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 pending and are examined on the merits.
Priority
Acknowledgment is made of a claim for foreign to KR10-2022-0078370 filed 06/27/2022. Receipt is acknowledged of certified copies of papers required by 37 CFR 1.55.
As recorded on the 2/1/2023 filing receipt, the instant application claims the benefit of priority to provisional application 63/339,228, filed 05/06/2022. Accordingly, the effective filing date of the claimed invention is 05/06/2022. At this point in examination, all claims have been interpreted as being accorded this priority date. In future actions, the effective filing date of one or more claims may change, due to amendments to the claims, or further analysis of the disclosure(s) of the priority application(s).
Information Disclosure Statement
The information disclosure statement (IDS) submitted on 01/09/2023 is in compliance with the provisions of 37 CFR 1.97. Accordingly, the list of cited references was considered in full by the examiner. A signed copy of the corresponding 1449 form has been included with this Office action.
Drawings
The drawings filed 001/09/2023 are accepted.
Objection to the specification: title
The title should be amended to more specifically reflect the claims, particularly the independent claims and referencing steps/elements: setting the context of the invention, particular to all claims, and distinguishing the instant application from any related applications, for example a title including terms such as: molecule, material and neural network. The title should be "descriptive" and "as... specific as possible" (MPEP 606, 1st para. and 37 CFR 1.72; also MPEP 606.01 pertains).
Claim Interpretation
The following is a quotation of 35 U.S.C. 112(f):
(f) Element in Claim for a Combination. – An element in a claim for a combination may be expressed as a means or step for performing a specified function without the recital of structure, material, or acts in support thereof, and such claim shall be construed to cover the corresponding structure, material, or acts described in the specification and equivalents thereof.
The following is a quotation of pre-AIA 35 U.S.C. 112, sixth paragraph:
An element in a claim for a combination may be expressed as a means or step for performing a specified function without the recital of structure, material, or acts in support thereof, and such claim shall be construed to cover the corresponding structure, material, or acts described in the specification and equivalents thereof.
The claims in this application are given their broadest reasonable interpretation using the plain meaning of the claim language in light of the specification as it would be understood by one of ordinary skill in the art. The broadest reasonable interpretation of a claim element (also commonly referred to as a claim limitation) is limited by the description in the specification when 35 U.S.C. 112(f) or pre-AIA 35 U.S.C. 112, sixth paragraph, is invoked.
As explained in MPEP § 2181, subsection I, claim limitations that meet the following three-prong test will be interpreted under 35 U.S.C. 112(f) or pre-AIA 35 U.S.C. 112, sixth paragraph:
(A) the claim limitation uses the term “means” or “step” or a term used as a substitute for “means” that is a generic placeholder (also called a nonce term or a non-structural term having no specific structural meaning) for performing the claimed function;
(B) the term “means” or “step” or the generic placeholder is modified by functional language, typically, but not always linked by the transition word “for” (e.g., “means for”) or another linking word or phrase, such as “configured to” or “so that”; and
(C) the term “means” or “step” or the generic placeholder is not modified by sufficient structure, material, or acts for performing the claimed function.
Use of the word “means” (or “step”) in a claim with functional language creates a rebuttable presumption that the claim limitation is to be treated in accordance with 35 U.S.C. 112(f) or pre-AIA 35 U.S.C. 112, sixth paragraph. The presumption that the claim limitation is interpreted under 35 U.S.C. 112(f) or pre-AIA 35 U.S.C. 112, sixth paragraph, is rebutted when the claim limitation recites sufficient structure, material, or acts to entirely perform the recited function.
Absence of the word “means” (or “step”) in a claim creates a rebuttable presumption that the claim limitation is not to be treated in accordance with 35 U.S.C. 112(f) or pre-AIA 35 U.S.C. 112, sixth paragraph. The presumption that the claim limitation is not interpreted under 35 U.S.C. 112(f) or pre-AIA 35 U.S.C. 112, sixth paragraph, is rebutted when the claim limitation recites function without reciting sufficient structure, material or acts to entirely perform the recited function.
Claim limitations in this application that use the word “means” (or “step”) are being interpreted under 35 U.S.C. 112(f) or pre-AIA 35 U.S.C. 112, sixth paragraph, except as otherwise indicated in an Office action. Conversely, claim limitations in this application that do not use the word “means” (or “step”) are not being interpreted under 35 U.S.C. 112(f) or pre-AIA 35 U.S.C. 112, sixth paragraph, except as otherwise indicated in an Office action.
Analysis of instant claims
This application includes one or more claim limitations that do not use the word “means,” but are nonetheless interpreted under 35 U.S.C. 112(f) or pre-AIA 35 U.S.C. 112, sixth paragraph, because the claim limitations use a generic placeholder that is coupled with functional language without reciting sufficient structure to perform the recited function and the generic placeholder is not preceded by a structural modifier.
The following claim limitations have been analyzed and interpreted under 112/f.
In claims 5 and 18, the claims recite means (or an equivalent, nonce term, here "first encoder") and function and/or result (here encoding by extracting embedding information), but the recitation does not invoke 112/f because it is interpreted as a well-known process, e.g. the process of using an encoder. [23] of the specification pertains. MPEP 2181.I.A,3rd para. pertains with analogy to structures having "sufficiently definite meaning," such as "filters" and "brakes."
Also in claims 5 and 18, the claims recite means (or an equivalent, nonce term, here "second encoder") and function and/or result (here encoding as trained to extract a first feature), but the recitation does not invoke 112/f because it is interpreted as a well-known process, e.g. the process of using a trained encoder. [23] of the specification pertains. MPEP 2181.I.A,3rd para. pertains with analogy to structures having "sufficiently definite meaning," such as "filters" and "brakes."
In claims 7 and 19, the claims recite means (or an equivalent, nonce term, here "model") and function and/or result (here neural network-based prediction), but the recitation does not invoke 112/f because the claim is interpreted as reciting sufficient structure (in this instance in the form of subsequent algorithmic steps) not to invoke. [9-10] of the specification pertain. MPEP 2181.I.C pertains.
Because these claim limitations are not being interpreted under 35 USC 112(f), they are not being interpreted to cover the corresponding structure, material, or acts described in the specification as performing the claimed function, and equivalents thereof.
If applicant intends to have these limitations interpreted under 35 USC 112(f), applicant may: (1) amend the claim limitations to remove the structure, materials, or acts that perform the claimed function; or (2) present a sufficient showing that the claim limitations do not recite sufficient structure, materials, or acts to perform the claimed function and thus prevent invoking.
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 without significantly more.
The Supreme Court has established a two-step framework for this analysis, wherein a claim does not satisfy § 101 if (1) it is “directed to” a patent-ineligible concept, i.e., a law of nature, natural phenomenon, or abstract idea, and (2), if so, the particular elements of the claim, considered “both individually and as an ordered combination,” do not add enough to “transform the nature of the claim into a patent-eligible application.” Elec. Power Grp., LLC v. Alstom S.A., 830 F.3d 1350, 1353 (Fed. Cir. 2016) (quoting Alice, 134 S. Ct. at 2355). Applicant is also directed to MPEP 2106.
Step 1: The instantly claimed invention (claim(s) 1-13 being representative) is directed to a method and (claim(s) 14-20 being representative) is directed to a system. Therefore, the instantly claimed invention falls into one of the four statutory categories. [Step 1: YES]
Step 2A: First it is determined in Prong One whether a claim recites a judicial exception, and if so, then it is determined in in Prong Two if the recited judicial exception is integrated into a practical application of that exception.
Step 2A, Prong 1: Under the MPEP § 2106.04, the Step 2A (Prong 1) analysis requires determining whether a claim recites an abstract idea, law of nature, or natural phenomenon.
Claims 1, 14, and 15 recite predicting one or more synthetic path descriptor candidates representing multi-step synthetic paths corresponding to the target molecule descriptor using a neural network-based predictive model, the neural network-based predictive model being trained based on a synthetic path descriptor obtained by converting multi-step synthetic path data; the limitation predicting one or more path using a trained neural network model does not provide how the neural network operates or how the prediction is made, and the plain meaning of “predicting” encompasses mental observations or evaluations (e.g., a computer programmer’s mental prediction of path), a process that is practically performed in human mind (mental process).
Claims 5 and 18 recite a first decoder to extract embedding information from the target molecule descriptor; a second encoder that is trained to extract a first feature from the embedding information; a trained decoder to restore a character string corresponding to the synthetic path descriptor based on the first feature and information on second tokens other than the first tokens among the tokens constituting the synthetic path descriptor; the claim does not limit how the trained encoder and decoder functions; the trained encoders and decoder are used to generally apply the abstract idea. See MPEP 2106.05(f); the limitation extract embedding information and a first feature and restore a character, given the plain meaning of extracting and restoring, encompasses observation, evaluation, judgment, and opinion (See MPEP 2106.04(a)(2), subsection III.) performable by human mind (mental process), since human mind is capable of extract and restore data.
Claims 5 and 18 further recite the neural network-based predictive model is trained by updating one or more weights of the predictive model based on a difference between the synthetic path descriptor and the character string; the limitation updating the weights of the predictive model is considered a mathematical calculation, and as such, falls into mathematical concepts groupings of abstract ideas.
Claim 6 recites that the second encoder is configured to extract a feature from a first, second, and a third sequence; the claim does not limit how the encoder functions; the encoders is used to generally apply the abstract idea. See MPEP 2106.05(f); the limitation extracts a feature, given the plain meaning of extracting encompasses observation, evaluation, judgment, and opinion (See MPEP 2106.04(a)(2), subsection III.) performable by human mind (mental process), since human mind is capable of extract data.
Claims 7 and 19 recite a first and second model to predict the synthetic path and a second target molecule; the limitation to predict is considered a mental process.
Claims 8 and 20 recite applying the first target molecule descriptor to the first model (mental process of applying data to a model); determining whether a candidate material corresponding to the synthetic path descriptor is a starting material of which a chemical characteristic and a structure are known (mental process of determining a starting material); predicting the second target molecule descriptor by applying the synthetic path descriptor to the second predictive model in response to a determination that the candidate material is the starting material (mental process of predicting data by applying data); and predicting the one or more synthetic path descriptor candidates based on whether the first target molecule descriptor matches the second target molecule descriptor (mental process of predicting a path).
Claim 9 recites recite classifying one or more characters representing an atom type of the target material and one or more characters representing a chemical bond of the target material in the first target molecule descriptor as tokens (mental process of classifying data); and applying at least a portion of the tokens to the first predictive model (mental process of applying data to a model).
Claim 10 recites determining whether the candidate material is the starting material using data registered in a material database (mental process of determining a starting material).
Claim 11 recites removing, in response to a determination that the candidate material is not the starting material, the synthetic path descriptor corresponding to the candidate material from the synthetic path descriptor candidates (mental process of removing data).
Claim 12 recites providing a synthesis recipe corresponding to the target material based on the synthetic path descriptor candidates (mental process of providing a recipe based on the result of an analysis).
Claim 13 recites converting the multi-step synthetic path data into the synthetic path descriptor in a form of a character string (mental process of converting one type of data to another).
Claims 2, 3, 4, and 16-17 provide additional information about the judicial exceptions.
The identified claims recite a law of nature, a natural phenomenon (product of nature) and/or fall into one of the groups of abstract ideas of mathematical concepts, mental processes, and/or certain methods of organizing human activity for the reasons set forth above. See MPEP 2106.04 (a)(2) III and MPEP 2106.04 (b) I. Therefore, claims are directed to one or more judicial exception(s) and require further analysis in Prong Two. [Step 2A, Prong 1: YES]
Step 2A: Prong 2: Under the MPEP § 2106.04, the Step 2A, Prong 2 analysis requires identifying whether there are any additional elements recited in the claim beyond the judicial exception(s), and evaluating those additional elements to determine whether they integrate the exception into a practical application of the exception. This judicial exception is not integrated into a practical application for the following reasons.
The additional elements of claim(s) 1-20 include the following.
Claims 1, 14, and 15 recite receiving a target molecule descriptor; using a trained neural network, outputting the one or more synthetic path descriptor candidates.
Claims 5 and 18 recite a first and second trained encoder, a trained decoder.
Claim 7 recites receive a first target molecule descriptor, receive the synthetic path descriptor.
Claim 8 recites acquiring the synthetic path descriptor.
Claim 13 recites receiving the multi-step synthetic path data.
Claim 14 recites a non-transitory computer-readable storage medium storing instructions that, when executed by at least one processor, cause the at least one processor.
Claim 15 recites an apparatus for predicting a synthesis path, the apparatus comprising: a user interface configured to receive a target molecule descriptor corresponding to a target material; at least one memory storing at least one program; and at least one processor configured to execute the at least one program.
The additional elements of an apparatus comprising a user interface, memory, a processor, a program, and a non-transitory computer-readable storage medium are generic computer components and/or processes. There are no limitations that indicate that a user interface, memory, a processor, a program, and a non-transitory computer-readable storage medium require anything other than generic computing systems. The courts have found the use of a computer or other machinery in its ordinary capacity for economic or other tasks (e.g., to receive, store, or transmit data) or simply adding a general-purpose computer or computer components after the fact to an abstract idea (e.g., a fundamental economic practice or mathematical equation) does not integrate a judicial exception into a practical application. See MPEP 2106.05(f).
Furthermore, the additional elements of receiving/acquiring data and outputting amount to necessary data gathering and outputting. The courts have found the limitations that amount to necessary data gathering and outputting are insignificant extra-solution activity that do not integrate a recited judicial exception into a practical application in Mayo, 566 U.S. at 79, 101 USPQ2d at 1968 and O/P Techs., Inc. v. Amazon.com, Inc., 788 F.3d 1359, 1363, 115 USPQ2d 1090, 1092-93 (Fed. Cir. 2015) (see MPEP 2106.05(g)).
Furthermore, the additional element of using a trained neural network, a first and second trained encoder, and a trained decoder merely indicates a field of use or technological environment in which the judicial exception is performed and thus fails to add an inventive concept to the claims. See MPEP 2106.05(h).
Therefore, the additionally recited elements amount to insignificant extra-solution activity and, as such, the claims as a whole do no integrate the abstract idea into practical application.
MPEP 2106.04(d).I lists the following example considerations for evaluating whether a judicial exception is integrated into a practical application:
An improvement in the functioning of a computer or an improvement to other technology or another technical field, as discussed in MPEP §§ 2106.04(d)(1) and 2106.05(a);
Applying or using a judicial exception to effect a particular treatment or prophylaxis for a disease or medical condition, as discussed in MPEP § 2106.04(d)(2);
Implementing a judicial exception with, or using a judicial exception in conjunction with, a particular machine or manufacture that is integral to the claim, as discussed in MPEP § 2106.05(b);
Effecting a transformation or reduction of a particular article to a different state or thing, as discussed in MPEP § 2106.05(c); and
Applying or using the judicial exception in some other meaningful way beyond generally linking the use of the judicial exception to a particular technological environment, such that the claim as a whole is more than a drafting effort designed to monopolize the exception, as discussed in MPEP § 2106.05(e).
In Step 2A, Prong 1 above, claim steps and/or elements were identified as part of one or more judicial exceptions (JEs).
In Step 2B below, any remaining steps and/or elements are therefore in addition to the identified JE(s). Any such additional steps and additional elements are further discussed in Step 2B.
Here in Step 2A, Prong 2, no additional step or element clearly demonstrates integration of the JE(s) into a practical application.
At this point in examination it is not yet the case that any of the Step 2A, Prong 2 considerations enumerated above clearly demonstrates integration of the identified JE(s) into a practical application. Referring to the considerations above, none of 1. an improvement, 2. treatment, 3. a particular machine or 4. a transformation is clear in the record.
For example, regarding the first consideration at MPEP 2106.04(d)(1), the record, including for example the specification, does not yet clearly disclose an explanation of improvement over the previous state of the technology field. The claims do not yet clearly result in such an improvement (e.g. specification: [4, 87]).
In conclusion regarding Prong 2, claims 1-20 are directed to an abstract idea. [Step 2A, Prong 2: NO]
Step 2B: In the second step it is determined whether the claimed subject matter includes additional elements that amount to significantly more than the judicial exception. An inventive concept cannot be furnished by an abstract idea itself. See MPEP § 2106.05.
The additional elements of claim(s) 1-20 include the following.
Claims 1, 14, and 15 recite receiving a target molecule descriptor; using a trained neural network, outputting the one or more synthetic path descriptor candidates.
Claims 5 and 18 recite a first and second trained encoder, a trained decoder.
Claim 7 recites receive a first target molecule descriptor, receive the synthetic path descriptor.
Claim 8 recites acquiring the synthetic path descriptor.
Claim 13 recites receiving the multi-step synthetic path data.
Claim 14 recites a non-transitory computer-readable storage medium storing instructions that, when executed by at least one processor, cause the at least one processor.
Claim 15 recites an apparatus for predicting a synthesis path, the apparatus comprising: a user interface configured to receive a target molecule descriptor corresponding to a target material; at least one memory storing at least one program; and at least one processor configured to execute the at least one program.
The additional elements of an apparatus comprising a user interface, memory, a processor, a program, and a non-transitory computer-readable storage medium are conventional computer components and/or processes. The courts have found the use of a computer or other machinery in its ordinary capacity for economic or other tasks (e.g., to receive, store, or transmit data) or simply adding a general-purpose computer or computer components after the fact to an abstract idea (e.g., a fundamental economic practice or mathematical equation) does not provide significantly more. See Affinity Labs v. DirecTV, 838 F.3d 1253, 1262, 120 USPQ2d 1201, 1207 (Fed. Cir. 2016) (cellular telephone); TU Communications LLC v. AV Auto, LLC, 823 F.3d 607,613,118 USPQ2d 1744, 1748 (Fed. Cir. 2016) (computer server and telephone unit).
Furthermore, the additional elements of receiving/acquiring data and outputting amount to necessary data gathering and outputting. The courts have found the limitations that amount to necessary data gathering and outputting are insignificant extra-solution activity that do not amount to significantly more (see MPEP 2106.05(g)).
Furthermore, the additional element of using a trained neural network, a first and second trained encoder, and a trained decoder merely indicates field of use or technological environment in which the judicial exception is performed and thus fails to add an inventive concept to the claims. See MPEP 2106.05(h).
Therefore, the additional element is not sufficient to amount to significantly more than the judicial exception.
Taken alone, the additional elements do not amount to significantly more than the above-identified judicial exception(s). Even when viewed as a combination, the additional elements fail to transform the exception into a patent-eligible application of that exception. Thus, the claims as a whole do not amount to significantly more than the exception itself. [Step 2B: NO]
Therefore, the instantly rejected claims are not drawn to eligible subject matter as they are directed to an abstract idea without significantly more.
Claim Rejections - 35 USC § 103
In the event the determination of the status of the application as subject to AIA 35 U.S.C. 102 and 103 (or as subject to pre-AIA 35 U.S.C. 102 and 103) is incorrect, any correction of the statutory basis (i.e., changing from AIA to pre-AIA ) for the rejection will not be considered a new ground of rejection if the prior art relied upon, and the rationale supporting the rejection, would be the same under either status.
The following is a quotation of 35 U.S.C. 103 which forms the basis for all obviousness rejections set forth in this Office action:
A patent for a claimed invention may not be obtained, notwithstanding that the claimed invention is not identically disclosed as set forth in section 102, if the differences between the claimed invention and the prior art are such that the claimed invention as a whole would have been obvious before the effective filing date of the claimed invention to a person having ordinary skill in the art to which the claimed invention pertains. Patentability shall not be negated by the manner in which the invention was made.
The factual inquiries for establishing a background for determining obviousness under 35 U.S.C. 103 are summarized as follows:
1. Determining the scope and contents of the prior art.
2. Ascertaining the differences between the prior art and the claims at issue.
3. Resolving the level of ordinary skill in the pertinent art.
4. Considering objective evidence present in the application indicating obviousness or nonobviousness.
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 1-3 and 12-16 are rejected under 35 U.S.C. 103 as being unpatentable over Yang (US12027240B2 as cited on the attached Form 892), in view of Manica (US20220359045A1 as cited on the attached Form 892).
Regarding claims 1, 14, and 15, Yang discloses a retrosynthesis processing method and apparatus, the method comprising: determining molecular representation information of a target molecule; inputting the molecular representation information (for example, synthetic path descriptor) into a target neural network; and performing, via the target neural network, retrosynthesis processing on the target molecule based on the molecular representation information of the target molecule, to obtain a respective retrosynthesis reaction of the target molecule for each step of the retrosynthesis processing wherein the target neural network is obtained by training a predetermined neural network according to a sample cost dictionary that is generated by concurrently performing retrosynthesis reaction training on each of a plurality of sample molecules, and the respective retrosynthesis reaction is performed according to a preset retrosynthesis reaction architecture (for example, training on synthetic path descriptor obtained by converting multi-step synthetic path data), (claim 1). Yang further discloses providing an effective retrosynthesis route when performing retrosynthesis processing on the target molecule, to accurately obtain the retrosynthesis reaction in each step, thereby implementing the automated design of the retrosynthesis route (for example, outputting synthetic path descriptors) (col. 21, para. 3); reading on limitations of a method of predicting a synthetic path, the method comprising: receiving a target molecule descriptor corresponding to a target material; predicting one or more synthetic path descriptor candidates representing multi-step synthetic paths corresponding to the target molecule descriptor using a neural network-based predictive model, and outputting the one or more synthetic path descriptor candidates.
Further regarding limitations of the neural network-based predictive model being trained based on a synthetic path descriptor obtained by converting multi-step synthetic path data Manica discloses a method for predicting at least one aspect of a chemical reaction (abstract). Manica further discloses that the computer-implemented method may be based on pretrained natural language processing models (e.g., a transformer model) that takes string representations of educts (SMILES separated by specific characters, e.g., “.”) and predicts the product (SMILES) of the chemical reaction of interest to explore synthesis routes in the biochemical space. Once the model is pretrained it can be used for de novo enzyme design fixing the information of the educts and products of interest [0057].
Manica further discloses that the encoder uses deep neural network layers and converts the input tokens (symbols or groups of symbols obtained by tokenizing the one or more string representation of substances provided as input to corresponding hidden vectors [0106] [0130].
Regarding claims 2, 3 and 16, Yang discloses that the molecular representation information may be a simplified molecular-input line-entry system that clearly describes a molecular structure by using ASCII strings, that is, describing a three-dimensional chemical structure by using a string of characters. A charge density, free valence, and a bond level are all closely related to the properties of a molecule (for example, different tokens) (col. 7, para. 1); reading on limitations of wherein the synthetic path descriptor comprises one or more tokens among: a molecular structure descriptor representing molecular structure information of reactants; a notation character for distinguishing synthesis steps before and after synthesis of the multi-step synthetic paths; a delimiting parenthesis for defining a synthesis order of the reactants used for each synthesis step of the multi-step synthetic paths; a separator for distinguishing between the molecular structure descriptor and the delimiting parenthesis or distinguishing between a plurality of molecular structure descriptors; and a reaction descriptor corresponding to a reaction scheme used for each synthesis step of the multi-step synthetic paths and wherein the molecular structure descriptor comprises a token representing at least one of types of atoms included in the reactants, bonding information comprising a bonding structure of the atoms, an aromatic compound corresponding to the molecular structure information, or an isomer corresponding to the molecular structure information.
Regarding claim 12, Manica discloses that the method can be used to plan various synthesis steps [0052]. Manica further discloses exploring the enzyme space in various ways, for example, propose synthesis route to generate desired products [0056]. Manita further discloses that once the model is pretrained it can be used for de novo enzyme design fixing the information of the educts and products of interest [0057-0062]; reading on limitations of providing a synthesis recipe corresponding to the target material based on the synthetic path descriptor candidates.
Regarding claim 13, Manica discloses the SMILES sequence of the educts is given as follows: “C(C(C(C(C(CO)O)O)O)O)O.O” and the abbreviated amino acid sequence is given as follows: “MPGQQATKHE . . . DGGYTTR”. The product prediction model then predicts the product as “C1C(C(C(C(O1)(CO)O)O)O)O” [0062], where the method may be used to predict complex, multi-step synthesis plans [0067]; reading on limitations of receiving the multi-step synthetic path data corresponding to chemical materials; and converting the multi-step synthetic path data into the synthetic path descriptor in a form of a character string.
In KSR Int 'l v. Teleflex, the Supreme Court, in rejecting the rigid application of the teaching, suggestion, and motivation test by the Federal Circuit, indicated that “The principles underlying [earlier] cases are instructive when the question is whether a patent claiming the combination of elements of prior art is obvious. When a work is available in one field of endeavor, design incentives and other market forces can prompt variations of it, either in the same field or a different one. If a person of ordinary skill can implement a predictable variation, § 103 likely bars its patentability.” KSR Int'l v. Teleflex lnc., 127 S. Ct. 1727, 1740 (2007).
Applying the KSR standard to Yang and Manica, the examiner concludes that this combination contained comparable method that was improved in the same way as the claimed invention. Both Yang and Manica are directed to predicting synthetic paths. Yang only disclosed employing a neural network predictive model to predict synthetic path descriptor candidates. In the same field of research, Manica provided the more advanced neural network type, transformer-based neural network that was trained using converted synthetic path data. Combining the synthetic path neural network prediction model of Yang with the more advanced transformer-based neural network of Manica would have allowed for assembling a template-free multi-step retrosynthesis planning and more accurate predictive results. One ordinary skilled in the art would have been capable of applying enhanced method of Manica to the base method of Yang and the results would have been predictable to one ordinary skilled in the art. Therefore, the invention would have been prima facie obvious to one of skill in the art before the effective filing date of the claimed invention, absent evidence to the contrary.
Claims 4-6, and 17-18 are rejected under 35 U.S.C. 103 as being unpatentable over Yang, in view of Manica, as applied to claims 1-3, 12-16 above, in view of Irwin (Chemformer: a pre-trained transformer for computational chemistry, Machine Learning: Science and Technology. 3 (2022) 015022, pages 1-13 as cited on the attached Form 892), and further in view of Prat (US20230290114A1 as cited on the attached Form 892).
Claims 4-6, and 17-18, depends on claims 3 and 15, respectively. Limitations of claims 3 and 15 have been taught in the above rejections.
Regarding claims 4, 6, and 17, Yang discloses determining the molecular representation of the molecule/the simplified molecular-input line-entry systems (col. 16, para.4); Manica discloses the SMILES representation and tokenization of the descriptor [0112] and that the architecture is an encoder-decoder architecture [0041].
Further regarding limitations of the synthetic path descriptor in which a portion of tokens are masked, Irwin discloses the Chemformer model—a Transformer-based model which can be quickly applied to both sequence-to-sequence and discriminative cheminformatics tasks. Irwin further discloses that self-supervised pre-training can improve performance and significantly speed up convergence on downstream tasks (abstract). Irwin further discloses randomly masking input tokens and randomly generating another SMILE string (pg. 3, Fig. 2).
Regarding claims 5 and 18, Tang discloses multi-step retrosynthesis processing using a neural network (abstract); Manica discloses a transformer-based multi-step retrosynthesis processing with an encoder-decoder architecture. Irwin discloses using encoder stack, where the tokenized SMILES string of a molecule is prefixed with one or more task tokens and this sequence of tokens is passed through the model’s embedding layer, followed by the model’s encoder (pg. 5, para. 4).
Prat discloses a system and method for pharmacophore-conditioned generation of molecules that modifies a conditional variational autoencoder (CVAE) such that the latent space in generation of a molecule is not conditioned on the pharmacophore space of the molecule, allowing for generation of pharmacophore descriptors independently from the conditional on which CVAE has been trained, removing a substantial impediment to the use of CVAEs for exploration of pharmacophore descriptors of a molecule (abstract).
Prat further discloses that the molecular encoder comprising a first plurality of programming instructions to receive a molecular conditional distribution representing a range of distributions of the molecular structure of a molecule; and encode the molecular conditional distribution as a first vector representation (for example, sequence embedding); and a latent mapper comprising receive the encoded molecular conditional distribution along with a plurality of molecule representation samples (for example, sequence feature extraction) from a set comprising probable representations of the molecule [0017].
Prat further discloses that the decoder portion recreates the input molecule [0129]. Prat further discloses updating weights iteratively to minimize the losses [0147], where the output of the decoder is used to map atomic features such as atom density in latent space [0141], and where model may learn a latent distribution that governs molecular properties and provide a decoder which can construct chemically valid molecules from samples of the prior 1505. Latent samples are passed through a sequence of dense layers, after which the two different matrices (node feature matrix, N and edge feature tensor) are used to reconstruct the node feature and edge feature matrices [0135]; reading on limitations of first and second encoder for sequence embedding and feature extraction, respectively, and a trained decoder and updating weights.
Applying the KSR standard to Yang, Manica, Irwin, and Prat, examiner concludes that this combination contained comparable method that was improved in the same way as the claimed invention. Yang, Manica, and Irwin are directed to predicting synthetic paths. Yang and Manica disclosed employing a transformer-based neural network that was trained using converted synthetic path data to predict synthetic path descriptor candidates. In the same field of research, Irwin discloses the known technique of randomly masking tokens to force the model to understand deep, underlying contextual relationships. Prat disclosed the known technique of employing two encoders one for sequence embedding and one for extracting sequence features for computational efficiency and handling variable lengths. Combining the synthetic path neural transformer-based neural network prediction model of Yang and Manica with the known masking technique of Irwin and known usage of two encoders would have prevented the model from overfitting to specific sentence structure, making it robust when applied to real-world data. One ordinary skilled in the art would have been capable of applying enhanced method of Irwin and Prat to the base method of Yang and Manica and the results would have been predictable to one ordinary skilled in the art. Therefore, the invention would have been prima facie obvious to one of skill in the art before the effective filing date of the claimed invention, absent evidence to the contrary.
Claims 7 and 8 are rejected under 35 U.S.C. 103 as being unpatentable over Yang, in view of Manica, as applied to claims 1-3, 12-16 above, in view of Tal (US20230038256A1 as cited on the attached Form 892).
Claims 7 and 8 depends on claim 1. Limitations of claim 1 have been taught in the above rejections.
Regarding claim 7, Manica discloses that another model, e.g., the product prediction model, and/or multiple models are used in each step. For example, a product prediction module may be used for predicting the products of a first chemical reaction and the product prediction model may be used in a second iteration to predict the products to be generated using the products of the first reaction as the educts of the second reaction [0068] [0122].
Tal discloses a system and method that given one or more input molecules, produces a contextualized summary of characteristics of related target molecules, e.g., proteins (abstract).
Tal further discloses retrosynthesis for de novo drug design. one approach begins with preprocessing all the SMILES representations for reactants and products to convert to canonical form (SMILES to Mol & Mol to SMILES through a cheminformatics toolkit), remove duplicates & clean the data, augmenting SMILE equivalents via enumeration. Then, transformer models are used with multiple attention heads and a k-beam search is set up. Further, the models are conformed by optimizing on producing long-term reactants, ensuring the models are robust to different representations of a molecule, providing intrinsic recursion (using performers), and including further reagents such as catalysts and solvents [0131]; reading on limitations of wherein the neural network-based predictive model comprises: a first predictive model configured to receive a first target molecule descriptor corresponding to the target material and predict the synthetic path descriptor corresponding to the target material; and a second predictive model configured to receive the synthetic path descriptor and predict a second target molecule descriptor corresponding to the target material.
Regarding claim 8, Manica discloses another model, e.g., the product prediction model, and/or multiple models are used in each step. For example, a product prediction module may be used for predicting the products of a first chemical reaction and the product prediction model may be used in a second iteration to predict the products to be generated using the products of the first reaction as the educts of the second reaction [0068]. Manica further discloses that the synthesis plan is considered to be completed and the recursive use is automatically ended once a termination condition is met. The terminating condition is selected from a group consisting of: all predicted educts are commercially available, all predicted educts are non-toxic, all predicted educts are water soluble, all predicted educts meet a predefined requirement (e.g., in respect to purity, availability, price, storability), and a combination of two or more of the aforementioned conditions [0069]; reading on limitations of wherein the predicting of the one or more synthetic path descriptor candidates comprises: applying the first target molecule descriptor to the first predictive model; acquiring the synthetic path descriptor predicted by the first predictive model based on the first target molecule descriptor; determining whether a candidate material corresponding to the synthetic path descriptor is a starting material of which a chemical characteristic and a structure are known; predicting the second target molecule descriptor by applying the synthetic path descriptor to the second predictive model in response to a determination that the candidate material is the starting material; and predicting the one or more synthetic path descriptor candidates based on whether the first target molecule descriptor matches the second target molecule descriptor.
Applying the KSR standard to Yang, Manica, and Tal, examiner concludes that this combination contained comparable method that was improved in the same way as the claimed invention. Yang, Manica, and Tal are directed to predicting synthetic paths. Yang and Manica disclosed employing a transformer-based neural network that was trained using converted synthetic path data to predict synthetic path descriptor candidates and that multiple models can be used for each step of the process. In the same field of research, Tal disclosed the specifics of the first and second model to increase chemical plausibility. Combining the synthetic path neural transformer-based neural network prediction model of Yang and Manica with the known usage of multiple models of Tal would have enabled diverse route generation and would have prevented overfitting. One ordinary skilled in the art would have been capable of applying enhanced method of Tal to the base method of Yang and Manica and the results would have been predictable to one ordinary skilled in the art. Therefore, the invention would have been prima facie obvious to one of skill in the art before the effective filing date of the claimed invention, absent evidence to the contrary.
Claims 19-20 are rejected under 35 U.S.C. 103 as being unpatentable over Yang, in view of Manica, as applied to claims 1-3, 12-16 above, in view of Irwin, in view of Prat, and further in view of Tal.
Claims 19 and 20 depend on claim 17. Limitations of claim 17 have been taught in the above rejections.
Regarding claim 19, Manica discloses that another model, e.g., the product prediction model, and/or multiple models are used in each step. For example, a product prediction module may be used for predicting the products of a first chemical reaction and the product prediction model may be used in a second iteration to predict the products to be generated using the products of the first reaction as the educts of the second reaction [0068] [0122].
Tal discloses a system and method that given one or more input molecules, produces a contextualized summary of characteristics of related target molecules, e.g., proteins (abstract).
Tal further discloses retrosynthesis for de novo drug design, two approaches are described below. A first approach begins with preprocessing all the SMILES representations for reactants and products to convert to canonical form (SMILES to Mol & Mol to SMILES through a cheminformatics toolkit), remove duplicates & clean the data, augmenting SMILE equivalents via enumeration. Then, transformer models are used with multiple attention heads and a k-beam search is set up. Further, the models are conformed by optimizing on producing long-term reactants, ensuring the models are robust to different representations of a molecule, providing intrinsic recursion (using performers), and including further reagents such as catalysts and solvents [0131].
Regarding claim 20, Manica discloses another model, e.g., the product prediction model, and/or multiple models are used in each step. For example, a product prediction module may be used for predicting the products of a first chemical reaction and the product prediction model may be used in a second iteration to predict the products to be generated using the products of the first reaction as the educts of the second reaction [0068]. Manica further discloses that the synthesis plan is considered to be completed and the recursive use is automatically ended once a termination condition is met. The terminating condition is selected from a group consisting of: all predicted educts are commercially available, all predicted educts are non-toxic, all predicted educts are water soluble, all predicted educts meet a predefined requirement (e.g., in respect to purity, availability, price, storability), and a combination of two or more of the aforementioned conditions [0069].
Applying the KSR standard to Yang, Manica, Irwin, Prat, and Tal, examiner concludes that this combination contained comparable method that was improved in the same way as the claimed invention. Yang, Manica, Irwin, Prat, and Tal are directed to predicting synthetic paths. Yang and Manica disclosed employing a transformer-based neural network that was trained using converted synthetic path data to predict synthetic path descriptor candidates. In the same field of research, Irwin discloses the known technique of randomly masking tokens to force the model to understand deep, underlying contextual relationships. Prat disclosed the known technique of employing two encoders one for sequence embedding and one for extracting sequence features for computational efficiency and handling variable lengths. In the same field of research, Tal disclosed the specifics of the first and second model to increase chemical plausibility. Combining the synthetic path neural transformer-based neural network prediction model of Yang, Manica, Irwin, Prat with the known usage of multiple models of Tal would have enabled diverse route generation and would have prevented overfitting. One ordinary skilled in the art would have been capable of applying enhanced method of Tal to the base method of Yang, Manica, Irwin, and Prat and the results would have been predictable to one ordinary skilled in the art. Therefore, the invention would have been prima facie obvious to one of skill in the art before the effective filing date of the claimed invention, absent evidence to the contrary.
Claims 9-11 are rejected under 35 U.S.C. 103 as being unpatentable over Yang, in view of Manica, as applied to claims 1-3, 12-16 above, in view of Irwin, in view of Prat, and further in view of Banatao (US 12499419 B2 as cited on the attached Form 892).
Claims 9-11 depend on claim 8. Limitations of claim 8 have been taught in the above rejections.
Regarding claim 9, Yang discloses that the molecular representation information may be a simplified molecular-input line-entry system that clearly describes a molecular structure by using ASCII strings, that is, describing a three-dimensional chemical structure by using a string of characters. A charge density, free valence, and a bond level are all closely related to the properties of a molecule (for example, different tokens) (col. 7, para. 1). Banatao discloses a method for generating spectroscopic data includes inputting, by a computing device, a text string comprising a structural representation of a material to an encoder of a natural language processing (NLP) model implemented with a deep neural network. The method includes generating, using the encoder of the NLP model, an encoded representation of the text string. The text string may include latent chemical bond information of the material (abstract; claim 1).
Banatao further discloses that the text string is a first text string, the material is a first material, the encoded representation is a first encoded representation, and the spectrum array is a first spectrum array, and the method includes inputting a second text string including a structural representation of a second material to the encoder of the NLP model, the second material including an additive (col. 2, para. 3).
Banatao further discloses two encoders in the seq-to-seq model architecture (col. 16, para, 2-3; figure 2), and a decoder that includes a self-attention layer and a feed-forward neural network (col. 15; L. 1-13).
Banatao further discloses processing the output of the encoder by a decoder to generate a new text string comprising a new structural representation of the material; wherein the new text string includes one or more tokens based at least in part on the latent chemical bond information (claim 10); reading on limitations of wherein the applying of the first target molecule descriptor to the first predictive model comprises: classifying one or more characters representing an atom type of the target material and one or more characters representing a chemical bond of the target material in the first target molecule descriptor as tokens; and applying at least a portion of the tokens to the first predictive model.
Regarding claim 10, Banatao discloses that the system implementing the trained encoder 224 may obtain sample data 226 from which to generated encoded representations 228. The sample data 226 can be categorized by material classification, for example, when the trained encoder 224 has been trained to predict spectra for materials within a specific material classification. The sample data 226 can be obtained from material databases (col. 17, para.1); reading on limitations of wherein the determining of whether the candidate material is the starting material comprises: determining whether the candidate material is the starting material using data registered in a material database.
Regarding claim 11, Banatao discloses input/output mapping using vector operations, machine learning models implemented in artificial neural networks trained to generate predicted spectral features from the encoded representation 228, or pre-processing operations (for example, removing unmatched descriptors) to reduce the size of the encoded representation (col. 17, para. 2; figure 2); reading on limitations of removing, in response to a determination that the candidate material is not the starting material, the synthetic path descriptor corresponding to the candidate material from the synthetic path descriptor candidates.
Applying the KSR standard to Yang, Manica, Irwin, Prat, and Banatao, examiner concludes that this combination contained comparable method that was improved in the same way as the claimed invention. Yang, Manica, Irwin, Prat, and Banatao are directed to predicting synthetic paths. Yang and Manica disclosed employing a transformer-based neural network that was trained using converted synthetic path data to predict synthetic path descriptor candidates. In the same field of research, Irwin discloses the known technique of randomly masking tokens to force the model to understand deep, underlying contextual relationships. Prat disclosed the known technique of employing two encoders one for sequence embedding and one for extracting sequence features for computational efficiency and handling variable lengths. In the same field of research, Banatao disclosed classifying characters of the descriptor as an atom type and and a chemical bond of the target material. Combining the synthetic path neural transformer-based neural network prediction model of Yang, Manica, Irwin, Prat with the known technique of classifying descriptors based on atoms and chemical bonds of Banatao would have minimized grammatically invalid outputs. One ordinary skilled in the art would have been capable of applying enhanced method of Banatao to the base method of Yang, Manica, Irwin, and Prat and the results would have been predictable to one ordinary skilled in the art. Therefore, the invention would have been prima facie obvious to one of skill in the art before the effective filing date of the claimed invention, absent evidence to the contrary.
Conclusion
No claims are allowed.
Any inquiry concerning this communication or earlier communications from the examiner should be directed to GHAZAL SABOUR whose telephone number is (703)756-1289. The examiner can normally be reached M-F 7:30-5:00.
Examiner interviews are available via telephone, in-person, and video conferencing using a USPTO supplied web-based collaboration tool. To schedule an interview, applicant is encouraged to use the USPTO Automated Interview Request (AIR) at http://www.uspto.gov/interviewpractice.
If attempts to reach the examiner by telephone are unsuccessful, the examiner’s supervisor, Larry D. Riggs can be reached at (571) 270-3062. The fax phone number for the organization where this application or proceeding is assigned is 571-273-8300.
Information regarding the status of published or unpublished applications may be obtained from Patent Center. Unpublished application information in Patent Center is available to registered users. To file and manage patent submissions in Patent Center, visit: https://patentcenter.uspto.gov. Visit https://www.uspto.gov/patents/apply/patent-center for more information about Patent Center and https://www.uspto.gov/patents/docx for information about filing in DOCX format. For additional questions, contact the Electronic Business Center (EBC) at 866-217-9197 (toll-free). If you would like assistance from a USPTO Customer Service Representative, call 800-786-9199 (IN USA OR CANADA) or 571-272-1000.
/G.S./Examiner, Art Unit 1686
/G. STEVEN VANNI/Primary patents examiner, Art Unit 1686
Prosecution Insights