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 1-20 are currently pending and under examination herein.
Claims 1-20 are rejected.
Priority
The instant applications claims priority as a CON of PCT/JP2021/021850 filed 09 June 2021 and foreign priority to JP2020-100403 filed 09 June 2020. Receipt is acknowledged of certified copies of papers required by 37 CFR 1.55. In this action, claims 1-20 are examined as though they had an effective filing date of 09 June 2020. 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(s) (IDS) submitted on 07 December 2022, 06 January 2023, 15 August 2024, 15 June 2026 are in compliance with the provisions of 37 CFR 1.97. Accordingly, the information disclosure statements are being considered by the examiner.
Drawings
The drawings filed 07 December 2022 are accepted.
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.
Claims 6-17 and 19-20 are rejected under 35 U.S.C. 112(b) as being indefinite for failing to particularly point out and distinctly claim the subject matter which the inventor or a joint inventor regards as the invention.
Claims 6-9 and 12-15 recite “the score”. These claims depend on claim 1 which recites generating a plurality of score two or more times. It is unclear if these limitations are referring to a specific instance of the score out the multiple plurality. The metes and bounds of the claims are unclear rendering the claims indefinite. Claims 10 and 11 depend on Claim 7, and thus contain the above issues due to said dependence.
Claims 6, 12, 15-17 recite “the structural formula”. These claims depend on claim 1 which recites generating a plurality of structural formulas two or more times. It is unclear if these limitations are referring to a specific instance of the structural formula out the multiple plurality. The metes and bounds of the claims are unclear rendering the claims indefinite. Claims 7-11 depend on Claim 6, and thus contain the above issues due to said dependence.
Claims 19 and 20 recite “making the one or more processors” perform a function. It is unclear what making processers perform a function indicates. It is unclear what entity or external agent is acting on the processors to get them to perform the recited functions. The metes and bound the limitations are therefore unclear rendering the claim indefinite.
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 invention is directed to an abstract idea and a natural law without significantly more. In accordance with MPEP § 2106, claims found to recite statutory subject matter (Step 1: YES) are then analyzed to determine if the claims recite any concepts that equate to an abstract idea or natural law (Step 2A, Prong 1). Claims 1-18 and 20 are directed to systems and Claims 19 is directed to a method. In the instant application, the claims recite the following limitations that equate to an abstract idea:
Claim 1 recites the limitation - acquire a plurality of latent variables; generate a plurality of structural formulas by inputting the plurality of latent variables, respectively, into a model; calculate a plurality of scores by evaluating the plurality of structural formulas, respectively; the one or more processors execute processing of the acquisition of the plurality of latent variables, the generation of the plurality of structural formulas, and the calculation of the plurality of scores, at least two times or more; and the one or more processors acquire, based on the acquired plurality of latent variables and the calculated plurality of scores, the plurality of latent variables in any of the execution of at least second time or thereafter. Based on the broadest reasonable interpretation, acquiring variables, generating formulas, and calculating scores could include equations and could practically be done by the human mind. This draws the limitation to a mathematical concept and a mental process, which classifies the limitation as an abstract idea. MPEP 2106.04(a)(2) states claims can recite a mental process even if they are claimed as being performed on a computer. Additionally, MPEP 2106.04(a)(2) states that a mathematical concept need not be expressed in mathematical symbols, because "[w]ords used in a claim operating on data to solve a problem can serve the same purpose as a formula.
Claim 2 recites the limitation - the one or more processors acquire the plurality of latent variables based on a metaheuristic algorithm. Based on the broadest reasonable interpretation, acquiring variables based on an algorithm could include equations and could practically be done by the human mind. This draws the limitation to a mathematical concept and a mental process, which classifies the limitation as an abstract idea.
Claim 3 recites the limitation - the metaheuristic algorithm is at least one of a particle swarm optimization, an artificial bee colony method, a simulated annealing method, a Metropolis method, a Monte Carlo method, a hill climbing method, an ant colony optimization method, a harmony search, a cuckoo search, a spiral optimization method, a firefly algorithm, a genetic algorithm, an immune algorithm, a covariance matrix adaptation evolution strategy, an evolution strategy, or an amoeba method/Nelder-Mead method. This limitation specifies the possible algorithms used in the judicial exception of claim 2. The refined acquiring indicated by this limitation still represents a judicial expectation.
Claim 5 recites the limitation - the one or more processors acquire the plurality of latent variables based on a metaheuristic algorithm. Based on the broadest reasonable interpretation, acquiring latent variables could include equations and could practically be done by the human mind. This draws the limitation to a mathematical concept and a mental process, which classifies the limitation as an abstract idea.
Claim 6 recites the limitation - the one or more processors calculate the score based on a three-dimensional structure of a compound expressed by the structural formula. Based on the broadest reasonable interpretation, calculating the score could include equations and could practically be done by the human mind. This draws the limitation to a mathematical concept and a mental process, which classifies the limitation as an abstract idea.
Claim 8 recites the limitation - the one or more processors calculate the score based on a potential. Based on the broadest reasonable interpretation, calculating the score could include equations and could practically be done by the human mind. This draws the limitation to a mathematical concept and a mental process, which classifies the limitation as an abstract idea.
Claim 9 recites the limitation - the one or more processors calculate the score based on at least any of a docking position, a docking direction, or internal coordinates. Based on the broadest reasonable interpretation, calculating the score could include equations and could practically be done by the human mind. This draws the limitation to a mathematical concept and a mental process, which classifies the limitation as an abstract idea.
Claim 12 recites the limitation - the one or more processors calculate the score based on at least one of a three-dimensional structure, an avoidance structure, or an easiness of bonding with respect to predetermined protein, of a compound expressed by the structural formula. Based on the broadest reasonable interpretation, calculating the score could include equations and could practically be done by the human mind. This draws the limitation to a mathematical concept and a mental process, which classifies the limitation as an abstract idea.
Claim 13 recites the limitation - the score is determined based on a plurality of properties. Based on the broadest reasonable interpretation, calculating the score could include equations and could practically be done by the human mind. This draws the limitation to a mathematical concept and a mental process, which classifies the limitation as an abstract idea.
Claim 14 recites the limitation - the one or more processors calculate a plurality of kinds of the score. Based on the broadest reasonable interpretation, calculating the score could include equations and could practically be done by the human mind. This draws the limitation to a mathematical concept and a mental process, which classifies the limitation as an abstract idea.
Claim 15 recites the limitation - the score is an evaluation value based on a property of a compound expressed by the structural formula. This limitation specifies the calculated score in the judicial exception of claim 1. The refined calculation indicated by this limitation still represents a judicial expectation.
Claim 16 recites the limitation - the structural formula is information indicating at least either a molecular structure or a crystal structure. This limitation specifies the generated formula in the judicial exception of claim 1. The refined formula generated indicated by this limitation still represents a judicial expectation.
Claim 17 recites the limitation - the structural formula is information expressed by a graph. This limitation specifies the generated formula in the judicial exception of claim 1. The refined formula generated indicated by this limitation still represents a judicial expectation.
Claim 18 recites the limitation - the acquisition of the plurality of latent variables in the execution of the first time is for acquiring initial values of the plurality of latent variables. This limitation specifies the variables acquired in the judicial exception of claim 1. The refined determination indicated by this limitation still represents a judicial expectation.
Claim 19 recites the limitation - acquire a plurality of latent variables; generate a plurality of structural formulas by inputting the plurality of latent variables, respectively, into a model; and calculate a plurality of scores by evaluating the plurality of structural formulas, respectively, wherein: processing of the acquisition of the plurality of latent variables, the generation of the plurality of structural formulas, and the calculation of the plurality of scores, at least two times or more; and acquire, based on the acquired plurality of latent variables and the calculated plurality of scores, the plurality of latent variables in any of the execution of at least second time or thereafter. Based on the broadest reasonable interpretation, acquiring variables, generating formulas, and calculating scores could include equations and could practically be done by the human mind. This draws the limitation to a mathematical concept and a mental process, which classifies the limitation as an abstract idea.
Claim 20 recites the limitation - acquire a plurality of latent variables; generate a plurality of structural formulas by inputting the plurality of latent variables, respectively, into a model; and calculate a plurality of scores by evaluating the plurality of structural formulas, respectively, wherein: processing of the acquisition of the plurality of latent variables, the generation of the plurality of structural formulas, and the calculation of the plurality of scores, at least two times or more; and acquire, based on the acquired plurality of latent variables and the calculated plurality of scores, the plurality of latent variables in any of the execution of at least second time or thereafter.
These limitations recite concepts of determining, generating, and calculating information that are so generically recited that they can be practically performed in the human mind as claimed, which falls under the “Mental processes” and “Mathematical concepts” grouping of abstract ideas. Therefore, these limitations fall under the “Mental process” and “Mathematical concepts” groupings of abstract ideas. As such, claims 1-20 recite an abstract idea (Step 2A, Prong 1: YES).
Claims found to recite a judicial exception under Step 2A, Prong 1 are then further analyzed to determine if the claims as a whole integrate the recited judicial exception into a practical application or not (Step 2A, Prong 2). These judicial exceptions are not integrated into a practical application because the claims do not recite an additional element that reflects an improvement to technology (MPEP § 2106.04(d)(1)). Rather, the claims provide insignificant extra-solution activity (MPEP § 2106.05(g)) and provide mere instructions to apply a judicial exception (MPEP § 2106.05(f)). Specifically, the claims recite the following additional elements:
Claim 1 recites one or more memories; and one or more processors
Claim 4 recites the one or more processors execute at least any processing of the acquisition of the plurality of latent variables, the generation of the plurality of structural formulas, or the calculation of the plurality of scores, in a parallel manner.
Claim 7 recites the one or more processors calculate the score by performing a simulation of a docking of the compound.
Claim 10 recites the one or more processors execute the simulation of the docking by using a first method which executes a global search, and a second method which executes a search in a more local manner than the first method.
Claim 11 recites the one or more processors execute at least the search using the first method in a parallel manner.
Claim 19 recites processors
Claim 20 recites a non-transitory computer readable medium storing a program; and processors.
There are no limitations that indicate that the determining, generating, and calculating information and applying algorithms require anything other than generic computing systems. As such, these limitations equate to mere instructions to implement the abstract idea on a generic computer that the courts have stated does not render an abstract idea eligible in Alice Corp., 573 U.S. at 223, 110 USPQ2d at 1983. See also 573 U.S. at 224, 110 USPQ2d at 1984. There is no indication that these steps are affected by the judicial exception in any way and thus do not integrate the recited judicial exception into a practical application. As such, claims 1-20 are directed to an abstract idea (Step 2A, Prong 2: NO).
Claims found to be directed to a judicial exception are then further evaluated to determine if the claims recite an inventive concept that provides significantly more than the judicial exception itself (Step 2B). The claims do not include additional elements that are sufficient to amount to significantly more than the judicial exception because the claims recite conventional additional elements that equate to mere instructions to apply the recited exception in a generic way or in a generic computing environment. The claims also recite conventional additional elements that represent insignificant extra-solution activities.
As discussed above, there are no additional limitations to indicate that the claimed determining, generating, and calculating information and applying algorithms require anything other than generic computer components in order to carry out the recited abstract idea in the claims. 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. Alice Corp., 573 U.S. at 223, 110 USPQ2d at 1983. See also 573 U.S. at 224, 110 USPQ2d at 1984. MPEP 2106.05(f) discloses that mere instructions to apply the judicial exception cannot provide an inventive concept to the claims. As specified in MPEP 2106.05(g), extra-solution activities can be understood as incidental to the primary process or product that are merely a nominal or tangential addition to the claim. Insignificant extra-solution activities include mere data gathering, selecting a particular data source or type of data to be manipulated, and displaying information. Additionally, Korb et al. (2014, Expert Opinion on Drug Discovery, Vol. 9: 1121-1131) teach the use of computers and parallel processing techniques in molecular docking simulation and scoring is well understood, routine, and conventional (Page 1122, Column 2, Paragraph 2: The term grid computing refers to the paradigm of distributing a software task across a network of computers or nodes. Virtual screening methods, such as docking, where each node on the grid processes a small number of compounds, so as to enable the screening of large datasets across the grid, are ideally suited to grid computing; Page 1126, Column 1, Paragraph 2: In the area of molecular docking, most GPU-based approaches concentrate on accelerating the conformational sampling step and the evaluation of the scoring function).
The additional elements do not comprise an inventive concept when considered individually or as an ordered combination that transforms the claimed judicial exception into a patent-eligible application of the judicial exception. Therefore, the claims do not amount to significantly more than the judicial exception itself (Step 2B: No). As such, Claims 1-20 are not patent eligible.
Claim Rejections - 35 USC § 102
The following is a quotation of the appropriate paragraphs of 35 U.S.C. 102 that form the basis for the rejections under this section made in this Office action:
A person shall be entitled to a patent unless –
(a)(1) the claimed invention was patented, described in a printed publication, or in public use, on sale, or otherwise available to the public before the effective filing date of the claimed invention.
Claims 1-3, 6-10, 12-16, and 18-20 are rejected under 35 U.S.C. 102(a)(1) as being anticipated by Sattarov et al. (2019, Journal of Chemical Information and Modeling, Vol. 59: 1182–1196). Italicized text from reference art.
The applicable claims include:
Claim 1. An inferring device comprising: one or more memories; and one or more processors configured to: (i) acquire a plurality of latent variables; (ii) generate a plurality of structural formulas by inputting the plurality of latent variables, respectively, into a model; and (iii) calculate a plurality of scores by evaluating the plurality of structural formulas, respectively, wherein: (iv) the one or more processors execute processing of the acquisition of the plurality of latent variables, the generation of the plurality of structural formulas, and the calculation of the plurality of scores, at least two times or more; and (v) the one or more processors acquire, based on the acquired plurality of latent variables and the calculated plurality of scores, the plurality of latent variables in any of the execution of at least second time or thereafter.
Claim 2. The inferring device according to claim 1, wherein the one or more processors acquire the plurality of latent variables based on a metaheuristic algorithm.
Claim 3. The inferring device according to claim 2, wherein the metaheuristic algorithm is at least one of a particle swarm optimization, an artificial bee colony method, a simulated annealing method, a Metropolis method, a Monte Carlo method, a hill climbing method, an ant colony optimization method, a harmony search, a cuckoo search, a spiral optimization method, a firefly algorithm, a genetic algorithm, an immune algorithm, a covariance matrix adaptation evolution strategy, an evolution strategy, or an amoeba method/Nelder-Mead method.
Claim 6. The inferring device according to claim 1, wherein the one or more processors calculate the score based on a three-dimensional structure of a compound expressed by the structural formula.
Claim 7. The inferring device according to claim 6, wherein the one or more processors calculate the score by performing a simulation of a docking of the compound.
Claim 8. The inferring device according to claim 7, wherein the one or more processors calculate the score based on a potential.
Claim 9. The inferring device according to claim 7, wherein the one or more processors calculate the score based on at least any of a docking position, a docking direction, or internal coordinates.
Claim 10. The inferring device according to claim 7, wherein the one or more processors execute the simulation of the docking by using a first method which executes a global search, and a second method which executes a search in a more local manner than the first method.
Claim 12. The inferring device according to claim 1, wherein the one or more processors calculate the score based on at least one of a three-dimensional structure, an avoidance structure, or an easiness of bonding with respect to predetermined protein, of a compound expressed by the structural formula.
Claim 13. The inferring device according to claim 1, wherein the score is determined based on a plurality of properties.
Claim 14. The inferring device according to claim 1, wherein the one or more processors calculate a plurality of kinds of the score.
Claim 15. The inferring device according to claim 1, wherein the score is an evaluation value based on a property of a compound expressed by the structural formula.
Claim 16. The inferring device according to claim 1, wherein the structural formula is information indicating at least either a molecular structure or a crystal structure.
Claim 18. The inferring device according to claim 1, wherein the acquisition of the plurality of latent variables in the execution of the first time is for acquiring initial values of the plurality of latent variables.
Claim 19. An inferring method comprising: (i) making one or more processors acquire a plurality of latent variables; (ii) making the one or more processors generate a plurality of structural formulas by inputting the plurality of latent variables, respectively, into a model; and (iii) making the one or more processors calculate a plurality of scores by evaluating the plurality of structural formulas, respectively, wherein:(iv) the one or more processors execute processing of the acquisition of the plurality of latent variables, the generation of the plurality of structural formulas, and the calculation of the plurality of scores, at least two times or more; and (v) the one or more processors acquire, based on the acquired plurality of latent variables and the calculated plurality of scores, the plurality of latent variables in any of the execution of at least second time or thereafter.
Claim 20. A non-transitory computer readable medium storing a program, the program configured to: (i) making one or more processors acquire a plurality of latent variables; (ii) making the one or more processors generate a plurality of structural formulas by inputting the plurality of latent variables, respectively, into a model; and (iii) making the one or more processors calculate a plurality of scores by evaluating the plurality of structural formulas, respectively, wherein: (iv) the one or more processors are made to execute processing of the acquisition of the plurality of latent variables, the generation of the plurality of structural formulas, and the calculation of the plurality of scores, at least two times or more; and (v) the one or more processors are made to acquire, based on the acquired plurality of latent variables and the calculated plurality of scores, the plurality of latent variables in any of the execution of at least second time or thereafter.
Regarding Claims 1, 19 and 20, Sattarov et al. teach (Claim 1.i) acquire a plurality of latent variables (Page 1184, Column 2, Paragraph 3: A part of the autoencoder called encoder computes the function for each sample s, where d is the code, called also latent vector; Page 1185, Column 1, Paragraph 2: autoencoders work with data represented as vectors in a multidimensional space; Page 1186, Column 2, Paragraphs 2: The autoencoder was trained in the batch mode, where batches of “one-hot”-encoded sequences). The latent vector contains a plurality of dimensions of data which is interpreted to be equivalent to a plurality of latent variables. The autoencoder was also run in batches of inputs which is interpreted as generating a plurality of latent variables. Sattarov et al. also teach (Claim 1.ii) generate structural formulas by inputting the latent variables into a model (Page 1188, Column 1, Paragraph 2: Given a latent vector, it can be interpreted by the trained decoder to produce a SMILES string for a new molecule. Since the GTM activity landscapes allows us to locate zones enriched with molecules with properties of interest, we can generate new latent vectors for them and, hence, the SMILES strings of novel molecules with desired properties; Page 1187, Figure 5: Generation of the focused library of novel active structures). A SMILES string is interpreted as a representation of a chemical structure (also see Page 1185, Figure 2). Sattarov et al. also teach (Claim 1.iii) calculate scores by evaluating the structural formulas (Page 1188, Column 1, paragraph 5: For each valid chemical structure, the Synthetic Accessibility score (SA) was estimated; Page 1188, Column 1, paragraph 6: By trial and error, the fitness score Fit was empirically defined). Besides the two scores indicated from the art above, multiple other assessments of the generated structures are calculated as values, which are interpreted as scores. Sattarov et al. also teach (Claim 1.iv) the acquisition of latent variables, the generation of structural formulas, and the calculation of scores occur at least two times or more (Page 1189, Column 2, Paragraph 2: For the autoencoder training, an ensemble of “one-hot”-encoded canonical SMILES strings was divided into the training set with 1211352 molecules and the validation set). The steps of the method occurred at least two times because there was at least a training and validation run of the methods. Sattarov et al. also teach (Claim 1.v) acquire, based on the acquired latent variables and the calculated scores, the latent variables in the execution of the second or further time. The training of the autoencoder (see claim 1.iv above) will impact how the trained encoder performs (i.e. the initial parameters of the model are adjusted) when processing the next set/conducting the next run. This is interpreted as equivalent to the acquired variables and calculating scores of the first run impact the acquiring the variable of the next run (i.e. the encoder is trained during the training run). Additionally, Sattarov et al. teach their methods were performed by a computer which contains memory and at least one processor (Page 1184, Column 2, Paragraph 1: We used RDKit60 canonical SMILES in this study; Page 1190, Column 1, Paragraph 2: <10 h on Nvidia GTX1080). The computer is considered synonymous with the inferring device and is shown performing processor based methods. Computers also inherently contain non-transitory computer readable medium. Claim 19 recites the limitations of claim 1 directed to a method and claim 20 recites the limitations of claim 1 directed to a non-transitory computer readable medium.
Regarding Claim 2, Sattarov et al. teach acquire the latent variables based on a metaheuristic algorithm (Page 1188, Column 1, Paragraph 6: The GA-sampling explores the latent space by means of the genetic algorithm with the latent-space vectors coded as chromosomes). Exploring the latent space is involved in selection (i.e. acquiring) of the latent vectors.
Regarding Claim 3, Sattarov et al. teach the metaheuristic algorithm is a genetic algorithm (Page 1188, Column 1, Paragraph 6: The GA-sampling explores the latent space by means of the genetic algorithm with the latent-space vectors coded as chromosomes).
Regarding Claim 6, Sattarov et al. teach calculate the score based on a three-dimensional structure of a compound expressed by the structural formula (Page 1189, Column 2, Paragraph 1: the docking score for the current ligand can be directly estimated). The chemical structural formula are converted into 3d structure for docking to generate the docking score (Page 1188, Column 2, Paragraph 7: ligands selected for docking were subjected to an automated conversion to protonated initial 3D structures).
Regarding Claim 7, Sattarov et al. teach calculate the score by performing a simulation of a docking of the compound (Page 1189, Column 2, Paragraph 1: the docking score for the current ligand can be directly estimated).
Regarding Claim 8, Sattarov et al. teach calculate the score based on a potential. The docking simulations that the score is based on (see regarding claim 7 above) are in silico so they represent only the potential for the compound to dock in vitro/vivo (Page 1192, Column 2, Paragraph 2: Pending real synthesis and experimental validation, we have applied alternative chemoinformatics approaches that do not rely on machine-learned models (pharmacophore screening and docking) to validate that the generated compounds could be binders to the adenosine A2A receptor). Therefore, generating the score based on a docking simulation is interpreted as calculating the score based on a potential (i.e. based on a possibility). Additionally, the scores discussed above (Regarding claim 1.iii) are based on in silico modeling and therefore represent scores based on a potential. No limiting definition of potential was found in the specification.
Regarding Claim 9, Sattarov et al. teach the one or more processors calculate the score based on a docking position, a docking direction, or internal coordinates (Page 1188, Column 2, Paragraph 7: Next, antechamber and other utilities were used to assign GAFF88 ligand types and to automatically set associated FF parameters to the internal coordinates found in the ligands; Page 1192, Figure 11: Binding site of the A2a receptor cocrystallized with ligand (3EML PDB structure) overlapped with the developed pharmacophore filter). Docking position is interpreted as synonymous with binding site.
Regarding Claim 10, Sattarov et al. teach the simulation of the docking uses a first method which executes a global search, and a second method which executes a search in a more local manner than the first method (Page 1188, Column 2, Paragraph 5: Two different docking approaches were applied: the in-house program S4MPLE, a general tool which handles docking simulations as a special case of the wider range of conformational sampling problems it may tackle, versus the state-of-art FlexX docking software licensed by BioSolveIT; Page 1189, Column 1, Paragraph 2: by default, all degrees of freedom are considered in S4MPLE. For FlexX, the active site was automatically defined by the subset of protein residues interacting with the adenosine ligand present in 2YDO according to the procedure implemented in LeadIt). The art indicated above utilizes two docking methods, S4MPLE and FlexX. S4MPLE was interpreted to be more global than FlexX because as indicated by the text above, it is a more general tool, considers all degrees of freedom, and does not have the specificity of FlexX which automatically defines a subset of residues to consider. No limiting definition was found in the specification for global search or search in a more local manner.
Regarding Claim 12, Sattarov et al. teach calculate the score based on a three-dimensional structure, an avoidance structure, or an easiness of bonding with respect to predetermined protein, of a compound expressed by the structural formula (Page 1189, Column 2, Paragraph 1: the docking score for the current ligand can be directly estimated). The chemical structural formula are converted into 3d structure for docking to generate the docking score (Page 1188, Column 2, Paragraph 7: ligands selected for docking were subjected to an automated conversion to protonated initial 3D structures). This indicates the score is based on docking which is based on at least one 3d structure. Additionally, (Page 1193, Column 1, paragraph 1: We have also found that S4MPLE reproduces correct binding modes of ligands; Page 1193, Figure 13: Histogram of docking scores of the generated compounds and compounds with experimentally measured activity for the binding pocket of the adenosine receptor A2a) indicates the scores are also calculated based on ligand binding, which is interpreted as equivalent to an easiness of bonding with respect to predetermined protein. This is related to the docking score that are generated (see Figure 13)
Regarding Claim 13, Sattarov et al. teach the score is determined based on a plurality of properties (Page 1187, Column 1, Paragraph 2: Besides visualization, such landscapes can be used for predicting the activity values or class labels for new molecules by projecting their descriptors onto the GTM and taking the local average property as predicted value; Page 1188, Column 1, Paragraph 2: Since the GTM activity landscapes allows us to locate zones enriched with molecules with properties of interest, we can generate new latent vectors for them and, hence, the SMILES strings of novel molecules with desired properties. Several properties of the libraries, such as synthetic accessibility and internal diversity of compounds, were assessed and compared with corresponding properties of the ChEMBL23 database). In addition to the properties indicated above that are related to the scores generated for the compounds, the properties related docking as indicated by claim 9 are also related to the determination of the docking score. No explicit definition of property was found in the specification.
Regarding Claim 14, Sattarov et al. teach calculate a plurality of kinds of the score (Page 1188, Column 1, paragraph 5: For each valid chemical structure, the Synthetic Accessibility score (SA) was estimated; Page 1188, Column 1, paragraph 6: By trial and error, the fitness score Fit was empirically defined). Besides the two scores indicated from the art above, multiple other assessments of the generated structures are calculated as values, which are interpreted as scores. There is also the docking score that is calculated (Page 1189, Column 2, Paragraph 1: the docking score for the current ligand can be directly estimated).
Regarding Claim 15, Sattarov et al. teach the score is an evaluation value based on a property of a compound expressed by the structural formula (Page 1188, Column 1, paragraph 3: Several properties of the libraries, such as synthetic accessibility and internal diversity of compounds, were assessed and compared with corresponding properties of the ChEMBL23 database; Page 1188, Column 1, paragraph 5: For each valid chemical structure, the Synthetic Accessibility score (SA) was estimated; Page 1188, Column 1, paragraph 6: By trial and error, the fitness score Fit was empirically defined). Besides the two scores indicated from the art above, multiple other assessments of the generated structures are calculated as values, which are interpreted as scores. Additionally, the properties of a compound are inherent to the compound (i.e. a score based on a compound is inherently based on the properties of the compound). The compounds are expressed as chemical formula (SMILES, see regarding claim 1).
Regarding Claim 16, Sattarov et al. teach the structural formula is information indicating at least either a molecular structure or a crystal structure (Page 1184, Column 1 Paragraph 5: To be processed by seq2seq models, molecules must be represented as sequences of characters−such as the Simplified Molecular-Input Line-Entry System (SMILES) strings). The SMILES used represent a structural formula and a molecular structure. Additionally, the methods also utilize formulas related to crystal structures (Page 1189, Column 1, Paragraph 2: Therefore, a selection of relevant residues that have at least one atom at less than 10 Å from any of the cocrystallized 2YOD ligand was used as site model for S4MPLE calculations).
Regarding Claim 18, Sattarov et al. teach the acquisition of latent variables in the execution of the first time is for acquiring initial values of the plurality of latent variables (Page 1186, Column 1, Paragraph 1: the vector is used as an initial state for the decoder, which carries out reconstruction of the sequence). The first run of the model, prior to when it is considered a trained model, which includes acquiring latent variables, is interpreted as the initial set of latent variables. The initial set of latent variables could also be considered the latent variables generated on the first run prior to any selections that are made. No limiting definition for initial values was found in the specification.
Claim Rejections - 35 USC § 103
The following is a quotation of 35 U.S.C. 103 which forms the basis for all obviousness rejections set forth in this Office action:
A patent for a claimed invention may not be obtained, notwithstanding that the claimed invention is not identically disclosed as set forth in section 102, if the differences between the claimed invention and the prior art are such that the claimed invention as a whole would have been obvious before the effective filing date of the claimed invention to a person having ordinary skill in the art to which the claimed invention pertains. Patentability shall not be negated by the manner in which the invention was made.
The factual inquiries 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 1-16 and 18-20 are rejected under 35 U.S.C. 103 as being unpatentable over Sattarov et al. (2019, Journal of Chemical Information and Modeling, Vol. 59: 1182–1196), as applied in the 102 rejection above, in view of Zhang et al. (2014, Journal of Chemical Information and Modeling, Vol. 54: 324−337). Italicized text from reference art.
The applicable claims include:
Claim 1. An inferring device comprising: one or more memories; and one or more processors configured to: (i) acquire a plurality of latent variables; (ii) generate a plurality of structural formulas by inputting the plurality of latent variables, respectively, into a model; and (iii) calculate a plurality of scores by evaluating the plurality of structural formulas, respectively, wherein: (iv) the one or more processors execute processing of the acquisition of the plurality of latent variables, the generation of the plurality of structural formulas, and the calculation of the plurality of scores, at least two times or more; and (v) the one or more processors acquire, based on the acquired plurality of latent variables and the calculated plurality of scores, the plurality of latent variables in any of the execution of at least second time or thereafter.
Claim 2. The inferring device according to claim 1, wherein the one or more processors acquire the plurality of latent variables based on a metaheuristic algorithm.
Claim 3. The inferring device according to claim 2, wherein the metaheuristic algorithm is at least one of a particle swarm optimization, an artificial bee colony method, a simulated annealing method, a Metropolis method, a Monte Carlo method, a hill climbing method, an ant colony optimization method, a harmony search, a cuckoo search, a spiral optimization method, a firefly algorithm, a genetic algorithm, an immune algorithm, a covariance matrix adaptation evolution strategy, an evolution strategy, or an amoeba method/Nelder-Mead method.
Claim 4. The inferring device according to claim 1, wherein the one or more processors execute at least any processing of the acquisition of the plurality of latent variables, the generation of the plurality of structural formulas, or the calculation of the plurality of scores, in a parallel manner.
Claim 5. The inferring device according to claim 4, wherein the one or more processors acquire the plurality of latent variables based on a metaheuristic algorithm.
Claim 6. The inferring device according to claim 1, wherein the one or more processors calculate the score based on a three-dimensional structure of a compound expressed by the structural formula.
Claim 7. The inferring device according to claim 6, wherein the one or more processors calculate the score by performing a simulation of a docking of the compound.
Claim 8. The inferring device according to claim 7, wherein the one or more processors calculate the score based on a potential.
Claim 9. The inferring device according to claim 7, wherein the one or more processors calculate the score based on at least any of a docking position, a docking direction, or internal coordinates.
Claim 10. The inferring device according to claim 7, wherein the one or more processors execute the simulation of the docking by using a first method which executes a global search, and a second method which executes a search in a more local manner than the first method.
Claim 11. The inferring device according to claim 10, wherein the one or more processors execute at least the search using the first method in a parallel manner.
Claim 12. The inferring device according to claim 1, wherein the one or more processors calculate the score based on at least one of a three-dimensional structure, an avoidance structure, or an easiness of bonding with respect to predetermined protein, of a compound expressed by the structural formula.
Claim 13. The inferring device according to claim 1, wherein the score is determined based on a plurality of properties.
Claim 14. The inferring device according to claim 1, wherein the one or more processors calculate a plurality of kinds of the score.
Claim 15. The inferring device according to claim 1, wherein the score is an evaluation value based on a property of a compound expressed by the structural formula.
Claim 16. The inferring device according to claim 1, wherein the structural formula is information indicating at least either a molecular structure or a crystal structure.
Claim 18. The inferring device according to claim 1, wherein the acquisition of the plurality of latent variables in the execution of the first time is for acquiring initial values of the plurality of latent variables.
Claim 19. An inferring method comprising: (i) making one or more processors acquire a plurality of latent variables; (ii) making the one or more processors generate a plurality of structural formulas by inputting the plurality of latent variables, respectively, into a model; and (iii) making the one or more processors calculate a plurality of scores by evaluating the plurality of structural formulas, respectively, wherein:(iv) the one or more processors execute processing of the acquisition of the plurality of latent variables, the generation of the plurality of structural formulas, and the calculation of the plurality of scores, at least two times or more; and (v) the one or more processors acquire, based on the acquired plurality of latent variables and the calculated plurality of scores, the plurality of latent variables in any of the execution of at least second time or thereafter.
20. A non-transitory computer readable medium storing a program, the program configured to: (i) making one or more processors acquire a plurality of latent variables; (ii) making the one or more processors generate a plurality of structural formulas by inputting the plurality of latent variables, respectively, into a model; and (iii) making the one or more processors calculate a plurality of scores by evaluating the plurality of structural formulas, respectively, wherein: (iv) the one or more processors are made to execute processing of the acquisition of the plurality of latent variables, the generation of the plurality of structural formulas, and the calculation of the plurality of scores, at least two times or more; and (v) the one or more processors are made to acquire, based on the acquired plurality of latent variables and the calculated plurality of scores, the plurality of latent variables in any of the execution of at least second time or thereafter.
Regarding Claims 1, 19 and 20, Sattarov et al. teach (Claim 1.i) acquire a plurality of latent variables (Page 1184, Column 2, Paragraph 3: A part of the autoencoder called encoder computes the function for each sample s, where d is the code, called also latent vector; Page 1185, Column 1, Paragraph 2: autoencoders work with data represented as vectors in a multidimensional space; Page 1186, Column 2, Paragraphs 2: The autoencoder was trained in the batch mode, where batches of “one-hot”-encoded sequences). The latent vector contains a plurality of dimensions of data which is interpreted to be equivalent to a plurality of latent variables. The autoencoder was also run in batches of inputs which is interpreted as generating a plurality of latent variables. Sattarov et al. also teach (Claim 1.ii) generate structural formulas by inputting the latent variables into a model (Page 1188, Column 1, Paragraph 2: Given a latent vector, it can be interpreted by the trained decoder to produce a SMILES string for a new molecule. Since the GTM activity landscapes allows us to locate zones enriched with molecules with properties of interest, we can generate new latent vectors for them and, hence, the SMILES strings of novel molecules with desired properties; Page 1187, Figure 5: Generation of the focused library of novel active structures). A SMILES string is interpreted as a representation of a chemical structure (also see Page 1185, Figure 2). Sattarov et al. also teach (Claim 1.iii) calculate scores by evaluating the structural formulas (Page 1188, Column 1, paragraph 5: For each valid chemical structure, the Synthetic Accessibility score (SA) was estimated; Page 1188, Column 1, paragraph 6: By trial and error, the fitness score Fit was empirically defined). Besides the two scores indicated from the art above, multiple other assessments of the generated structures are calculated as values, which are interpreted as scores. Sattarov et al. also teach (Claim 1.iv) the acquisition of latent variables, the generation of structural formulas, and the calculation of scores occur at least two times or more (Page 1189, Column 2, Paragraph 2: For the autoencoder training, an ensemble of “one-hot”-encoded canonical SMILES strings was divided into the training set with 1211352 molecules and the validation set). The steps of the method occurred at least two times because there was at least a training and validation run of the methods. Sattarov et al. also teach (Claim 1.v) acquire, based on the acquired latent variables and the calculated scores, the latent variables in the execution of the second or further time. The training of the autoencoder (see claim 1.iv above) will impact how the trained encoder performs (i.e. the initial parameters of the model are adjusted) when processing the next set/conducting the next run. This is interpreted as equivalent to the acquired variables and calculating scores of the first run impact the acquiring the variable of the next run (i.e. the encoder is trained during the training run). Additionally, Sattarov et al. teach their methods were performed by a computer which contains memory and at least one processor (Page 1184, Column 2, Paragraph 1: We used RDKit60 canonical SMILES in this study; Page 1190, Column 1, Paragraph 2: <10 h on Nvidia GTX1080). The computer is considered synonymous with the inferring device and is shown performing processor based methods. Computers also inherently contain non-transitory computer readable medium. Claim 19 recites the limitations of claim 1 directed to a method and claim 20 recites the limitations of claim 1 directed to a non-transitory computer readable medium.
Regarding Claim 2, Sattarov et al. teach acquire the latent variables based on a metaheuristic algorithm (Page 1188, Column 1, Paragraph 6: The GA-sampling explores the latent space by means of the genetic algorithm with the latent-space vectors coded as chromosomes). Exploring the latent space is involved in selection (i.e. acquiring) of the latent vectors.
Regarding Claim 3, Sattarov et al. teach the metaheuristic algorithm is a genetic algorithm (Page 1188, Column 1, Paragraph 6: The GA-sampling explores the latent space by means of the genetic algorithm with the latent-space vectors coded as chromosomes).
Regarding Claim 5, Sattarov et al. teach acquire the plurality of latent variables based on a metaheuristic algorithm (Page 1188, Column 1, Paragraph 6: The GA-sampling explores the latent space by means of the genetic algorithm with the latent-space vectors coded as chromosomes). Exploring the latent space in involved in selection (i.e. acquiring) of the latent vectors.
Regarding Claim 6, Sattarov et al. teach calculate the score based on a three-dimensional structure of a compound expressed by the structural formula (Page 1189, Column 2, Paragraph 1: the docking score for the current ligand can be directly estimated). The chemical structural formula are converted into 3d structure for docking to generate the docking score (Page 1188, Column 2, Paragraph 7: ligands selected for docking were subjected to an automated conversion to protonated initial 3D structures).
Regarding Claim 7, Sattarov et al. teach calculate the score by performing a simulation of a docking of the compound (Page 1189, Column 2, Paragraph 1: the docking score for the current ligand can be directly estimated).
Regarding Claim 8, Sattarov et al. teach calculate the score based on a potential. The docking simulations that the score is based on (see regarding claim 7 above) are in silico so they represent only the potential for the compound to dock in vitro/vivo (Page 1192, Column 2, Paragraph 2: Pending real synthesis and experimental validation, we have applied alternative chemoinformatics approaches that do not rely on machine-learned models (pharmacophore screening and docking) to validate that the generated compounds could be binders to the adenosine A2A receptor). Therefore, generating the score based on a docking simulation is interpreted as calculating the score based on a potential (i.e. based on a possibility). Additionally, the scores discussed above (Regarding claim 1.iii) are based on in silico modeling and therefore represent scores based on a potential. No limiting definition of potential was found in the specification.
Regarding Claim 9, Sattarov et al. teach the one or more processors calculate the score based on a docking position, a docking direction, or internal coordinates (Page 1188, Column 2, Paragraph 7: Next, antechamber and other utilities were used to assign GAFF88 ligand types and to automatically set associated FF parameters to the internal coordinates found in the ligands; Page 1192, Figure 11: Binding site of the A2a receptor cocrystallized with ligand (3EML PDB structure) overlapped with the developed pharmacophore filter). Docking position is interpreted as synonymous with binding site.
Regarding Claim 10, Sattarov et al. teach the simulation of the docking uses a first method which executes a global search, and a second method which executes a search in a more local manner than the first method (Page 1188, Column 2, Paragraph 5: Two different docking approaches were applied: the in-house program S4MPLE, a general tool which handles docking simulations as a special case of the wider range of conformational sampling problems it may tackle, versus the state-of-art FlexX docking software licensed by BioSolveIT; Page 1189, Column 1, Paragraph 2: by default, all degrees of freedom are considered in S4MPLE. For FlexX, the active site was automatically defined by the subset of protein residues interacting with the adenosine ligand present in 2YDO according to the procedure implemented in LeadIt). The art indicated above utilizes two docking methods, S4MPLE and FlexX. S4MPLE was interpreted to be more global than FlexX because as indicated by the text above, it is a more general tool, considers all degrees of freedom, and does not have the specificity of FlexX which automatically defines a subset of residues to consider. No limiting definition was found in the specification for global search or search in a more local manner.
Regarding Claim 12, Sattarov et al. teach calculate the score based on a three-dimensional structure, an avoidance structure, or an easiness of bonding with respect to predetermined protein, of a compound expressed by the structural formula (Page 1189, Column 2, Paragraph 1: the docking score for the current ligand can be directly estimated). The chemical structural formula are converted into 3d structure for docking to generate the docking score (Page 1188, Column 2, Paragraph 7: ligands selected for docking were subjected to an automated conversion to protonated initial 3D structures). This indicates the score is based on docking which is based on at least one 3d structure. Additionally, (Page 1193, Column 1, paragraph 1: We have also found that S4MPLE reproduces correct binding modes of ligands; Page 1193, Figure 13: Histogram of docking scores of the generated compounds and compounds with experimentally measured activity for the binding pocket of the adenosine receptor A2a) indicates the scores are also calculated based on ligand binding, which is interpreted as equivalent to an easiness of bonding with respect to predetermined protein. This is related to the docking score that are generated (see Figure 13)
Regarding Claim 13, Sattarov et al. teach the score is determined based on a plurality of properties (Page 1187, Column 1, Paragraph 2: Besides visualization, such landscapes can be used for predicting the activity values or class labels for new molecules by projecting their descriptors onto the GTM and taking the local average property as predicted value; Page 1188, Column 1, Paragraph 2: Since the GTM activity landscapes allows us to locate zones enriched with molecules with properties of interest, we can generate new latent vectors for them and, hence, the SMILES strings of novel molecules with desired properties. Several properties of the libraries, such as synthetic accessibility and internal diversity of compounds, were assessed and compared with corresponding properties of the ChEMBL23 database). In addition to the properties indicated above that are related to the scores generated for the compounds, the properties related docking as indicated by claim 9 are also related to the determination of the docking score. No explicit definition of property was found in the specification.
Regarding Claim 14, Sattarov et al. teach calculate a plurality of kinds of the score (Page 1188, Column 1, paragraph 5: For each valid chemical structure, the Synthetic Accessibility score (SA) was estimated; Page 1188, Column 1, paragraph 6: By trial and error, the fitness score Fit was empirically defined). Besides the two scores indicated from the art above, multiple other assessments of the generated structures are calculated as values, which are interpreted as scores. There is also the docking score that is calculated (Page 1189, Column 2, Paragraph 1: the docking score for the current ligand can be directly estimated).
Regarding Claim 15, Sattarov et al. teach the score is an evaluation value based on a property of a compound expressed by the structural formula (Page 1188, Column 1, paragraph 3: Several properties of the libraries, such as synthetic accessibility and internal diversity of compounds, were assessed and compared with corresponding properties of the ChEMBL23 database; Page 1188, Column 1, paragraph 5: For each valid chemical structure, the Synthetic Accessibility score (SA) was estimated; Page 1188, Column 1, paragraph 6: By trial and error, the fitness score Fit was empirically defined). Besides the two scores indicated from the art above, multiple other assessments of the generated structures are calculated as values, which are interpreted as scores. Additionally, the properties of a compound are inherent to the compound (i.e. a score based on a compound is inherently based on the properties of the compound). The compounds are expressed as chemical formula (SMILES, see regarding claim 1).
Regarding Claim 16, Sattarov et al. teach the structural formula is information indicating at least either a molecular structure or a crystal structure (Page 1184, Column 1 Paragraph 5: To be processed by seq2seq models, molecules must be represented as sequences of characters−such as the Simplified Molecular-Input Line-Entry System (SMILES) strings). The SMILES used represent a structural formula and a molecular structure. Additionally, the methods also utilize formulas related to crystal structures (Page 1189, Column 1, Paragraph 2: Therefore, a selection of relevant residues that have at least one atom at less than 10 Å from any of the cocrystallized 2YOD ligand was used as site model for S4MPLE calculations).
Regarding Claim 18, Sattarov et al. teach the acquisition of latent variables in the execution of the first time is for acquiring initial values of the plurality of latent variables (Page 1186, Column 1, Paragraph 1: the vector is used as an initial state for the decoder, which carries out reconstruction of the sequence). The first run of the model, prior to when it is considered a trained model, which includes acquiring latent variables, is interpreted as the initial set of latent variables. The initial set of latent variables could also be considered the latent variables generated on the first run prior to any selections that are made. No limiting definition for initial values was found in the specification.
Sattarov et al. does not teach acquisition of the plurality of latent variables, the generation of the plurality of structural formulas, or the calculation of the plurality of scores, are processing in a parallel manner (Claim 4). Sattarov et al. also does not teach at least the search using the first method in a parallel manner (Claim 11).
Regarding Claim 4, Zhang et al. teach acquisition of the plurality of latent variables, the generation of the plurality of structural formulas, or the calculation of the plurality of scores, are processing in a parallel manner (Page 326, Column 1, Paragraph 4: On the basis of the workflow (Figure 1), four parallel programs, preReceptors, preLigands, VinaLC, and mmgbsa have been developed to perform the calculations at different steps (Figure 2); Page 335, Column 1, Paragraph 4: To implement a docking and rescoring pipeline, such that better lead compounds can be discovered, requires parallel processing). Parallel processes include includes generation structures, scoring, and docking. Additionally, it would be obvious to apply the parallel processing methods of Zhang et al. to any processing step due to the benefits of parallel processing (see reason to combine).
Regarding Claim 11, Zhang et al. teach at least the search using the first method in a parallel manner (Page 326, Column 1, Paragraph 4: On the basis of the workflow (Figure 1), four parallel programs, preReceptors, preLigands, VinaLC, and mmgbsa have been developed to perform the calculations at different steps (Figure 2); Page 335, Column 1, Paragraph 4: To implement a docking and rescoring pipeline, such that better lead compounds can be discovered, requires parallel processing). Parallel processes include includes generation structures, scoring, and docking. Additionally, it would be obvious to apply the parallel processing methods of Zhang et al. to any processing step due to the benefits of parallel processing (see reason to combine).
It would have been obvious to one of ordinary skill in the art at the time of the effective filing date to combine the methods of Sattarov et al. and Zhang et al. because Zhang et al. teach parallel processes greatly benefit the molecule development pipelines, which is the major focus of Sattarov et al. (Page 334, Column 2, Paragraph 3: Why Parallel Programming Is Essential for the Docking and Rescoring Pipeline? High throughput virtual screening of large databases is a very popular practice in computer-aided drug design. However, limited computational resources can cap the database size that can be screened in practice). Additionally, Sattarov et al. suggest using parallel processes for their methods (Page 1186, Column 1, Paragraph 3: Subsequently, it is decoded by four parallel dense layers). Therefore, it would have been obvious to someone of ordinary skill in the art at the time of the effective filling date to combine the methods from the references indicated above. Furthermore, one of ordinary skill in the art would predict that the methods could be readily combined with a reasonable expectation of success because both are within the same technical field – computer modeling and simulations to assess the binging ability of compounds. Accordingly, Claims 1-16 and 18-20 taken as a whole would have been prima facie obvious before the effective filing date and are rejected under 35 U.S.C. 103.
Claims 1-3, 6-10, and 12-20 are rejected under 35 U.S.C. 103 as being unpatentable over Sattarov et al., as applied to claims 1-16 and 18-20 above, in view of Torng and Altman (2019, Journal of Chemical Information and Modeling, Vol. 59: 4131-4149). Italicized text from reference art.
The applicable claims include:
Claims 1-16 and 18-20 are displayed above.
Claim 17. The inferring device according to claim 1, wherein the structural formula is information expressed by a graph.
Regarding Claims 1-3, 6-10, 12-16, and 18-20, the limitation are taught by Sattarov et al. as indicated above.
Sattarov et al. does not teach the structural formula is information expressed by a graph (Claim 17).
Regarding Claim 10, Torng and Altman teach multiple docking methods that include local and global searching (Page 4147, Column 1, Paragraph 3: Such design can simultaneously allow detailed characterization at the local amino acid environment level and global flexibility at the pocket level).
Regarding Claim 17, Torng and Altman teach the structural formula is information expressed by a graph (Page 4133, Column 1, Paragraph 2: Specifically, in Step I, an unsupervised pocket graph autoencoder is trained on a representative druggable pocket set to embed protein pockets into a fixed-size latent space that preserve the pocket properties. In Step II, we constructed a pocket Graph-CNN and a ligand Graph-CNN to extract features from the pocket graphs and 2D ligand graphs, respectively).
It would have been obvious to one of ordinary skill in the art at the time of the effective filing date to combine the methods of Sattarov et al. and Torng and Altman because Torng and Altman teach the methods based on using graphs with an autoencoder structure because of the increased information contained in a graph and successfully demonstrated predicting binding, which is also a major focus of Sattarov et al. (Page 4146, Column 2, Paragraph 3: Our results show that graph-autoencoders can learn meaningful fixed-size representation for protein pockets of varying sizes that reflects protein family similarities. Such representation has the potential to enable efficient pocket similarity search, pocket classification, and can serve as input for downstream machine learning algorithms. We further demonstrate that for the task of predicting protein-ligand interactions, the Graph-CNN framework achieved better or comparable performance to other structure-based methods without relying on target-ligand complexes, and showed advantages over ligand-based methods on challenging cases where actives and decoys are similar in chemical descriptor space). Additionally, Sattarov et al. suggest using graphs for their methods (Page 1183, Column 1, Paragraph 2: An alternative approach based on directly generating molecular graphs instead of SMILES should also be mentioned). Therefore, it would have been obvious to someone of ordinary skill in the art at the time of the effective filling date to combine the methods from the references indicated above. Furthermore, one of ordinary skill in the art would predict that the methods could be readily combined with a reasonable expectation of success because both are based on autoencoders with the same structure in terms of encoder and decoder. Also both are within the same technical field - making predictions using machine learning models based on molecular structure data. Accordingly, Claims 1-3, 6-10, and 12-20 taken as a whole would have been prima facie obvious before the effective filing date and are rejected under 35 U.S.C. 103.
Double Patenting
The nonstatutory double patenting rejection is based on a judicially created doctrine grounded in public policy (a policy reflected in the statute) so as to prevent the unjustified or improper timewise extension of the “right to exclude” granted by a patent and to prevent possible harassment by multiple assignees. A nonstatutory double patenting rejection is appropriate where the conflicting claims are not identical, but at least one examined application claim is not patentably distinct from the reference claim(s) because the examined application claim is either anticipated by, or would have been obvious over, the reference claim(s). See, e.g., In re Berg, 140 F.3d 1428, 46 USPQ2d 1226 (Fed. Cir. 1998); In re Goodman, 11 F.3d 1046, 29 USPQ2d 2010 (Fed. Cir. 1993); In re Longi, 759 F.2d 887, 225 USPQ 645 (Fed. Cir. 1985); In re Van Ornum, 686 F.2d 937, 214 USPQ 761 (CCPA 1982); In re Vogel, 422 F.2d 438, 164 USPQ 619 (CCPA 1970); In re Thorington, 418 F.2d 528, 163 USPQ 644 (CCPA 1969).
A timely filed terminal disclaimer in compliance with 37 CFR 1.321(c) or 1.321(d) may be used to overcome an actual or provisional rejection based on nonstatutory double patenting provided the reference application or patent either is shown to be commonly owned with the examined application, or claims an invention made as a result of activities undertaken within the scope of a joint research agreement. See MPEP § 717.02 for applications subject to examination under the first inventor to file provisions of the AIA as explained in MPEP § 2159. See MPEP § 2146 et seq. for applications not subject to examination under the first inventor to file provisions of the AIA . A terminal disclaimer must be signed in compliance with 37 CFR 1.321(b).
The filing of a terminal disclaimer by itself is not a complete reply to a nonstatutory double patenting (NSDP) rejection. A complete reply requires that the terminal disclaimer be accompanied by a reply requesting reconsideration of the prior Office action. Even where the NSDP rejection is provisional the reply must be complete. See MPEP § 804, subsection I.B.1. For a reply to a non-final Office action, see 37 CFR 1.111(a). For a reply to final Office action, see 37 CFR 1.113(c). A request for reconsideration while not provided for in 37 CFR 1.113(c) may be filed after final for consideration. See MPEP §§ 706.07(e) and 714.13.
The USPTO Internet website contains terminal disclaimer forms which may be used. Please visit www.uspto.gov/patent/patents-forms. The actual filing date of the application in which the form is filed determines what form (e.g., PTO/SB/25, PTO/SB/26, PTO/AIA /25, or PTO/AIA /26) should be used. A web-based eTerminal Disclaimer may be filled out completely online using web-screens. An eTerminal Disclaimer that meets all requirements is auto-processed and approved immediately upon submission. For more information about eTerminal Disclaimers, refer to www.uspto.gov/patents/apply/applying-online/eterminal-disclaimer.
Claims 1, 6-9, and 13-20 are provisionally rejected on the ground of nonstatutory double patenting as being unpatentable over claim 1-2, 8-13, and 15-20 of copending Application No. 18076640 (reference application). Although the claims at issue are not identical, they are not patentably distinct from each other (see table below).
This is a provisional nonstatutory double patenting rejection because the patentably indistinct claims have not in fact been patented.
Instant Application Claims
Anticipating Reference Application 18076640 Claims
1, 19, and 20
1-2, 19, and 20
6
8
7
9
8
10
9
11
13
12
14
13
15
15
16
16
17
17
18
18
Conclusion
No claims are allowed.
Any inquiry concerning this communication or earlier communications from the examiner should be directed to BLAKE H ELKINS whose telephone number is (571)272-2649. The examiner can normally be reached Monday-Thursday 8-5PM.
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, Karlheinz Skowronek can be reached at (571) 272-9047. 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.
/B.H.E./Examiner, Art Unit 1687
/Karlheinz R. Skowronek/Supervisory Patent Examiner, Art Unit 1687
Prosecution Insights