample protein molecule based on the sample protein molecule in the training sample, the adjacency matrix of the sample protein molecule characterizing sample amino acid molecules contained in the sample protein molecule and the molecular structure distance between every two sample amino acid molecules; determining an adjacency matrix of the sample alternative molecule based on the sample alternative molecule in the training sample, the adjacency matrix of the sample alternative molecule characterizing sample alternative atoms contained in the sample alternative molecule and chemical bond structures between the sample alternative atoms; and performing feature extraction processing on the adjacency matrix of the sample protein molecule and the adjacency matrix of the sample alternative molecule respectively to obtain the protein feature information of the sample protein molecule and the molecular feature information of the sample alternative molecule”
Claims 4 and 14: “predicting for a sample amino acid molecule of the sample amino acid molecules contained in the sample protein molecule, based on the protein feature information and the molecular feature information, using the molecular binding model, a binding distance between the sample amino acid molecular and the sample alternative atom contained in the sample alternative molecule after a virtual binding of the sample protein molecule and the sample alternative molecule so as to obtain a plurality of binding distances; and determining, based on the plurality of binding distances, the embedding feature information and the eutectic feature information between the sample protein molecule and the sample alternative molecule”
Claims 5 and 15: “determining, based on the binding distance with a minimum value between a specified sample amino acid molecule and the sample alternative atom, the embedding feature information between the sample protein molecule and the sample alternative molecule; the specified sample amino acid molecule being one of the sample amino acid molecules contained in the sample protein molecule; and determining, based on the sample amino acid molecule, the sample alternative atom, and a binding distance between the sample amino acid molecule and the sample alternative atom, the eutectic feature information between the sample protein molecule and the sample alternative molecule”
Claims 6 and 16: “determining, based on a first error value between the embedding feature information and an embedding target, a first training loss of the molecular binding model; determining, based on a second error value between the eutectic feature information and a eutectic target, a second training loss of the molecular binding model; determining, based on a third error value between the binding activity feature information and an activity target, a third training loss of the molecular binding model; and determining, based on the first training loss, the second training loss and the third training loss, the training loss of the molecular binding model”
Claims 7 and 17: “using a trained neighborhood consensus model to match a sample amino acid molecule contained in the sample protein molecular with a sample alternative atom contained in the sample alternative molecule to obtain matching feature information between the sample protein molecule and the sample alternative molecule as the eutectic target, the matching feature information characterizing a matching distance between the sample amino acid molecule and the sample alternative atom; using a cross entropy function to determine the second error value between the eutectic feature information and the eutectic target; and using the second error value as the second training loss of the molecular binding model”
Claim 8 and 18: “determining, based on the third error value between the binding activity feature information and the activity target, the third training loss of the molecular binding model, comprises: using an activity prediction model to predict, based on the binding activity feature information, a sample activity value of the sample protein molecule and the sample alternative molecule after virtual binding; and determining, based on the third error value between the sample activity value and the corresponding reference activity value, the third training loss of the molecular binding model”
Claims 9 and 19: “adjusting model parameters of the molecular binding model during determining that the training loss of the molecular binding model does not meet the training target”
Claim 10: “when the training losses comprise a first training loss, a second training loss and a third training loss, the first training loss being determined based on the embedding feature information, the second training loss being determined based on the eutectic feature information and the third training loss being determined based on the binding activity feature information, determining whether the first training loss converges, whether the second training loss converges, and whether the third training loss converges, respectively; and adjusting the model parameters of the molecular binding model when in the first training loss, the second training loss and the third training loss, there is at least one training loss that does not converge”
The abstract ideas in the claims are evaluated under Broadest Reasonable Interpretation (BRI) and determined herein to each cover mental processes and mathematic concepts because the claims recite no more than using machine learning, mathematical concepts, to analyze information and determine features of said data. Those features are used to make predictions. All the steps on the data are either able to be performed mentally, such as determining that the training loss meets a target, or are mathematical concepts because they comprise using an algorithm to calculate a number.
With respect to step (2A)(2): The claims must therefore be examined further to determine whether they integrate that abstract idea into a practical application (MPEP 2106.04(d)). The claimed additional elements are analyzed alone or in combination to determine if the judicial exception is integrated into a practical application (MPEP 2106.04(d).I.; MPEP 2106.05(a-h)). If the claim contains no additional elements beyond the judicial exception, the claim fails to integrate the abstract idea into a practical application (MPEP 2106.04(d).III).
Claims 1, 11, and 20 recite the following additional elements that are not abstract ideas:
method executed by a computer device
a computing device comprising one or more processors and one or more memories, the one or more memories storing at least one computer program, the at least one computer program being loaded and executed by one or more processors
a non-transitory computer readable storage medium, storing at least one computer program, the at least one computer program being loaded and executed by a processor
obtaining protein feature information of sample protein molecules and molecular feature information of sample alternative molecules
The step of obtaining protein feature information and molecular feature information is directed to data gathering because it gathers the data on which the judicial exceptions are performed. Data gathering does not impose any meaningful limitation on the abstract idea, or how the abstract idea is performed. Data gathering steps are not sufficient to integrate an abstract idea into a practical application (MPEP 2106.05(g)). The elements of a computer device executing the method, a computing device comprising one or more processors and one or more memories, and a non-transitory computer readable storage medium are all directed to a generic computer or elements of a generic computer. The courts have weighed in and consistently maintained that when, for example, a memory, display, processor, machine, etc. ... are recited so generically (i.e., no details are provided) that they represent no more than mere instructions to apply the judicial exception on a computer, and these limitations may be viewed as nothing more than generally linking the use of the judicial exception to the technological environment of a computer (see MPEP 2106.05(f)).
Dependent claims 2, 8, 12, and 18 recite further steps of data gathering and thus do not integrate the judicial exceptions into a practical application.
None of these dependent claims recite additional elements, alone or in combination, which would integrate a judicial exception into a practical application.
Lastly, the claims have been evaluated with respect to step (2B): Because the claims recite an abstract idea, and do not integrate that abstract idea into a practical application, the claims lack a specific inventive concept. Under said analysis, Applicant is reminded that the judicial exception alone cannot provide that inventive concept or practical application (MPEP 2106.05). Identifying whether the additional elements beyond the abstract idea amount to such an inventive concept requires considering the additional elements individually and in combination to determine if they provide significantly more than the judicial exception (MPEP 2106.05.A i-vi).
With respect to the instant claims, the additional elements described above do not rise to the level of significantly more than the judicial exception. As set forth in the MPEP at 2106.05(d).I, determinations of whether or not additional elements (or a combination of additional elements) may provide significantly more and/or an inventive concept rests in whether or not the additional elements (or combination of elements) represent well-understood, routine, conventional activity. Said assessment is made by a factual determination stemming from a conclusion that an element (or combination of elements) is widely prevalent or in common use in the relevant industry, which is determined by either a citation to an express statement in the specification or to a statement made by an applicant during prosecution that demonstrates a well-understood, routine or conventional nature of the additional element(s); a citation to one or more of the court decisions as discussed in MPEP 2106(d)(II) as noting the well-understood, routine, conventional nature of the additional element(s); a citation to a publication that demonstrates the well-understood, routine, conventional nature of the additional element(s); and/or a statement that the examiner is taking official notice with respect to the well-understood, routine, conventional nature of the additional element(s).
With respect to claims 1, 11, and 20: The additional elements of obtaining protein feature information of sample protein molecules and molecular feature information of sample alternative molecules, a computing device comprising one or more processors and one or more memories, the one or more memories storing at least one computer program, the at least one computer program being loaded and executed by one or more processors, and a non-transitory computer readable storage medium do not rise to the level of significantly more than the judicial exception. With respect to the computing device, as exemplified in the MPEP at 2106.05(f) with reference to Alice Corp. 573 US at 223, 110 USPQ2d at 1983 “claims that amount to nothing more than an instruction to apply the abstract idea using a generic computer do not render an abstract idea eligible”. Therefore, the device constitutes no more than a general link to a technological environment, which is insufficient to constitute an inventive concept that would render the claims significantly more than the abstract idea (see MPEP 2105(b)I-III). With respect to obtaining the feature information, the prior art to Kimber et al. (“Deep Learning in Virtual Screening: Recent Applications and Developments”) discloses that virtual screening is a prominent method for rationalizing and speeding up drug development (page 1, last paragraph). Kimber et al. discloses that virtual screening is an integral part of drug discovery and that the methods are often divided into two major categories, structure-based methods and ligand-based methods (page 2, paragraph 2). Kimber et al. discloses that the most commonly used structure-based technique is molecular docking which predicts one or several binding poses of a query ligand in the receptor structure and estimates their binding affinity (page 2, paragraph 3). As such, it is recognized that these additional limitations are routine, well understood, and conventional in the art. These limitations do not improve the functioning of a computer, or comprise an improvement to any other technical field, they do not require or set forth a particular machine, they do not affect a transformation of matter, nor do they provide a non-conventional or unconventional step. As such, these limitations fail to rise to the level of significantly more.
With respect to claims 2 and 12: The additional element of obtaining training samples, the training sample comprising the sample protein molecule and the sample alternative molecule does not rise to the level of significantly more than the judicial exception. The prior art to Kimber et al. discloses that virtual screening is a prominent method for rationalizing and speeding up drug development (page 1, last paragraph). Kimber et al. discloses that virtual screening is an integral part of drug discovery and that the methods are often divided into two major categories, structure-based methods and ligand-based methods (page 2, paragraph 2). Kimber et al. discloses that the most commonly used structure-based technique is molecular docking which predicts one or several binding poses of a query ligand in the receptor structure and estimates their binding affinity (page 2, paragraph 3). As such, it is recognized that these additional limitations are routine, well understood, and conventional in the art. These limitations do not improve the functioning of a computer, or comprise an improvement to any other technical field, they do not require or set forth a particular machine, they do not affect a transformation of matter, nor do they provide a non-conventional or unconventional step. As such, these limitations fail to rise to the level of significantly more.
With respect to claims 8 and 18: The additional element of obtaining tramples, the training sample comprising the sample protein molecule, the sample alternative molecule and a reference activity value of the sample protein molecule and the sample alternative molecule after virtual binding does not rise to the level of significantly more than the judicial exception. The prior art to Kimber et al. discloses that virtual screening is a prominent method for rationalizing and speeding up drug development (page 1, last paragraph). Kimber et al. discloses that virtual screening is an integral part of drug discovery and that the methods are often divided into two major categories, structure-based methods and ligand-based methods (page 2, paragraph 2). Kimber et al. discloses that the most commonly used structure-based technique is molecular docking which predicts one or several binding poses of a query ligand in the receptor structure and estimates their binding affinity (page 2, paragraph 3). As such, it is recognized that these additional limitations are routine, well understood, and conventional in the art. These limitations do not improve the functioning of a computer, or comprise an improvement to any other technical field, they do not require or set forth a particular machine, they do not affect a transformation of matter, nor do they provide a non-conventional or unconventional step. As such, these limitations fail to rise to the level of significantly more.
The claims have all been examined to identify the presence of one or more judicial exceptions. Each additional limitation in the claims has been addressed, alone and in combination, to determine whether the additional limitations integrate the judicial exception into a practical application. Each additional limitation in the claims has been addressed, alone and in combination, to determine whether those additional limitations provide an inventive concept which provides significantly more than those exceptions. Individually, the limitations of the claims and the claims as a whole have been found lacking.
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.
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, 2, 6-12, and 16-20 are rejected under 35 U.S.C. 103 as being unpatentable over Ford et al. (US 11929152 B1, filed 06/09/2020) in view of Chinta et al. (“Machine Learning Derived Quantitative Structure Property Relationship (QSPR) to Predict Drug Solubility in Binary Solvent Systems”).
Regarding claims 1, 11, and 20, Ford et al. teaches a method for training molecular binding models, executed by a computer device (Figure 11; column 12, line 61), comprising:
obtaining protein feature information of sample protein molecules and molecular feature information of sample alternative molecules (column 2, line 20);
inputting the protein feature information and the molecular feature information into a molecular binding model (column 2, line 20; line 38; column 9 , line 48), and using the molecular binding model to determine binding activity feature information (column 2, line 38), embedding feature information between the sample protein molecule and the sample alternative molecule (column 6, line 56-column 7, line 14), the binding activity feature information characterizing activities of the sample protein molecule and the sample alternative molecule after virtual binding, the embedding feature information characterizing a degree of binding between the sample protein molecule and the sample alternative molecule (column 9, line 52);
determining a training loss of the molecular binding model based on the binding activity feature information and the embedding feature information (column 7, line 42); and
outputting the molecular binding model as a trained molecular binding model when the training loss of the molecular binding model meets a training target: Ford et al. teaches a four-part loss function that uses cross-entropy being used to fit a sequence recovery task on the enzyme, the molecular adjacency matrix, the molecular node identity recovery task on the substrate, and the prediction of substrate-enzyme functionality, the final total loss being a weighted summation of the individual cross entropy losses (column 7, line 42) and training the model using self-supervised input recovery tasks coupled to a multi-task loss function (column 7, line 14), and teaches an output trained machine learning model (column 2, line 45).
Furthermore, Ford et al. teaches that the computer system includes one or more processors coupled to a system memory (column 12, line 66), such as non-transitory computer storage media (column 14, line 25).
Ford et al. does not teach the claim elements of using the molecular binding model to determine eutectic feature information between the sample protein molecule and the sample alternative molecule, the eutectic feature information characterizing whether a eutectic structure exists between the sample protein molecule and the sample alternative molecule.
However, Chinta et al. teaches machine learning derived quantitative structure property relationships to predict drug solubility in binary solvent systems. Chinta et al. teaches that prediction of drug solubility is a crucial problem in pharmaceutical industries for both drug delivery and discovery purposes (Abstract) because it is important to achieve the desired concentration of drug in circulation for achieving a required pharmacological response. As drugs can reach the receptors through aqueous media, aqueous soluble drugs are preferred for clinical purposes, thus leading to several approaches to increase drug solubility, such as usage of Deep Eutectic Solvents (page 3082, column 1). Chinta et al. teaches a machine learning approach for correlating drug solubility to structural features, comprising feature selection, prediction based on the identified features, and a prediction error based clustering approach (page 3083, column 2, last paragraph – page 3084, column 1, paragraph 1).
Therefore, it would have been prima facie obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to have incorporated the drug solubility eutectic feature information of Chinta et al. to the method of Ford et al. because Ford et al. is directed to predictions of enzyme and substrate combinations and interactions (column 2, line 1), and teaches that predicting new substrates for a given enzyme has potential applications in developing drug discovery panels (column 2, line 5). Chinta et al. that prediction of drug solubility is a crucial problem in pharmaceutical industries for both drug delivery and discovery purposes (Abstract) because it is important to achieve the desired concentration of drug in circulation for achieving a required pharmacological response. Chinta et al. teaches that drugs can reach the receptors through aqueous media, aqueous soluble drugs are preferred for clinical purposes, thus leading to several approaches to increase drug solubility (page 3082, column 1). Thus, one of ordinary skill in the art would have a reasonable expectation of success of success of predicting eutectic feature information as well as interaction and binding information by combining the prior art references and would be motivated to do so in order to develop possible drug panels with aqueous applications for preferred administration of the drug.
Regarding claims 2 and 12, Ford et al. teaches the method of claim 1 and the device of claim 11 in view of Chinta et al. Ford et al. also teaches obtaining training data comprising enzyme and substrate data and teaches extracting features from this data (column 7, line 16; Figure 3; Figure 4).
Regarding claims 6 and 16, Ford et al. teaches the method of claim 1 and the device of claim 11. Ford et al. teaches a four-part loss function that uses cross-entropy to fit a sequence recovery task on the enzyme, the molecular adjacency matrix, and the molecular node identify recovery task on the substrate, and the prediction of substrate-enzyme functionality. Ford et al. teaches that the final total loss is a weighted summation of the individual cross entropy losses (column 7, line 42).
Ford et al. does not teach the claim element of determining, based on a second error value between the eutectic feature information and a eutectic target, a second training loss of the molecular binding model.
However, Chinta et al. teaches predicting drug solubility in binary solvent systems using prediction error of the sample with respect to a cluster and teaches calculating the root-mean-square error of the model based on the prediction errors (page 3087, column 2).
Regarding claims 7 and 17, Ford et al. teaches the method of claim 6 and the device of claim 16 in view of Chinta et al. Ford et al. also teaches that the model is trained using self-supervised input recovery tasks coupled to a multi-task loss function and that an enzyme is matched to each substrate (column 7, line 15), and that from the functional pairs, matching of the set of available enzymes is made against the set of available substrates (column 7, line 30). Ford et al. teaches that a four-part loss function that uses cross-entropy is used to fit a sequence recovery task on the enzyme, the molecular adjacency matrix, and the molecular node identify recovery task on the substrate, and the prediction of substrate-enzyme functionality. Ford et al. teaches that the final total loss is a weighted summation of the individual cross entropy losses (column 7, line 42).
Ford et al. does not teach the claim elements of determining, based on the second error value between the eutectic feature information and the eutectic target, the second training loss of the molecular binding model, or using a trained neighborhood consensus model.
However, Chinta et al. teaches that K nearest neighbors is the most frequency used testing approach to find which model should be used to predict the output of a new data point (page 3087, column 1, paragraph 2). Chinta et al. also teaches that data separated out for K-fold validation is associated with averaged models based on their prediction errors, so that any new sample can use the data samples as neighbors for selecting a suitable model. Chinta et al. teaches that the final pair of models can be considered as global models for predicting drug solubility in binary solvent systems (page 3089, column 1, paragraph 3).
Regarding claims 8 and 18, Ford et al. teaches the method of claim 6 and the device of claim 16 in view of Chinta et al. Ford et al. also teaches obtaining training data comprising enzyme and substrate data including extracting substrate/product lines (column 7, line 16; Figure 3; Figure 4). Ford et al. teaches that the enzyme vector, substrate vector, and on/off functionality are concatenated and fed to a model to generate a representative enzyme and substrate and an indication of reactivity (column 7, line 6). Ford et al. also teaches using a four-part loss function that uses cross-entropy to fit a sequency recovery task on the enzyme, the molecular adjacency matrix, the molecular node identity recovery task on the substrate, and the prediction of substrate-enzyme functionality, and that the final total loss is weighted summation of the individual cross entropy losses (column 7, line 42).
Ford et al. does not teach the claim elements of a reference activity value and determining the error value between the sample activity value and the reference activity value.
However, Chinta et al. teaches comparing experimentally obtained data as reference data to predicted data (page 3091, column 2, Section Supporting Information).
Regarding claims 9 and 19, the claims are directed to adjusting model parameters of the molecular binding model during determining that the training loss of the molecular binding model does not meet the training target. Ford et al. teaches the method of claim 1 and the device of claim 11.
Ford et al. does not teach the claim elements of adjusting model parameters of the molecular binding model during determining that the training loss of the molecular binding model does not meet the training target.
However, Chinta et al. teaches model optimization wherein model parameters are adjusted and the calculation is terminated based on a criterion, but when the root-mean-square error is not less than a predefined limit, another iteration of the calculations is performed (page 3087, column 2).
Regarding claim 10, Ford et al. teaches the method of claim 9 and the device of claim 19 in view of Chinta et al. Ford et al. teaches a four-part loss function that uses cross-entropy to fit a sequence recovery task on the enzyme, the molecular adjacency matrix, and the molecular node identify recovery task on the substrate, and the prediction of substrate-enzyme functionality. Ford et al. teaches that the final total loss is a weighted summation of the individual cross entropy losses (column 7, line 42).
Ford et al. does not teach the claim element of adjusting the model parameters of the molecular binding model during determining that the training loss of the molecular binding model does not meet the training target.
However, Chinta et al. teaches a prediction error based clustering algorithm for drug solubility problems wherein the model is initialized with the model parameters in the previous run and the parameters are adjusted until convergence is reached (Figure 5), and teaches that the clustering procedure is repeated until all models over all folds convergence within a predefined similarity metric (page 3088, column 2, Section Drug Solubility Estimation Using Multiple Models, paragraph 2).
Claims 3-5 and 13-15 are rejected under 35 U.S.C. 103 as being unpatentable over Ford et al. in view of Chinta et al., as applied to claims 1, 2, 6-12, and 16-20 above, and further in view of Feinberg et al. “PotentialNet for Molecular Property Prediction”, ACS Central Science, published 2018).
Regarding claims 3 and 13, Ford et al. teaches the method of claim 2 and the device of claim 12. Ford et al. also teaches a molecular adjacency matrix and teaches using the molecular adjacency matrix and protein data to perform feature extraction (column 7, line 42).
Neither Ford et al. nor Chinta et al. teach the claim element of determining an adjacency matrix of the sample protein molecule based on the sample protein molecule in the training sample, the adjacency matrix of the sample protein molecule characterizing sample amino acid molecules contained in the sample protein molecule and the molecular structure distance between every two sample amino acid molecules.
However, Feinberg et al. teaches machine learning for molecular property prediction and teaches the key parameters for the multiparameter optimization of drug discovery ranging from solubility to protein-ligand binding to in vivo toxicity. Feinberg et al. teaches structure-based scoring models (page 1522, column 1, Section II.B.) using adjacency matrices of the atoms of a protein (page 1522, column 2; page 1523, column 1).
Regarding claims 4 and 14, Ford et al. teaches the method of claim 1 and the device of claim 11 in view of Chinta et al.
Neither Ford et al. nor Chinta et al. teach the claim elements of predicting a binding distance.
However, Feinberg et al. teaches generating a distance matrix and teaches generating a spatial graph convolution, based on notions of adjacency predicated on Euclidean distance (page 1522, column 1, Section II.B.1 – column 2). Furthermore, Feinberg et al. teaches the spatial gated graph neural network in stage 1 performing graph convolutions over only binds which derives node atom node feature maps, and in stage 2 performing both bond-based and spatial distance-based propagation of information. Then in the final stage, a graph gather operation is conducted over the ligand atoms, whose features are derived from bonded ligand information and spatial proximity to protein atoms (page 1522, Figure 2 description) and teaches using the model to perform solubility predictions (page 1526, column 2, Section IV.C.3).
Regarding claims 5 and 15, Ford et al. teaches the method of claim 4 and the device of claim 14 in view of Chinta et al. and further in view of Feinberg et al.
Neither Ford et al. nor Chinta et al. teach the claim elements of determining embedding feature information and the eutectic feature information based on the plurality of binding distances.
However, Feinberg et al. teaches determining eutectic solubility information based on bonds and binned distances (page 1526, column 1, Section IV.C, paragraph 3), and teaches embedding protein and molecule information in a spatial graph convolution, a graph convolution based on notions of adjacency predicated on Euclidean distance (page 1522, column 2) and teaches that in a distance matrix, pairwise distances below a threshold value can be used (page 1522, column 1, Section II.B.1).
Therefore, it would have been prima facie obvious to one of ordinary skill in the art to incorporate the adjacency matrix and distance-based predictions of Feinberg et al. to the method of Ford et al. in view of Chinta et al. because Ford et al. is directed to predictions of enzyme and substrate combinations and interactions (column 2, line 1), and teaches that predicting new substrates for a given enzyme has potential applications in developing drug discovery panels (column 2, line 5). Feinberg et al. teaches that the arc of drug discovery entails a multiparameter optimization problem spanning vast length scales with key parameters ranging from solubility to protein-ligand binding, to in vivo toxicity. Feinberg et al. teaches that the PotentialNet family of graph convolutions is specifically designed for and achieves state-of-the-art performance for protein-ligand binding affinity (Abstract). Thus, one of ordinary skill in the art would have a reasonable expectation of success of predicting enzyme and substrate interactions by including parameters of Feinberg et al. by combining the prior art references and would be motivated to do so in order to achieve state-of-the-art performance for predicting enzyme and substrate interactions.
Conclusion
No claims are allowed.
Any inquiry concerning this communication or earlier communications from the examiner should be directed to Emilie A Smith whose telephone number is (571)272-7543. The examiner can normally be reached 9am - 5pm.
Examiner interviews are available via telephone, in-person, and video conferencing using a USPTO supplied web-based collaboration tool. To schedule an interview, applicant is encouraged to use the USPTO Automated Interview Request (AIR) at http://www.uspto.gov/interviewpractice.
If attempts to reach the examiner by telephone are unsuccessful, the examiner’s supervisor, 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.
/E.A.S./Examiner, Art Unit 1686
/OLIVIA M. WISE/Supervisory Patent Examiner, Art Unit 1685
Prosecution Insights