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 .
Applicant's response filed 06/24/2026 has been fully considered. The following rejections
and/or objections are either reiterated or newly applied.
Status of Claims
Claims 1-5, 7-8, 10, 12-13, and 15-21 pending and examined on the merits.
Claims 6, 9, 11, and 14 canceled (claim 14 is newly canceled).
Priority
The instant application filed on 12/3/2021 is a 371 national stage entry of PCT/JP2020/022576 having an international filing date of 6/8/2020, and claims the benefit of priority to Japanese Patent Application No. 2019-106814 filed on 6/7/2019. Thus, the effective filing date of the claims is 6/7/2019.
The applicant is reminded that amendments to the claims and specification must comply with 35 U.S.C. § 120 and 37 C.F.R. § 1.121 to maintain priority to an earlier-filed application. Claim amendments may impact the effective filing date if new subject matter is introduced that lacks support in the originally filed disclosure. If an amendment adds limitations that were not adequately described in the parent application, the claim may no longer be entitled to the priority date of the earlier filing.
Information Disclosure Statement
The IDS filed on 6/24/2026 has been entered and considered. A signed copy of the corresponding 1449 form has been included with this Office action.
Specification
The objection to the specification is withdrawn in view of Applicant's amendments to para.0309 filed on 6/24/2026.
Claim Interpretation
Claims 1-2, 13-15 no longer invoke 35 USC 112(f) because the amendments to the claims, filed 6/24/2026, now include sufficient structure for performing the recited steps. Claim 14 no longer invokes due to being canceled.
Withdrawn Rejections
35 USC § 112(a)
The rejection of claims 1-2, 13-15 under 35 USC 112(a) withdrawn in view of Applicant's claim amendments and remarks filed on 6/24/2026. Claim 14 is canceled.
35 USC § 112(b)
The rejection of claims 1-5, 7-8, 10, and 12-21 under 35 USC 112(b) withdrawn in view of Applicant's claim amendments and remarks filed on 6/24/2026. Claim 14 is canceled.
35 USC § 101
The rejection of claim 14 under 35 USC 101 withdrawn due to being canceled.
35 USC § 103
The rejection of claim 14 under 35 USC 103 withdrawn due to being canceled.
Claim Rejections - 35 USC § 112
The following is a quotation of 35 U.S.C. 112(b):
(b) CONCLUSION.—The specification shall conclude with one or more claims particularly pointing out and distinctly claiming the subject matter which the inventor or a joint inventor regards as the invention.
The following is a quotation of 35 U.S.C. 112 (pre-AIA ), second paragraph:
The specification shall conclude with one or more claims particularly pointing out and distinctly claiming the subject matter which the applicant regards as his invention.
Claims 3 and 16-21 rejected under 35 U.S.C. 112(b) or 35 U.S.C. 112 (pre-AIA ), second paragraph, as being indefinite for failing to particularly point out and distinctly claim the subject matter which the inventor or a joint inventor (or for applications subject to pre-AIA 35 U.S.C. 112, the applicant), regards as the invention. The rejection of claims 3-5 and 16-21 under 35 USC 112(b) are newly recited necessitated by claim amendments.
Claim 3 recites "the characteristic of the sequences is a characteristic including positions of constituent elements in the sequences and anteroposterior relationship of the constituent elements". There is insufficient antecedent basis for the two instances of "the sequences". It is not clear if "the sequences" here are referring to the set of "the first sequences", "the second sequences", "the virtual mutated sequence information", "sequence information of an actual antigen-binding molecule", "the new virtual mutated sequence information", or "a target antigen-binding molecule represented by a virtual sequence". To further prosecution, "the sequences" is interpreted as "sequence information of an actual antigen-binding molecule", based on the context of sequence characteristics from claim 1.
Claims 16, 18, and 20 recite "the antigen-binding molecules are acquired from a sequence of panning in which a subsequent round of panning is carried out with a target antigen for the antigen-binding molecules that have appeared in an operation of panning in the previous round". It is not clear if "the antigen-binding molecules" are from the first or second sequence information that is acquired in claim 1, 13, and 14, respectively.
Claims 17, 19, and 21 depend from claims 16, 18, and 20, respectively, and therefore are also rejected under 35 USC 112(b).
Response to Arguments under 35 USC § 112
Applicant’s arguments filed 6/24/2026 are fully considered and were found persuasive: 112f) Claims 1-2 and 13-15 no longer invoke 35 USC 112(f) because the amendments to the claims, filed 6/24/2026, now include sufficient structure for performing the recited steps. Claim 14 no longer invokes due to being canceled; 112a) The rejection of claims 1-2, 13-15 under 35 USC 112(a) withdrawn in view of Applicant's claim amendments and remarks filed on 6/24/2026. Claim 14 is canceled; and 112b) The rejection of claims 1-5, 7-8, 10, and 12-21 under 35 USC 112(b) withdrawn in view of Applicant's claim amendments and remarks filed on 6/24/2026. Claim 14 is canceled.
However, the rejection of claims 16-21 under 35 USC 112(a), and 3-5 and 16-21 under 35 USC 112(b), are newly recited above necessitated by claim amendments.
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 1-5, 7-8, 10, 12-13, and 15-21 are rejected under 35 U.S.C. 101 because the claimed invention is directed to an abstract idea of a mental process, a mathematical concept, organizing human activity, or a law of nature or natural phenomenon without significantly more. The rejection of claims 1-5, 7-8, 10, 12-13, and 15-21 under 35 USC 101 contains newly recited portions necessitated by claim amendments filed 6/24/2026.
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, law of nature or natural phenomenon (Step 2A, Prong 1). In the instant application, the claims recite the following limitations that equate to an abstract idea:
Claims 1-2, 13, and 15: “mutate the first sequences represented by the first sequence information, thereby generating virtual mutated sequence information indicating mutated virtual sequences obtained by mutating at least one of constituent amino acids in at least one of the first sequences represented by the first sequence information”; “generate new virtual mutated sequence information based on the new first trained model”; and “generate a target antigen-binding molecule represented by a virtual sequence based on the predicted values for the characterization of antigen-binding molecules with sequences represented by the new virtual mutated sequence information” provides an evaluation (generating information about a virtual mutated sequence or a target antigen-binding molecule) that may be performed in the human mind and is therefore considered a mental process, which is an abstract idea.
“estimate the predicted values for the characterization of the antigen-binding molecules with sequences represented by the inputted virtual mutated sequence information” and “estimate the predicted values for the characterization of antigen-binding molecules with sequences represented by the new virtual mutated sequence information” provides an evaluation (estimating values about antigen-binding molecules) that may be performed in the human mind and is therefore considered a mental process, which is an abstract idea.
Claim 3: “the characteristic of the sequences is a characteristic including positions of constituent elements in the sequences and anteroposterior relationship of the constituent elements” provides for organizing information (structuring data) that may be performed in the human mind and is therefore considered a mental process, which is an abstract idea.
Claim 4: “the virtual sequence information is generated by changing at least one of constituent elements in preset sites including one or more of the constituent elements in a sequence” provides for organizing information (structuring data) that may be performed in the human mind and is therefore considered a mental process, which is an abstract idea.
Claim 10: “according to the predicted values estimated, output on the basis of virtual mutated sequence information and the predicted values” provides for organizing information (outputting data) that may be performed in the human mind and is therefore considered a mental process, which is an abstract idea.
Claims 17, 19, and 21: “the sequence of panning includes a plurality of rounds of panning, in the first round of panning, a collection of a plurality of molecules is subjected to panning, and in the second or later round of panning, a collection of antigen-non-binding molecules in the previous round is not subjected to panning” provides an evaluation (filtering out antigen-non-binding molecules) that may be performed in the human mind and is therefore considered a mental process, which is an abstract idea.
These recitations are similar to the concepts of collecting information, analyzing it, and displaying certain results of the collection and analysis in Electric Power Group, LLC, v. Alstom (830 F.3d 1350, 119 USPQ2d 1739 (Fed. Cir. 2016)), organizing and manipulating information through mathematical correlations in Digitech Image Techs., LLC v Electronics for Imaging, Inc. (758 F.3d 1344, 111 U.S.P.Q.2d 1717 (Fed. Cir. 2014)) and comparing information regarding a sample or test to a control or target data in Univ. of Utah Research Found. v. Ambry Genetics Corp. (774 F.3d 755, 113 U.S.P.Q.2d 1241 (Fed. Cir. 2014)) and Association for Molecular Pathology v. USPTO (689 F.3d 1303, 103 U.S.P.Q.2d 1681 (Fed. Cir. 2012)) that the courts have identified as concepts that can be practically performed in the human mind or are mathematical relationships. Therefore, these limitations fall under the “Mental process” and “Mathematical concepts” groupings of abstract ideas. Additionally, while claims 1-2 recite performing some aspects of the analysis on “An information processing system comprising at least one processor and at least one memory including computer program code”, there are no additional limitations that indicate that this requires anything other than carrying out the recited mental processes or mathematical concepts in a generic computer environment. Merely reciting that a mental process is being performed in a generic computer environment does not preclude the steps from being performed practically in the human mind or with pen and paper as claimed. If a claim limitation, under its broadest reasonable interpretation, covers performance of the limitation in the mind but for the recitation of generic computer components, then it falls within the “Mental processes” grouping of abstract ideas. As such, claims 1-5, 7-8, 10, 12-13, and 15-21 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). The judicial exceptions listed above are not integrated into a practical application because the claims do not recite an additional element or elements that reflects an improvement to technology. Specifically, the claims recite the following additional elements:
Claims 1-2, 13, and 15: “acquire first sequence information representing first sequences including some or all of sequences of a plurality of antigen-binding molecules from a learning dataset storage” and “acquire second sequence information representing second sequences including some or all of the sequences of a plurality of antigen-binding molecules and results of characterization of the antigen-binding molecules represented by the second sequences from the learning dataset storage, wherein the characterization of the antigen-binding molecules include evaluation of affinity, evaluation of pharmacological activity, evaluation of physical properties, evaluation of kinetics, and evaluation of safety of the antigen-binding molecules” provides insignificant extra-solution activities (acquiring data from a dataset storage is a pre-solution activity involving data gathering steps) that do not serve to integrate the judicial exceptions into a practical application.
“input the virtual mutated sequence information into the second trained model, execute arithmetic processing of the second trained model” and “input the new virtual mutated sequence information into the new second trained model” are generally linking the abstract idea identified in the independent claims to the technological environment of machine learning models.
“perform machine learning using the first sequence information to generate a first trained model that has learned characteristics of the first sequences”; “use the first trained model to mutate the first sequences”; “perform machine learning using the second sequence information and the results of characterization to generate a second trained model that has learned predicted values for the characterization of the antigen-binding molecules”; “perform further machine learning for the first trained model and the second trained model based on sequence information of an actual antigen-binding molecule having actually measured characteristics and characteristic information indicating the actually measured characteristics to generate a new first trained model and a new second trained model”; and “generate new virtual mutated sequence information based on the new first trained model” are generally linking the abstract idea identified in the independent claims to the technological environment of machine learning models.
Claims 16-21: “the antigen-binding molecules are acquired from a sequence of panning in which a subsequent round of panning is carried out with a target antigen for the antigen-binding molecules that have appeared in an operation of panning in the previous round” (claims 16, 18, and 20) and “the sequence of panning includes a plurality of rounds of panning, in the first round of panning, a collection of a plurality of molecules is subjected to panning” (claims 17, 19, and 21) provides insignificant extra-solution activities (acquiring data from a dataset storage is a pre-solution activity involving data gathering steps) that do not serve to integrate the judicial exceptions into a practical application.
The steps for acquiring data from a dataset storage do not serve to integrate the recited judicial exceptions into a practical application because they are pre-solution activities involving data gathering and manipulation steps (see MPEP 2106.04(d)(2)). Additionally, the deep learning model or probability model of claims 7-8 and machine learning model 12 (the models) is used to generally apply the abstract idea (i.e., generating information about a virtual mutated sequence or a target antigen-binding molecule) without placing any limitation on how the models operates to "generate a target antigen-binding molecule represented". The claim omits any details as to how the models solve a technical problem and instead recites only the idea of a solution or outcome. See MPEP 2106.05(f). Therefore, the limitation represents no more than mere instructions to implement the abstract idea, which is equivalent to adding the words “apply it” to the recited judicial exception. In addition, the claim confines the use of the recited judicial exception recited in claim 1 to the technological environment of a model by generally linking the use of the judicial exception to the recited models. Therefore, this general model recitation does not integrate the judicial exception into a practical application. See MPEP 2106.05(h). Therefore, it can also be viewed as nothing more than an attempt to generally link the use of the judicial exception to a particular field of use or a technological environment. Furthermore, the limitations regarding implementing program instructions do not indicate that they require anything other than mere instructions to implement the abstract idea in a generic way or in a generic computing environment. As such, this limitation equates 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. Therefore, claims 1-5, 7-8, 10, 12-13, and 15-21 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 additional elements that are insignificant extra-solution activities that do not serve to integrate the recited judicial exceptions into a practical application, or equate to mere instructions to apply the recited exception in a generic way or in a generic computing environment.
As discussed above, there are no additional elements to indicate that the claimed “information processing system comprising at least one processor and at least one memory including computer program code” requires 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. MPEP 2106.05(f) discloses that mere instructions to apply the judicial exception cannot provide an inventive concept to the claims. Additionally, the limitations for: changing, inputting, and outputting data; and measuring protein characteristics are insignificant extra-solution activities that do not serve to integrate the recited judicial exceptions into a practical application. Furthermore, no inventive concept is claimed by the limitations as they are demonstrated to be well-understood, routine, and conventional.
Specifically for the characterization of antigen-binding molecules in claims 1-2, 13, and 15-21 as demonstrated by Gao et al.: Page 1 Abstract "Modern immunoassay methods and techniques are important tools in labs of basic biology, biomedicine, clinical medicine, and even in home tests, such as pregnant test, many of which utilize antibody—antigen binding mechanism as their foundational principle in common. Meanwhile, compared with polyclonal anitbody, monoclonal antibody shows obvious advantages in their application of modern immunoassays. Furthermore, the progress of other technologies have also promoted the development of modern immunoassays robustly, and widely made it extend into more research and industry fields. In this review, we will first look back to the discovery of antibody-antigen binding mechanism, antibody structure, and the development of monoclonal antibody technology. Then, a brief description of different classical immunoassays will be introduced, such as enzyme-linked immunosorbent assay, flow cytometry, Immunoprecipitation, and lateral flow immunoassays" (Gao et al. "A brief review of monoclonal antibody technology and its representative applications in immunoassays." Journal of immunoassay and immunochemistry 39.4 (2018): 351-364);
and for the acquisition of sequence information via panning in Claims 16-21 as demonstrated by Abbott et al.: Under "Approaches to fine epitope mapping and their limitations", page 3 Peptide-based approaches: "In these methods, overlapping peptides are synthesized that cover the whole sequence of the antigen and are then immobilized onto a solid surface as an array and the binding to the antibody of interest is determined in an ELISA format.42-45 It is ideally suited to situations where the epitope is a linear peptide sequence, although it is also possible to constrain peptides via one or more disulphide bonds to mimic discontinuous and conformation-dependent epitopes.46 It is simple and quick to perform. A variation on the use of synthetic peptide libraries is the use of random phage libraries that can be linear or constrained by disulphide bonds.47-51 In this approach, very large phage libraries can be ‘panned’ with the antibodies and the recombinant peptide genes from binding phage particles can be sequenced to determine the sequences and therefore putative epitopes or ‘mimotopes’ as they are sometimes called" (Abbott et al. "Current approaches to fine mapping of antigen–antibody interactions." Immunology 142.4 (2014): 526-535).
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 1-5, 7-8, 10, 12-13, and 15-21 are not patent eligible.
Response to Arguments under 35 USC § 101
Applicant’s arguments filed 6/24/2026 are fully considered but they are not persuasive.
Applicant asserts "that amended claim 1 is not directed to an Abstract idea because amended claim 1 as whole integrates the so-called abstract idea, if any, into a practical application, which is to generate a target antigen-binding molecule represented by a virtual sequence" (Remarks 6/24/2026 page 17). Examiner concedes that the amended independent claims (1-2, 13, and 15) do not recite explicit special mathematical formulas, however, the identified judicial exceptions noted above are not integrated into a practical application because the improvement has to be rooted in an additional element and cannot come from the judicial exception itself (i.e. cannot argue for "a better algorithm" etc.). Additionally, Examiner notes that usage of models are simply generally linking the abstract idea identified in claim 1 to the technological environment of machine learning models.
Applicant further asserts that "using the first trained model to mutate the first sequences represented by the first sequence information, thereby generating virtual mutated sequence information indicating mutated virtual sequences obtained by mutating at least one of constituent amino acids in at least one of the first sequences represented by the first sequence information" amounts to significantly more than the abstract idea (Remarks 6/24/2026 pages 17-18). The Examiner notes that MPEP 2106(I) states that if the claims are directed to a judicial exception, the second part of the Mayo test is to determine whether the claim recites additional elements that amount to significantly more than the judicial exception. Id. citing Mayo, 566 U.S. at 72-73, 101 USPQ2d at 1966). In the “search for an ‘inventive concept’” (the second part of the Alice/Mayo test), the additional elements identified 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 because acquiring data from a dataset storage are all well-understood, routine, and conventional techniques (as identified above) that are insignificant extra-solution activities that do not serve to integrate the recited judicial exceptions into a practical application. Therefore, combining insignificant extra-solution activities with any of the identified judicial exceptions would not result in patent eligible subject matter because integrating well-understood, routine, and conventional techniques does not yield “significantly more” to a mental process, a mathematical concept, organizing human activity, or a law of nature or natural phenomenon.
Applicant also asserts that the additional element of "characterization of the antigen-binding molecules include evaluation of affinity, evaluation of pharmacological activity, evaluation of physical properties, evaluation of kinetics, and evaluation of safety of the antigen-binding molecules" amounts to significantly more than the abstract idea (Remarks 6/24/2026 page 18). Examiner notes above that these characteristics are well-understood, and have been routinely measured as basic laboratory tests as evidenced by Gao et al. (Journal of immunoassay and immunochemistry 39.4 (2018): 351-364): page 1 Abstract "Modern immunoassay methods and techniques are important tools in labs of basic biology, biomedicine, clinical medicine, and even in home tests, such as pregnant test, many of which utilize antibody—antigen binding mechanism as their foundational principle in common. Meanwhile, compared with polyclonal antibody, monoclonal antibody shows obvious advantages in their application of modern immunoassays. Furthermore, the progress of other technologies have also promoted the development of modern immunoassays robustly, and widely made it extend into more research and industry fields. In this review, we will first look back to the discovery of antibody-antigen binding mechanism, antibody structure, and the development of monoclonal antibody technology. Then, a brief description of different classical immunoassays will be introduced, such as enzyme-linked immunosorbent assay, flow cytometry, Immunoprecipitation, and lateral flow immunoassays"
Therefore, the rejection of independent claims 1-2, 13, and 15 under 35 USC 101 is maintained. All other claims depend from these independent claims; therefore, their rejection is likewise maintained.
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.
Claims 1-5, 7-8, 10, 12-13, and 15-21 rejected under 35 U.S.C. 103 as being unpatentable over Bremel et al. (US-20170039314) in view of Shim et al. (US-20170362306) and Karimi et al. (Karimi et al., DeepAffinity: interpretable deep learning of compound–protein affinity through unified recurrent and convolutional neural networks, Bioinformatics, Volume 35, Issue 18, September 2019, Pages 3329–3338, doi.org/10.1093/bioinformatics/btz111). The rejection of claims 1-2, 13, and 15 under 35 USC 103 contains newly recited portions necessitated by claim amendments, and the rest are previously recited.
Regarding claims 1-2, 13, and 15, Bremel teaches acquiring sequence information representing first sequences including some or all of sequences of a plurality of antigen-binding molecules and perform machine learning using the first sequence information to generate a first trained model that has learned characteristics of the first sequence (Para.0016 "In some embodiments, the processes further comprise constructing a neural network via the computer, wherein the neural network is used to predict the binding affinity to one or more MHC [major histocompatibility complex] binding region. In some embodiments, the neural network provides a quantitative structure activity relationship. In some embodiments, the first three principal components represent more than 80% of physical properties of an amino acid." and Para.0018 " the neural network is trained to predict binding to more than one MHC binding region").
Bremel also teaches acquire second sequence information representing second sequences including some or all of the sequences of a plurality of antigen-binding molecules and results of characterization of the antigen-binding molecules represented by the second sequences from the learning dataset storage, wherein the characterization of the antigen-binding molecules include evaluation of affinity, evaluation of pharmacological activity, evaluation of physical properties, evaluation of kinetics, and evaluation of safety of the antigen-binding molecules (evaluation of affinity: Bremel, para.0016 "the neural network is used to predict the binding affinity to one or more MHC [major histocompatibility complex] binding region"; evaluation of pharmacological activity and safety: Bremel, para.0034 "In some embodiments, the processes further comprise identifying a combination of amino acid subsets and MHC binding partners which predispose a subject to a disease outcome. In some embodiments, the processes further comprise screening a population to identify individuals with a HLA haplotype which predisposes individuals with the HLA haplotype to a disease outcome. In some embodiments, the processes further comprising applying the information to design a clinical trial in which patients represent multiple HLA alleles with different binding affinity to said amino acid subset. In some embodiments, the processes further comprise excluding the subjects from a clinical trial"; evaluation of physical properties: Bremel, para.0016 "the neural network provides a quantitative structure activity relationship"; and evaluation of kinetics: Bremel, para.0113 "The predictions suggest a strongly negative correlation with cathepsin cleavage to amino acid position").
Bremel also teaches inputting the virtual mutated sequence information into the second trained model, execute arithmetic processing of the second trained model, and thereby estimate the predicted values for the characterization of the antigen-binding molecules with sequences represented by the inputted virtual mutated sequence information (Para.0012 "The present invention is directed to a method for identification in silico of peptides and sets of peptides internal to or on the surface of microorganisms and cells which have a high probability of being effective in stimulating humoral and cell mediated immune responses. The method combines multiple predictive tools to provide a composite of both topology and multiple sets of binding or affinity characteristics of specific peptides within an entire proteome. This allows us to predict and characterize specific peptides which are B-cell epitope sequences and MHC binding regions in their topological distribution and spatial relationship to each other").
Bremel also teaches: perform further machine learning for the first trained model and the second trained model based on sequence information of an actual antigen-binding molecule having actually measured characteristics and characteristic information indicating the actually measured characteristics to generate a new first trained model and a new second trained model; generating new virtual mutated sequence information based on the new first trained model; and inputting the new virtual mutated sequence information into the new second trained model to estimate the predicted values for the characterization of antigen-binding molecules with sequences represented by the new virtual mutated sequence information. These limitations are akin to iterative training of models from measured data, which is similar in nature to a cross validation, where training data from measurements is used for training then testing the models. An iteration of producing predicted molecules, gathering data to validate those predictions, then using this data to further improve the models is conventional scientific experimentation and therefore would be obvious to one of ordinary skill in the art (para.0225 "In developing NN predictive tools, a common feature is a process of cross validation of the results by use of “training sets” in the “learning” process. In practice, the prediction equations are computed using a subset of the training set and then tested against the remainder of the set to assess the reliability of the method. Binding affinities of peptides of known amino acid sequence have been determined experimentally and are publicly available at http://mhcbindingpredictions.immuneepitope.org/dataset.html. During training, the experimentally determined natural logarithm of the affinity of the particular peptide was used as the output layer. Most of the available training sets consist of about 450 peptides, whose binding affinity to various MHC molecules have been determined in the laboratory").
Bremel also teaches generate a target antigen-binding molecule represented by a virtual sequence based on the predicted values for the characterization of antigen-binding molecules with sequences represented by the new virtual mutated sequence information (para.0023 "In some embodiments, the amino acid sequence comprises the amino acid sequences of a class of proteins selected from the group consisting of membrane associated proteins in the proteome of a target source").
Bremel does not explicitly teach use the first trained model to mutate the first sequences represented by the first sequence information, thereby generating virtual mutated sequence information indicating mutated virtual sequences obtained by mutating at least one of constituent amino acids in at least one of the first sequences represented by the first sequence information; nor perform machine learning using the second sequence information and the results of characterization to generate a second trained model that has learned predicted values for the characterization of the antigen-binding molecules.
However, Shim teaches using the first trained model to mutate the first sequences represented by the first sequence information, thereby generating virtual mutated sequence information indicating mutated virtual sequences obtained by mutating at least one of constituent amino acids in at least one of the first sequences represented by the first sequence information (Para.0117 "First, for heavy chain and light chain CDR1s and CDR2s with only somatic hypermutation without recombination, CDR sequences were designed to have similar sequences and germline CDR sequences of the human-derived mature antibodies by introducing virtual mutations into the human germline CDR sequence, through simulation using a computer. 1,500 simulated sequences for each CDR were designed").
However, Karimi teaches perform machine learning using the second sequence information and the results of characterization to generate a second trained model that has learned predicted values for the characterization of the antigen-binding molecules (Page 3 col 2 section 2.3 paragraph 1 "We used a recurrent neural network (RNN) model, seq2seq (Sutskever et al., 2014), that has seen much success in natural language processing and was recently applied to embedding compound SMILES strings into fingerprints (Xu et al., 2017). A Seq2seq model is an auto-encoder that consists of two recurrent units known as the encoder and the decoder, respectively (see the corresponding box in Fig. 1). The encoder maps an input sequence (SMILES/SPS in our case) to a fixed-dimension vector known as the thought vector. Then the decoder maps the thought vector to the target sequence (again, SMILES/SPS here)").
Therefore, it would have been obvious to one of ordinary skill in the art as of the effective filing date of the claimed invention to modify the training data used for the neural network of Bremel with mutated sequence data as taught by Shim in order to design antibody libraries with excellent physical properties against many antigens (Abstract "The antibody library prepared according to the present invention contains antibodies having excellent physical properties against a plurality of antigens, thereby having functional diversity and containing a plurality of unique sequences, and thus can be favorably used as an antibody library"). One skilled in the art would have a reasonable expectation of success because both methods use binding affinity data and essentially are focused on designing antibody libraries.
Therefore, it would have been obvious to one of ordinary skill in the art as of the effective filing date of the claimed invention to modify the neural network model of Bremel and Shim with the antigen-binding molecule characterization data as taught by Karimi because it is the most appropriate model for learning compound/protein representations (page 4 col 2 paragraphs 1-2 "Third, whereas bi-directional GRUs lowered perplexity by about 2 ~3.5 folds and the default attention mechanism did much more for compounds or proteins, they together achieved the best performances (perplexity being 1.0002 for compound SMILES and 1.001 for protein SPS). Therefore, the last seq2seq variant, bidirectional GRUs with attention mechanism, is regarded the most appropriate one for learning compound/protein representations and adopted thereinafter."). One skilled in the art would have a reasonable expectation of success because both approaches have the goal of modeling protein binding affinity.
Regarding claim 3, Bremel in view of Shim and Karimi teach the methods of Claim 1 on which this claim depends. Shim also teaches the characteristic of the sequences is a characteristic including positions of the constituent elements in the sequences and anteroposterior relationship of the constituent element (Para.0117 "First, for heavy chain and light chain CDR1s and CDR2s with only somatic hypermutation without recombination, CDR [complementarity-determining region] sequences were designed to have similar sequences and germline CDR sequences of the human-derived mature antibodies by introducing virtual mutations into the human germline CDR sequence, through simulation using a computer. 1,500 simulated sequences for each CDR were designed").
Regarding claim 4, Bremel in view of Shim and Karimi teach the methods of Claim 1 on which this claim depends. Shim also teaches the virtual sequence information is generated by changing at least one of the constituent elements in preset sites including one or more of the constituent elements in a sequence (same excerpt as claim 3, para.0117, mutations within the CDR sequence constitutes preset sites).
Regarding claim 5, Bremel in view of Shim and Karimi teach the methods of Claim 4 on which this claim depends. Shim also teaches the at least one preset sites is included in a sequence of a heavy chain variable region, a light chain variable region, or a constant region of an antibody (same excerpt as claim 3, para.0117, CDRs are within both the variable heavy- and light- chain domains).
Regarding claims 7 and 8, Bremel in view of Shim and Karimi teach the method of Claim 1 on which these claims depend. Karimi also teaches a sequence model that performs the machine learning with use of a deep learning model in the form of a gate recurrent unit (Page 3 col 2 section 2.3 paragraph 1 "We choose gated recurrent unit (GRU) (Cho et al., 2014) as our default seq2seq model and treat the thought vectors as the representations learned from the SMILES/SPS inputs.").
Regarding claim 10, Bremel in view of Shim and Karimi teach the methods of Claim 1 on which this claim depends. Shim also teaches an output configured to, according to the predicted values estimated by the estimator, output on the basis of virtual sequence information and the predicted values (Para.0117 "First, for heavy chain and light chain CDR1s and CDR2s with only somatic hypermutation without recombination, CDR sequences were designed to have similar sequences and germline CDR sequences of the human-derived mature antibodies by introducing virtual mutations into the human germline CDR sequence, through simulation using a computer. 1,500 simulated sequences for each CDR were designed").
Regarding claim 12, Bremel in view of Shim and Karimi teach the methods of Claim 1 on which this claim depends. Bremel also teaches the sequence learner performs the machine learning on the basis of the sequence information represented by character strings, numeric vectors, or physical property values of constituent elements constituting sequences (Para.0203 provides Table 1 displaying the training resources utilized for NN training: "General immunology resources", "Amino acid physical properties", "Web NN & Training sets", etc.).
Regarding claim 15, Bremel in view of Shim and Karimi teach the methods of Claim 1 on which this claim depends. Bremel also teaches the antigen-binding molecule or protein is represented by a virtual sequence, and a predicted value for characterization has been estimated for the virtual sequence (Para.0012 "The present invention is directed to a method for identification in silico of peptides and sets of peptides internal to or on the surface of microorganisms and cells which have a high probability of being effective in stimulating humoral and cell mediated immune responses. The method combines multiple predictive tools to provide a composite of both topology and multiple sets of binding or affinity characteristics of specific peptides within an entire proteome. This allows us to predict and characterize specific peptides which are B-cell epitope sequences and MHC binding regions in their topological distribution and spatial relationship to each other.").
Regarding claims 16-21, Bremel in view of Shim and Karimi teach the methods of Claims 1, 13, and 15 on which these claims depend. Shim also teaches the antigen-binding molecules are acquired from a sequence of panning in which a subsequent round of panning is carried out with a target antigen for the antigen-binding molecules that have appeared in an operation of panning in the previous round, and the sequence of panning includes a plurality of rounds of panning, in the first round of panning, a collection of a plurality of molecules is subjected to panning, and in the second or later round of panning, a collection of antigen-non-binding molecules in the previous round is not subjected to panning (Para.0082 "The term “panning” refers to a process of selectively amplifying only those clones that bind to a specific molecule from a library of proteins, such as antibodies, displayed on a phage surface. The procedure is that a phage library is added to a target molecule immobilized on the surface to induce binding, unbound phage clones are removed by washing, only bound phage clones are eluted and again infect the E. Coli host, and target-bound phage clones are amplified using helper phages. In most cases, this process is repeated three to four times or more to maximize the percentage of bound clones" and Figure 2).
Response to Arguments under 35 USC § 103
Applicant’s arguments filed 6/24/2026 are fully considered but they are not persuasive.
Applicant asserts "that none of the cited art discloses, teaches, or suggests that the characterization of the antigen-binding molecules include evaluation of affinity, evaluation of pharmacological activity, evaluation of physical properties, evaluation of kinetics, and evaluation of safety of the antigen-binding molecules, as recited in amended claim 1" (Remarks 6/24/2026 page 18). Examiner notes above that Bremel in fact teaches these limitations (evaluation of affinity: Bremel, para.0016 "the neural network is used to predict the binding affinity to one or more MHC [major histocompatibility complex] binding region"; evaluation of pharmacological activity and safety: Bremel, para.0034 "In some embodiments, the processes further comprise identifying a combination of amino acid subsets and MHC binding partners which predispose a subject to a disease outcome. In some embodiments, the processes further comprise screening a population to identify individuals with a HLA haplotype which predisposes individuals with the HLA haplotype to a disease outcome. In some embodiments, the processes further comprising applying the information to design a clinical trial in which patients represent multiple HLA alleles with different binding affinity to said amino acid subset. In some embodiments, the processes further comprise excluding the subjects from a clinical trial"; evaluation of physical properties: Bremel, para.0016 "the neural network provides a quantitative structure activity relationship"; and evaluation of kinetics: Bremel, para.0113 "The predictions suggest a strongly negative correlation with cathepsin cleavage to amino acid position").
Therefore, the rejection of independent claims 1-2, 13, and 15 under 35 USC 103 is maintained. All other claims depend from these independent claims; therefore, their rejection is likewise maintained.
Conclusion
No claims are allowed.
Applicant's amendment necessitated the new ground(s) of rejection presented in this Office action. Accordingly, THIS ACTION IS MADE FINAL. See MPEP § 706.07(a). Applicant is reminded of the extension of time policy as set forth in 37 CFR 1.136(a).
A shortened statutory period for reply to this final action is set to expire THREE MONTHS from the mailing date of this action. In the event a first reply is filed within TWO MONTHS of the mailing date of this final action and the advisory action is not mailed until after the end of the TH REE-MONTH shortened statutory period, then the shortened statutory period will expire on the date the advisory action is mailed, and any nonprovisional extension fee (37 CFR 1.17(a)) pursuant to 37 CFR 1.136(a) will be calculated from the mailing date of the advisory action. In no event, however, will the statutory period for reply expire later than SIX MONTHS from the mailing date of this finaI action.
Inquiries
Any inquiry concerning this communication or earlier communications from the examiner should be directed to Robert A. Player whose telephone number is (571)272-6350. The examiner can normally be reached Mon-Fri, 8am-5pm.
Examiner interviews are available via telephone, in-person, and video conferencing using a USPTO supplied web-based collaboration tool. To schedule an interview, applicant is encouraged to use the USPTO Automated Interview Request (AIR) at http://www.uspto.gov/interviewpractice.
If attempts to reach the examiner by telephone are unsuccessful, the examiner’s supervisor, Larry D. Riggs can be reached on 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.
/R.A.P./Examiner, Art Unit 1686
/KAITLYN L MINCHELLA/Primary Examiner, Art Unit 1685
Prosecution Insights