DETAILED ACTION
Applicant's response, filed 9/30/2025, has been fully considered. The following rejections and/or
objections are either reiterated or newly applied. They constitute the complete set presently being
applied to the instant application.
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 .
Priority
The instant application claims benefit of priority to U.S. Provisional Application No. 62/831,663 filed on 04/09/2019. The claim to the benefit of priority is acknowledged. As such, the effective filing date of claims 1, 3-5, 7-19 and 88 is 04/09/2019.
Claim Status
Claims 1, 3-5, 7-19 and 88 are currently pending.
Claims 1, 3-5, 7-19 and 88 are rejected.
Nucleotide and/or Amino Acid Sequence Disclosures
Response to Amendment
In view of applicant’s amendments to the specification previous objections due to sequence disclosures are withdrawn.
Drawings
Response to Amendment
In view of applicant’s amendments to the drawings, previous objections to the drawings over sequence identifiers is withdrawn.
Specification
Response to Amendment
In view of applicant’s amendments to the specification previous objections regarding the use of trade names have been reviewed, updated, and provided below. Additionally, while Advanced Analytical Technologies does not seem to be a registered trade name Analytical Technologies is and does produce a sequencer, if this is the sequencer to which the specification is referring than such names need to be amended to their correct names and properly marked as trade names.
The use of the terms Rockland Immunochemicals, which is a trade name or mark used in commerce, has been noted in this application. The term should be accompanied by the generic terminology; furthermore the term should be capitalized wherever it appears or, where appropriate, include a proper symbol indicating use in commerce such as ™, SM , or ® following the term.
Although the use of trade names and marks used in commerce (i.e., trademarks, service marks, certification marks, and collective marks) are permissible in patent applications, the proprietary nature of the marks should be respected and every effort made to prevent their use in any manner which might adversely affect their validity as commercial marks.
Claim Rejections - 35 USC § 112
The following is a quotation of the first paragraph of 35 U.S.C. 112(a):
(a) IN GENERAL.—The specification shall contain a written description of the invention, and of the manner and process of making and using it, in such full, clear, concise, and exact terms as to enable any person skilled in the art to which it pertains, or with which it is most nearly connected, to make and use the same, and shall set forth the best mode contemplated by the inventor or joint inventor of carrying out the invention.
The following is a quotation of the first paragraph of pre-AIA 35 U.S.C. 112:
The specification shall contain a written description of the invention, and of the manner and process of making and using it, in such full, clear, concise, and exact terms as to enable any person skilled in the art to which it pertains, or with which it is most nearly connected, to make and use the same, and shall set forth the best mode contemplated by the inventor of carrying out his invention.
Claim 88 is rejected under 35 U.S.C. 112(a) or 35 U.S.C. 112 (pre-AIA ), first paragraph, as failing to comply with the written description requirement. The claim(s) contains subject matter which was not described in the specification in such a way as to reasonably convey to one skilled in the relevant art that the inventor or a joint inventor, or for applications subject to pre-AIA 35 U.S.C. 112, the inventor(s), at the time the application was filed, had possession of the claimed invention. Claim 88 is directed to a system for the method of claim 1. However, claim 1 and claim 88 contain the step of generating an antibody, and while claim 1 is a method claim, claim 88 is a system claim. Claim 88 is thus further examined for system components described within the specification and drawings that provides sufficient detail for performing the recited limitations. Within the specification and the drawings no mention of such components was provided nor described. Additionally, the specification does not provide an indication that the applicant had possession of a system that included the components for generating an antibody nor does it provide any indication that a processor alone is capable of generating an antibody. Here, the step is merely the transmission of instructions for generating a antibody that is within the functions that can be carried out by a processor.
Claim Rejections - 35 USC § 101
Response to Amendment
In view of applicant’s amendments to the claims, previous rejections under 35 U.S.C. 101 have been reviewed, updated, and provided below. For claims 1, 3-5, and 7-19, rejections under 35 U.S.C. 101 have been withdrawn.
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.
Claim 88 is rejected under 35 U.S.C. 101 because the claimed invention is directed to abstract ideas without significantly more. The claims recite a system for using sequence data to train a classification engine to determine affinity binding scores for a sequence to an antigen and selecting said sequence based upon an affinity binding score threshold and subsequently generating the antibody. The judicial exception is not integrated into a practical application because while claim 88 attempt to integrate the exception into a practical application, said application is either generically recited computer elements that do not add a meaningful limitation to the abstract idea or it is insignificant extra solution activity and merely implementing the abstract idea on a computer. The claims do not include additional elements that are sufficient to amount to significantly more than the judicial exception because the computer elements only store and retrieve information in memory as well as perform basic calculations that are known to be well-understood, routine and conventional computer functions as recognized by the decisions listed in MPEP § 2106.05(d).
Framework with which to Analyze Subject Matter Eligibility:
Step 1: Are the claims directed to a category of stator subject matter (a process, machine,
manufacture, or composition of matter)? [See MPEP § 2106.03]
Claims are directed to statutory subject matter, specifically a system (Claim 88)
Step 2A Prong One: Do the claims recite a judicially recognized exception, i.e., an abstract idea, a law of nature, or a natural phenomenon? [See MPEP § 2106.04(a)]
The claims herein recite abstract ideas, specifically mental processes and mathematical concepts.
With respect to the Step 2A Prong One evaluation, the instant claims are found herein
to recite abstract ideas that fall into the grouping of mental processes and mathematical
concepts.
Claim 88: Providing the second training data set to a classification engine to generate weights, determining a first affinity binding score, and selecting the proposed amino acid sequence based upon the binding score are processes of calculating, and comparing, information/data that can all be done with pen and paper or in the human mind and are therefore, abstract ideas, specifically mental processes. The first machine learning model classifying antigen binding specificity with a predetermined confidence or probability verbal articulation of a mathematical processes and is therefore an abstract idea, specifically a mathematical concept.
Step 2A Prong Two: If the claims recite a judicial exception under prong one, then is the judicial
exception integrated into a practical application? [See MPEP § 2106.04(d) and MPEP § 2106.05(a)-(c) & (e)-(h)]
Because the claims do recite judicial exceptions, direction under Step 2A Prong Two
provides that the claims must be examined further to determine whether they integrate the abstract ideas into a practical application.
The following claims recite the following additional elements in the form of non-abstract
elements:
Claim 88: A system, processors, memory, executable instructions, and machine learning models, are generic and nonspecific elements of a computer that do not improve the functioning of any computer or technology described herein [See MPEP § 2106.04(d)(1) and MPEP § 2106.05(d)]. Receiving an input amino acid sequence, receiving a first training data set, receiving a second training data set, generating a combinatorial mutagenesis library, and generating libraries through high throughput mutagenesis, deep mutational scanning, and combinatorial mutagenesis (merely describes the process by which the data was generated and aren’t actually being performed in the system) are insignificant extra solution activities, specifically mere data gathering (See Performing clinical tests on individuals to obtain input for an equation, In re Grams, 888 F.2d 835, 839-40; 12 USPQ2d 1824, 1827-28 (Fed. Cir. 1989) and Determining the level of a biomarker in blood, Mayo, 566 U.S. at 79, 101 USPQ2d at 1968. See also PerkinElmer, Inc. v. Intema Ltd., 496 Fed. App'x 65, 73, 105 USPQ2d 1960, 1966 (Fed. Cir. 2012) (assessing or measuring data derived from an ultrasound scan, to be used in a diagnosis)) [See MPEP § 2106.05(g)]. Generating an antibody is merely the transmission of instructions for generating an antibody that is within the functions that can be carried out by a processor and is therefore mere instructions to apply an exception [See MPEP § 2106.05(f)].
Step 2B: If the claims do not integrate the judicial exception, do the claims provide an inventive
concept? [See MPEP § 2106.05]
Because the additional claim elements do not integrate the abstract idea into a practical
application, the claims are further examined under Step 2B, which evaluates whether
the additional elements, individually and in combination, amount to significantly more
than the judicial exception itself by providing an inventive concept.
The claims do not recite additional elements that are sufficient to amount to
significantly more than the judicial exception because the claims recite additional
elements that are generic, conventional or nonspecific. These additional elements
include:
The additional elements of a system, processors, memory, executable instructions, and machine learning models are generic and nonspecific elements of a computer that are well-understood, routine and conventional within the art and therefore do not improve the functioning of any computer or technology described therein (Receiving or transmitting data over a network, e.g., using the Internet to gather data, Symantec, 838 F.3d at 1321, 120 USPQ2d at 1362 (utilizing an intermediary computer to forward information), Performing repetitive calculations, Flook, 437 U.S. at 594, 198 USPQ2d at 199 (recomputing or readjusting alarm limit values), and Storing and retrieving information in memory, Versata Dev. Group, Inc. v. SAP Am., Inc., 793 F.3d 1306, 1334, 115 USPQ2d 1681, 1701 (Fed. Cir. 2015)) [See § MPEP 2106.05(d)(II)]. Therefore, taken both individually and as a whole, the additional elements do not amount to significantly more than the judicial exception by providing an inventive concept.
The additional elements of providing an input amino acid sequence (Conventional Methods: Specification [117]-[120]), receiving an input amino acid sequence, receiving a first training data set (Conventional Methods: Specification [117]-[120]), receiving a second training data set (Conventional Methods: Specification [117]-[120]), generating a combinatorial mutagenesis library (Conventional Methods: Specification [103]-[109]), and generating libraries through high throughput mutagenesis (Conventional Methods: Specification [103]-[109]), deep mutational scanning (Conventional Methods: Specification [103]-[109]), and combinatorial mutagenesis (Conventional Methods: Specification [103]-[109]) are insignificant extra solutional activities, specifically mere data gathering, that are recognized as well understood, routine and conventional by the courts (See Performing clinical tests on individuals to obtain input for an equation, In re Grams, 888 F.2d 835, 839-40; 12 USPQ2d 1824, 1827-28 (Fed. Cir. 1989) and Determining the level of a biomarker in blood, Mayo, 566 U.S. at 79, 101 USPQ2d at 1968. See also PerkinElmer, Inc. v. Intema Ltd., 496 Fed. App'x 65, 73, 105 USPQ2d 1960, 1966 (Fed. Cir. 2012) (assessing or measuring data derived from an ultrasound scan, to be used in a diagnosis)) [See MPEP § 2106.05(g)]. Therefore, taken both individually and as whole, the additional elements do not amount to significantly more than the judicial exception by providing an inventive concept.
The additional element of generating an antibody (Tiller et al. 2019) is merely the transmission of instructions for generating an antibody that is within the functions that can be carried out by a processor and is therefore mere instructions to apply an exception [See MPEP § 2106.05(f)]. Therefore, taken both individually and as whole, the additional elements do not amount to significantly more than the judicial exception by providing an inventive concept.
Therefore, claim 88, when the limitations are considered individually and as a whole, is rejected under 35 USC § 101 as being directed to non-statutory subject matter.
Response to Arguments
Applicant's arguments filed 9/30/2025 have been fully considered but they are only partially persuasive. Applicant asserts that due to the amendments the rejection is traversed. For claims 1, 3-5, and 7-19 the rejection is withdrawn in view of the amendments, however claim 88 is not traversed as there is no support within the specification for how the system is to work without the necessary equipment for the synthesis of an antibody, i.e., the step which is used to ground the judicial exception within non-abstract additional elements is not described within the description such as to provide any indication that a processor alone is capable of generating an antibody. Therefore, this step is relegated to merely providing the transmission of instructions for generating a antibody that is within the functions that can be carried out by a processor and is therefore, a well-understood, routine and conventional computer element that does not improve the functioning of the computer.
Claim Rejections - 35 USC § 103
Response to Amendment
In view of applicant’s amendments to the claims, previous rejections under 35 U.S.C. 103 have been reviewed, updated, and provided below.
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.
Claims 1, 4-5, 7-11, 14, and 88 are rejected under 35 U.S.C. 103 as being unpatentable over Kuroda et al. (Protein Engineering, Design & Selection (2012) 507-521; previously cited), and Jain et al. (Bioinformatics (2017) 3758–3766; previously cited), in view of Tiller et al. (Annual Review of Biomedical Engineering (2015) 191-216; previously cited) and Liberis et al. (Bioinformatics (2018) 2944–2950; newly cited).
Claim 1 is directed to a method of using an amino acid sequence to generate training data for a machine learning algorithm to learn weights such that the model can determine a binding score for the proposed sequence and then select the sequence whose binding score satisfies a particular threshold.
Claim 88 is directed to a system for using an amino acid sequence to generate training data for a machine learning algorithm to learn weights such that the model can determine a binding score for the proposed sequence and then select the sequence whose binding score satisfies a particular threshold.
Kuroda et al. teaches on page 507, column 2, paragraph 2 “The sequences, structures and functions of antibodies have been extensively studied due to their growing importance as therapeutics and research tools”, in paragraph 3 “One central goal is to accurately predict the structures of antibodies from their sequences…”, and on page 508, column 2, paragraph 3 “Adopting these empirically determined sequence–structure relationships has made it possible to model the L1, L2, L3, H1 and H2 loops with high accuracy”, reading on providing an input amino acid sequence that represents an antigen binding portion of an antigen binding molecule. Kuroda et al. teaches on page 510, column 1, paragraph 2 “When modeling antibody structures from sequence…”, reading on wherein the antigen binding molecule comprises an antibody, or an antigen binding fragment thereof. Kuroda et al. teaches on page 507, column 2, paragraph 3 “…using experimentally determined or predicted structures of the antibody–antigen complex, computational methods can be used to predict mutations that may improve binding affinity, specificity or other properties such as solubility”, reading on selecting the proposed amino acid sequence for expression based on the first affinity binding score satisfying a threshold. Finally, Kuroda et al. teaches in Figure 1 the “Flow of computation antibody design”, specifically showcasing the “Experimental Validations” following “Computational Designs” that would necessarily require generating the antibody, thereby reading on generating an antibody comprising an amino acid sequence of a classified antigen-binder variant.
Kuroda et al. does not teach creating two datasets one for training a model’s parameters and the other for calculating binding affinity.
Jain et al. teaches on page 3760, column 2, paragraph 6 “The dataset was partitioned based on these clusters to generate training and cross-validation sets for coefficient determination using logistic regression. While differing in detail, a similar approach of partitioning sequences into training and test sets on the basis of similarity was also used in another study focused on developing machine learning models to predict aggregation risk of antibodies”, reads on generating a first training data set comprising a first plurality of variant sequences, each of the first plurality of variant sequences comprising a single site mutation in the input amino acid sequence of the antigen binding molecule, and generating a second training data set comprising a second plurality of sequences, each of the second plurality of sequences comprising a plurality of variants at positions based on enrichment scores of the first training data set comprising the first plurality of variant sequences. The use of the phrases training and logistic regression, and additionally on the same page first column, paragraph 5 “For each position i along the antibody sequence, a separate random forest regressor was trained to predict the SASA”, would inherently necessitate the use of data to learn weights used in logistic regression and random forests, reading on providing the second training data set to a classification engine comprising a first machine learning model to generate a plurality of weights and biases for the first machine learning model. Jain et al. teaches on page 3762, column 2, paragraph 3 “A few examples of such
approaches are the detection of aggregation-prone regions, modeling transfer solvation energies, estimating binding free energies…”, reading on determining, by the classification engine based on the plurality of weights and bias for the first machine learning model, a first affinity binding score for a proposed amino acid sequence to an antigen.
Tiller et al. teaches on page 195, paragraph 2 “Given the large size and complexity of antibodies, most design efforts have focused on redesigning or optimizing existing antibodies rather than on de novo design of new antibodies. These design methods vary greatly, and range from knowledge-based methods based on previous mutagenesis results to advanced computational methods based on first principles”, on page 203, paragraph 3 “There are two key approaches to controlling the level and type of effector functions for conventional antibodies, namely (a) engineering the sequences of Fc and hinge regions, and (b)modulating the amount and type of Fc glycosylation.The first approach builds on the identification of residues within the Fc domain that normally interact with Fcγ receptors and C1q (74–77). One method for identifying such residues is to make chimeras of different IgG isotypes that naturally display dissimilar abilities to elicit ADCC and CDC (78–80). Another method is to use systematic alanine mutagenesis of the Fc domain to identify residues involved in binding to different Fcγ receptors or C1q”, reading on high throughput mutagenesis in a mammalian cell expressing an input amino acid sequence of an antibody variant that cannot bind a target antigen to generate a population of mammalian cells expressing antibody variants that bind and cannot bind the target antigen, and deep mutational scanning of the population of mammalian cells to determine enrichment scores of positions in the input amino acid sequence capable of accepting a wide- range of mutations, and combinatorial mutagenesis and based on enrichment scores of the first training data set that indicate an amenability of positions to accepting mutations, and (ii) deep sequencing population of mammalian cells comprising binding and non-binding rationally designed antibody variants.
Liberis et al. teaches in the abstract “In this work, we present a sequence-based probabilistic machine learning algorithm for paratope prediction, named Parapred. Parapred uses a deep-learning architecture to leverage features from both local residue neighbourhoods and across the entire sequence. The method significantly improves on the current state-of-the-art methodology, and only requires a stretch of amino acid sequence corresponding to a hypervariable region as an input, without any information about the antigen”, and on page 2950, column 1, paragraph 5 “this work is the first application of modern deep learning (CNN- and RNN-based neural networks) to the paratope prediction problem”, reading on wherein the first machine learning model is selected from the group consisting of a long- short term memory recurrent neural network (LSTM-RNN). a convolutional neural network (CNN). a standard artificial neural network (ANN). a support vector machine (SVM). a random forest ensemble (RF). and a logistic regression model (LR). and wherein the first machine learning model classifies antigen binding specificity with a predetermined confidence or probability level: generating in silico, by a candidate identifying system, a combinatorial mutagenesis library comprising a plurality of variant amino acid sequences of the input amino acid sequence.
It would have been obvious at the time of invention to a person skilled in the art to modify the teachings of Kuroda et al. for the use of sequence information to predict or improve antibody binding, with the teachings of Jain et al. for the use of machine learning techniques as Kuroda et al. points out the use of machine learning for predicting antibody interactions, and Jain et al. does the same, Jain et al. merely specifically points to a specific type of machine learning expanding on a proposed method from Kuroda et al., to outline in detail the implementation of such a method. Additionally it would have been obvious to modify the teachings of the previous two with the teachings of Tiller et al. for the use of mutagenesis in the data generation and the teachings of Liberis et al. for the use of the specified machine learning teachniques as the former stands as a review of current methods and best practices within the field of antibody design and the latter explicitly states within the abstract “The method significantly improves on the current state-of-the-art methodology”. One would have had a reasonable expectation of success given that Kuroda et al. is a review of in silico methods for antibody design as is Tiller et al. and Jain et al. is a more recent publication utilizing more recent in silico methods for predicting antibody binding and Liberis et al. is showcasing applied models for in silico antibody design and binding prediction. Therefore, it would have been obvious to a person skilled in the art to modify the teachings of each and to be successful.
Claim 4 is directed to the method of claim 1 but further specifies determining a second score for the amino acid sequence using a second machine learning model and then selecting the sequence whose binding scores satisfy both thresholds.
Jain et al. teaches on page 3760, column 2, paragraph 5 “For each position i along the antibody sequence, a separate random forest regressor was trained to predict the SASA”, reading on determining, by the classification engine, a second affinity binding score for the proposed amino acid sequence using a second machine learning model of the classification engine and selecting the proposed amino acid sequence for expression based on the first affinity binding score and the second affinity binding score satisfying the threshold, as a random forest is an ensemble method inherently performing multiple predictions/classifications.
Tiller et al. teaches on page 195, paragraph 2 “Given the large size and complexity of antibodies, most design efforts have focused on redesigning or optimizing existing antibodies rather than on de novo design of new antibodies. These design methods vary greatly, and range from knowledge-based methods based on previous mutagenesis results to advanced computational methods based on first principles”, on page 203, paragraph 3 “There are two key approaches to controlling the level and type of effector functions for conventional antibodies, namely (a) engineering the sequences of Fc and hinge regions, and (b)modulating the amount and type of Fc glycosylation.The first approach builds on the identification of residues within the Fc domain that normally interact with Fcγ receptors and C1q (74–77). One method for identifying such residues is to make chimeras of different IgG isotypes that naturally display dissimilar abilities to elicit ADCC and CDC (78–80). Another method is to use systematic alanine mutagenesis of the Fc domain to identify residues involved in binding to different Fcγ receptors or C1q”, reading on wherein the second machine learning model is selected from the group consisting of a long-short term memory recurrent neural network (LSTM-RNN), or a convolutional neural network (CNN), a standard artificial neural network (ANN), a support vector machine (SVM), a random forest ensemble (RF), and a logistic regression model (LR).
Claim 5 is directed to the method of claim 1 but further specifies determining an affinity binding score for a plurality of sequences along with parameters for each of a plurality of sequences and selecting variants from the plurality based on the score and parameters of each of the sequences.
Jain et al. teaches on page 3760, column 2, paragraph 6 “The dataset was partitioned based on these clusters to generate training and cross-validation sets for coefficient determination using logistic regression. While differing in detail, a similar approach of partitioning sequences into training and test sets on the basis of similarity was also used in another study focused on developing machine learning models to predict aggregation risk of antibodies”, reading on determining, by the classification engine, an affinity binding score for each of a plurality of proposed amino acid sequences; determining, by a candidate selection engine, one or more parameters for each of the plurality of proposed amino acid sequences; and selecting, by the candidate selection engine, candidate variants from the plurality of proposed amino acid sequences based on the affinity binding score and the one or more parameters for each of the plurality of proposed amino acid sequences.
Claim 7 is directed to the method of claim 6 and thus claim 1, but further specifies that the confidence or probability level be above 0.5.
Jain et al. teaches on page 3760, column 2, paragraph 6 “The dataset was partitioned based on these clusters to generate training and cross-validation sets for coefficient determination using logistic regression”, reading on wherein the predetermined confidence or probability level is above 0.5.
Claim 8 is directed to the method of claim 5 and thus claim 1, but further specifies that the variant selected must satisfy a threshold for at least one or more additional parameters.
Kuroda et al. teaches on page 517, column 2, paragraph 3 “These methods were used to modify the structure of the designed proteins with the goal of the minimizing immunogenicity while retaining the thermal stability, solubility and high affinity toward cognate antibody obtained in the earlier study”, reading on wherein the candidate selection engine selects variants based on the proposed amino acid sequence satisfying a threshold for at least of one of the one or more additional parameters.
Claim 9 is directed to the method of claim 5 and thus claim 1, but further specifies that the selection of variants satisfy the threshold for all of the additional parameters.
Kuroda et al. teaches on page 517, column 2, paragraph 3 “These methods were used to modify the structure of the designed proteins with the goal of the minimizing immunogenicity while retaining the thermal stability, solubility and high affinity toward cognate antibody obtained in the earlier study”, reading on wherein the candidate selection engine selects variants based on the proposed amino acid sequence satisfying a threshold for each of the one or more additional parameters.
Claim 10 is directed to the method of claim 9 and thus claim 1, but further specifies that each of the additional parameters includes a value threshold.
Kuroda et al. teaches on page 517, column 2, paragraph 3 “These methods were used to modify the structure of the designed proteins with the goal of the minimizing immunogenicity while retaining the thermal stability, solubility and high affinity toward cognate antibody obtained in the earlier study”, for which solubility, immunogenicity, thermal stability and affinity would inherently be values with necessitated thresholds, reading on wherein the threshold for each of the one or more additional parameters includes a value threshold.
Claim 11 is directed to the method of claim 9 and thus claim 1, but further specifies that each of the additional parameters includes a variable or relative threshold.
Kuroda et al. teaches on page 517, column 2, paragraph 3 “These methods were used to modify the structure of the designed proteins with the goal of the minimizing immunogenicity while retaining the thermal stability, solubility and high affinity toward cognate antibody obtained in the earlier study”, all of these values can also be relative to each other, reading on wherein the threshold for each of the one or more additional parameters includes a variable or relative threshold.
Claim 14 is directed to the method of claim 5 and thus claim 1, but further specifies that the one or parameters comprise those from the specified group.
Kuroda et al. teaches on page 517, column 2, paragraph 3 “These methods were used to modify the structure of the designed proteins with the goal of the minimizing immunogenicity while retaining the thermal stability, solubility and high affinity toward cognate antibody obtained in the earlier study”, reading on wherein the one or more parameters comprise viscosity values, solubility values, stability values, pharmacokinetic values, and/or immunogenicity values.
Claim 3 is rejected under 35 U.S.C. 103 as being unpatentable over Jain et al. (Bioinformatics (2017) 3758–3766; previously cited), Kuroda et al. (Protein Engineering, Design & Selection (2012) 507-521; previously cited), Tiller et al. (Annual Review of Biomedical Engineering (2015) 191-216; previously cited) and Liberis et al. (Bioinformatics (2018) 2944–2950; newly cited) as applied to claims 1, 4-5, 7-11 , 14, and 88 above, and further in view of Sadelain et al. (Cancer Discovery (2013) 388-398; previously cited).
Claim 3 is directed to the method of claim 1 but further specifies that the antigen binding molecule be a chimeric antigen receptor.
Jain et al., Kuroda et al., Tiller et al., and Liberis et al. teach the method of claims 5 and 1 as previously described.
Jain et al., Kuroda et al., Tiller et al., and Liberis et al. do not teach that the antigen binding molecule be a chimeric antigen receptor.
Sadelain et al. teaches in the abstract “This review focuses on the design of CARs, including the requirements for optimal antigen recognition and different modalities to provide costimulatory support to targeted T cells, which include the use of second- and third generation CARs, costimulatory ligands, chimeric costimulatory receptors, and cytokines”, reading on wherein the antigen binding molecule comprises a chimeric antigen receptor.
It would have been obvious at the time of invention to a person skilled in the art to modify the teachings of Jain et al. and Kuroda et al. for the method of claims 1 and 5, with the teachings of Sadelain et al. for the design of chimeric antigen receptors as the latter is a review focusing on the governing parameters and requirements for the creation of effective chimeric antigen receptors which would be the same principles used in the selection/optimization of chimeric antibody design/classification. One would have had a reasonable expectation of success given that the latter is a review of the current understanding and technical details governing chimeric antigen receptors. Therefore, it would have been obvious to a person skilled in the art to modify the teachings of each and to be successful.
Claim 15 is rejected under 35 U.S.C. 103 as being unpatentable over Jain et al. (Bioinformatics (2017) 3758–3766; previously cited), Kuroda et al. (Protein Engineering, Design & Selection (2012) 507-521; previously cited), Tiller et al. (Annual Review of Biomedical Engineering (2015) 191-216; previously cited) and Liberis et al. (Bioinformatics (2018) 2944–2950; newly cited) as applied to claims 1, 4-5, 7-11 , 14, and 88, and further in view of Chen et al. (Immune Research (2010) 1-7; previously cited).
Claim 15 is directed to the method of claim 5 and thus claim 1, but further specifies that the one or more parameters comprise a Levenshtein distance value.
Jain et al., Kuroda et al., Tiller et al., and Liberis et al. teach the method of claims 5 and 1 as previously described.
Jain et al., Kuroda et al., Tiller et al., and Liberis et al. do not teach that the one or more parameters comprise a Levenshtein distance value.
Chen et al. teaches on page 3, column 1, paragraph 1 “Measuring the similarity of DNA sequences can be considered as a problem of comparing two strings. The Edit distance also known as the Levenshtein distance, can be considered as a classic measure of the similarity of two strings”, reading on wherein the one or more parameters comprise a Levenshtein distance value.
It would have been obvious at the time of invention to a person skilled in the art to modify the teachings of Jain et al. and Kuroda et al. for the method of claims 1 and 5, with the teachings of Chen et al. for the use of the Levenshtein distance value for parameters in antibody classification as the latter points out it is merely a measure of the similarity between two strings, and for a “system and method of classification of amino acid sequence of binding proteins”, a measure of similarity between two sequences would be an obvious means for basing classifications. One would have had a reasonable expectation of success given that the latter is paper on the classification of antibodies and is using a Levenshtein distance value in their classification. Therefore, it would have been obvious to a person skilled in the art to modify the teachings of each and to be successful.
Claims 16-18 are rejected under 35 U.S.C. 103 as being unpatentable over Jain et al. (Bioinformatics (2017) 3758–3766; previously cited), Kuroda et al. (Protein Engineering, Design & Selection (2012) 507-521; previously cited), Tiller et al. (Annual Review of Biomedical Engineering (2015) 191-216; previously cited) and Liberis et al. (Bioinformatics (2018) 2944–2950; newly cited) as applied to claims 1, 4-5, 7-11 , 14, and 88above, and further in view of Yadav et al. (Journal of Biological Chemistry (2015) 29732-29741; previously cited).
Claim 16 is directed to the method of claim 5 and thus claim 1 but further specifies that one of the parameters be charge value.
Jain et al., Kuroda et al., Tiller et al., and Liberis et al. teach the method of claims 5 and 1 as previously described.
Jain et al., Kuroda et al., Tiller et al., and Liberis et al. do not teach that one of the parameters be charge value.
Yadav et al. teaches on page 29732, column 2, paragraph 2 “Factors that affect antibody pharmacokinetics (PK)2 include antibody-specific properties (charge, hydrophobicity, target affinity, FcRn affinity, Fc-gamma receptor interactions, and glycosylation), target properties (expression level, turnover rate, and soluble versus membrane-associated), drug administration (dose and route), anti-therapeutic antibody formation, off-target/nonspecific binding, and disease state (healthy volunteers versus patients)”, reading on wherein the one or more parameters comprise charge value.
It would have been obvious at the time of invention to a person skilled in the art to modify the teachings of Jain et al. and Kuroda et al. for the method of claims 1 and 5, with the teachings of Yadav et al. for including fragment charge as the latter shows that it is an important factor in pharmacokinetic behavior, which they specify includes affinity on page 29732, column 2, paragraph 2. One would have had a reasonable expectation of success given both papers are focused on binding affinity and the secondary parameters surrounding it. Therefore, it would have been obvious to a person skilled in the art to modify the teachings of each and to be successful.
Claim 17 is directed to the method of claim 16 and thus claim 1 but further specifies that the charge value is a variable charge fragment.
Jain et al., Kuroda et al., Tiller et al., and Liberis et al. teach the method of claims 5 and 1 as previously described.
Jain et al., Kuroda et al., Tiller et al., and Liberis et al. do not teach that the charge value is a variable charge fragment.
Yadav et al. teaches in the abstract “The effect of modifying Fv charge on subcutaneous bioavailability was also examined, and in general bioavailability was inversely related to the direction of the Fv charge change. Thus, modifying Fv charge appears to impact antibody PKs”, reading on wherein the charge value is a variable fragment (Fv) charge value.
Claim 18 is directed to the method of claim 17 and thus claim 1, but further specifies that the charge value be between about 0 and about 6.2.
Jain et al., Kuroda et al., Tiller et al., and Liberis et al. teach the method of claims 5 and 1 as previously described.
Jain et al., Kuroda et al., Tiller et al., and Liberis et al. do not teach that the charge value is between about 0 and about 6.2.
Yadav et al. teaches on page 29737, columns 1-2, paragraph 2 “It is possible that charge changes within the acceptable range predicted by the model of 0–6.2 do not mediate a change in PK or that the mutations chosen had no significant impact on the overall charge distribution within the Fv, as indicated by the homology model for this variant compared with its parental antibody”, reading on wherein the Fv charge value is between about 0 and about 6.2.
Claim 19 is rejected under 35 U.S.C. 103 as being unpatentable over Jain et al. (Bioinformatics (2017) 3758–3766; previously cited), Kuroda et al. (Protein Engineering, Design & Selection (2012) 507-521; previously cited), Tiller et al. (Annual Review of Biomedical Engineering (2015) 191-216; previously cited), Liberis et al. (Bioinformatics (2018) 2944–2950; newly cited) and Yadav et al. (Journal of Biological Chemistry (2015) 29732-29741; previously cited) as applied to claims 1, 4-5, 7-11 , 14, 16-18 and 88 above, and further in view of Raybould et al. (bioRxiv (2018) 1-7).
Claim 19 is directed to the method of claim 16 and thus claim 1, but further specifies that the charge value is a variable fragment charge symmetry parameter value.
Jain et al., Kuroda et al., Tiller et al., and Liberis et al., and Yadav et al. teach the method of claims 1, 5, and 16 as previously described.
Jain et al., Kuroda et al., Tiller et al., and Liberis et al., and Yadav et al. do not teach that the charge value is a variable fragment charge symmetry parameter value.
Raybould et al. teaches on page 4, column 1, paragraph 3 “When mAbs have oppositely charged VH and VL chains, they typically have higher in vitro viscosity values. This aggregate-inducing electrostatic attraction is captured at the sequence level by the Fv Charge Symmetry Parameter (FvCSP) metric - the mAb tends to be more viscous if the product of net VH and VL charges is negative”, reading on wherein the charge value is a variable fragment charge symmetry parameter (FvCSP) value.
It would have been obvious at the time of invention to a person skilled in the art to modify the teachings of Jain et al., Kuroda et al., and Yadav et al. for the method of claims 1, 5, and 16, with the teachings of Raybould et al. for the use of FvCSP as a metric in antibody profiling, as the latter reference is showing how the metric can be used in antibody profiling on pages 3-4 under the subheading Charge, specifically on page 4, column 1, paragraph 3. One would have had a reasonable expectation of success given that all the references are centered on antibody design, classification/profiling, and therefore showcasing the important features and rules that govern antibodies. Therefore, it would have been obvious to a person skilled in the art to modify the teachings of each and to be successful.
Claims 12 and 13 are rejected under 35 U.S.C. 103 as being unpatentable over Kuroda et al. (Protein Engineering, Design & Selection (2012) 507-521; previously cited), Jain et al. (Bioinformatics (2017) 3758–3766; previously cited), Tiller et al. (Annual Review of Biomedical Engineering (2015) 191-216; previously cited), and Liberis et al. (Bioinformatics (2018) 2944–2950; newly cited) as applied to claim 1, 4-5, 7-11 , 14, and 88 above, and further in view of Bergstra et al. (Advances in Neural Information Processing Systems (2011) 1-9).
Claim 12 is directed to the method of claim 9 and thus claim 1, but further specifies that the threshold for one or more of the additional parameters is a value in the top 5-10%.
Jain et al., Kuroda et al., Tiller et al., and Liberis et al., and Yadav et al. teach the method of claims 1, 5, and 16 as previously described.
Jain et al., Kuroda et al., Tiller et al., and Liberis et al., and Yadav et al. do not teach that the threshold for one or more of the additional parameters is a value in the top 5-10%.
Claim 13 is directed to the method of claim 9 and thus claim 1, but further specifies that the threshold for one or more of the additional parameters is a value based on a number of standard deviations above the average.
Jain et al., Kuroda et al., Tiller et al., and Liberis et al., and Yadav et al. teach the method of claims 1, 5, and 16 as previously described.
Jain et al., Kuroda et al., Tiller et al., and Liberis et al., and Yadav et al. do not teach that the threshold for one or more of the additional parameters is a value based on a number of standard deviations above the average.
Bergstra et al. teaches on page 4, subsection 4, paragraph 2, the method of TPE, “Whereas the GP-based approach favored quite an aggressive y∗ (typically less than the best observed loss), the TPE algorithm depends on a y∗ that is larger than the best observed f(x) so that some points can be used to form (x). The TPE algorithm chooses y∗ to be some quantile γ of the observed y values, so that p(y < y∗) = γ, but no specific model for p(y) is necessary. By maintaining sorted lists of observed variables in H, the runtime of each iteration of the TPE algorithm can scale linearly in |H| and linearly in the number of variables (dimensions) being optimized”, meaning that the parameter estimation is based on a specified quantile of the observed values and that this quantile is specified by the user. While this itself does not directly read on the limitation it does provide an obviousness read on the limitation in that the quantile whether it be 5%, 10%, or some standard deviation above a specified value is obvious to try to optimize through the use of said TPE algorithm particularly as within the abstract Bergstra et al. teaches “The sequential algorithms are applied to the most difficult DBNlearning problems from [1] and find significantly better results than the best previously reported”.
It would have been obvious at the time of first filing to have modified the teachings of Jain et al., Kuroda et al., Tiller et al., and Liberis et al., and Yadav et al. for the method of claims 1, 5, and 9 with the teachings of Bergstra et al. for the use of user specified thresholds for parameter tuning and optimization as the latter teaches in the abstract “The sequential algorithms are applied to the most difficult DBNlearning problems from [1] and find significantly better results than the best previously reported”. One would have had a reasonable expectation of success given that the use of such algorithms would merely present a simple substitution of algorithms and would not interfere with the overall process but merely enable superior parameter optimization. Therefore, it would have been obvious at the time of first filing to have modified the teachings of each and to be successful.
Response to Arguments
Applicant's arguments filed 9/30/2025 have been fully considered but they are not persuasive. Applicant asserts on page 22 of the Remarks filed 9/30/2025 that the cited reference Kuroda et al. does not teach or suggest “how to identify a variant of an antibody that can selectively bind to a target antibody using the claimed method”. However, with the claims this is not specifically taught, while variants are examined these are merely variant sequences which can amount to nothing more than the various sequences under examination within Figure 1 of Kuroda et al. and are responsible for the various structures Kuroda et al. is examining for improved antibody design.
Furthermore, applicant asserts at the bottom of page 22 of the Remarks filed 9/30/2025 that the teachings of Juan et al. do not cover “a method of generating antigen binding molecule variants with improved binding affinity based on an input antibody or antigen binding fragment that cannot bind a target antigen”. This however is moot, as the reference is used for the support of the method of using machine learning in antibody design, the latter of which and the assertion of applicant, is taught within Kuroda et al. page 507, column 2, paragraphs 2-3 as previously cited.
Conclusion
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 THREE-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 final action.
Any inquiry concerning this communication or earlier communications from the examiner should be directed to KEENAN NEIL ANDERSON-FEARS whose telephone number is (571)272-0108. The examiner can normally be reached M-Th, alternate F, 8-5.
Examiner interviews are available via telephone, in-person, and video conferencing using a USPTO supplied web-based collaboration tool. To schedule an interview, applicant is encouraged to use the USPTO Automated Interview Request (AIR) at http://www.uspto.gov/interviewpractice.
If attempts to reach the examiner by telephone are unsuccessful, the examiner’s supervisor, Karlheinz Skowronek can be reached at 571-272-9047. The fax phone number for the organization where this application or proceeding is assigned is 571-273-8300.
Information regarding the status of published or unpublished applications may be obtained from Patent Center. Unpublished application information in Patent Center is available to registered users. To file and manage patent submissions in Patent Center, visit: https://patentcenter.uspto.gov. Visit https://www.uspto.gov/patents/apply/patent-center for more information about Patent Center and https://www.uspto.gov/patents/docx for information about filing in DOCX format. For additional questions, contact the Electronic Business Center (EBC) at 866-217-9197 (toll-free). If you would like assistance from a USPTO Customer Service Representative, call 800-786-9199 (IN USA OR CANADA) or 571-272-1000.
/K.N.A./Examiner, Art Unit 1687
/OLIVIA M. WISE/Supervisory Patent Examiner, Art Unit 1685