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 .
Remarks
In response to communications sent March 25, 2026, claim(s) 1-10 and 21-30 is/are pending in this application; of these claim(s) 1, 21, 25, and 28 is/are in independent form. Claim(s) 11-20 is/are cancelled.
Response to Amendment
The amendments to the specification and claims are acknowledged and have been entered into the record.
Drawings
The drawings were received on August 20, 2025. These drawings are accepted by the Examiner.
Terminal Disclaimer
The terminal disclaimer filed on March 23, 2026 disclaiming the terminal portion of any patent granted on this application which would extend beyond the expiration date of United States Patent No. 12,424,300 has been reviewed and is accepted. The terminal disclaimer has been recorded.
Response to Arguments
Applicant’s arguments, see page 1 lines 4-12, filed March 23, 2026, with respect to the drawings have been fully considered and are persuasive. The rejection of the drawings of December 29, 2025 has been withdrawn.
Applicant’s arguments, see page 1 lines 13-18, filed March 23, 2026, with respect to the claims 22 and 23 have been fully considered and are persuasive. The objection to claims 22 and 23 of December 29, 2025 has been withdrawn.
Applicant’s arguments, see page 1 line 19 to page 2 line 9, filed March 23, 2026, with respect to the claims 1-10 and 21-30 have been fully considered and are persuasive. The rejection of claims 1-10 and 21-30 of December 29, 2025 has been withdrawn.
Applicant’s arguments, see page 2 line 10 to page 4 line 23, filed March 23, 2026, with respect to the rejection(s) of claim(s) 1-4 and 21-30 under 35 U.S.C. § 103 have been fully considered and are persuasive. Therefore, the rejection has been withdrawn. However, upon further consideration, a new ground(s) of rejection is made in view of 35 U.S.C. § 102 and Zhang (Zhang, Zuobai, et al. "Pre-training protein encoder via siamese sequence-structure diffusion trajectory prediction." Advances in Neural Information Processing Systems 36 (2023): 43496-43524). Note that the Examiner disagrees with the arguments on page 4 line 24 to 5 line 4 because considerations of well-understood, routine, and conventional elements and integration into a practical application have different standards under 35 U.S.C. § 101 than obviousness under 35 U.S.C. § 103.
Applicant’s arguments, see page 2 line 10 to page 4 line 23, filed March 23, 2026, with respect to claims 5-10 have been fully considered and are persuasive. The rejection of claims 5-10 sent December 29, 2025 has been withdrawn.
Applicant’s arguments, see page 5 line 29-40, filed March 23, 2026, with respect to the claims 1-10 and 21-30 have been fully considered and are persuasive. The rejection of claims 1-10 and 21-30 of December 29, 2025 has been withdrawn due to the Terminal Disclaimer.
Applicant’s arguments, see page 5 line 41 to page 7 line 14, filed March 23, 2026, with respect to the rejection(s) of claim(s) 1-10 and 21-30 under non-statutory double patenting have been fully considered and are persuasive. Therefore, the rejection has been withdrawn. However, upon further consideration, a new ground(s) of rejection is made in view of U.S. Patent No. 12,437,837 in view of Zhang (Zhang, Zuobai, et al. "Pre-training protein encoder via siamese sequence-structure diffusion trajectory prediction." Advances in Neural Information Processing Systems 36 (2023): 43496-43524.).
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.
Claim 27 is 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.
Claim 27 recites the limitation "the ligand of an industrial enzyme" in lines 1-2. There is insufficient antecedent basis for this limitation in the claim.
Claim Rejections - 35 USC § 102
The following is a quotation of the appropriate paragraphs of 35 U.S.C. 102 that form the basis for the rejections under this section made in this Office action:
A person shall be entitled to a patent unless –
(a)(1) the claimed invention was patented, described in a printed publication, or in public use, on sale, or otherwise available to the public before the effective filing date of the claimed invention.
Claim(s) 1-4 and 21-30 is/are rejected under 35 U.S.C. 102(a)(1) as being anticipated by Zhang (Zhang, Zuobai, et al. "Pre-training protein encoder via siamese sequence-structure diffusion trajectory prediction." Advances in Neural Information Processing Systems 36 (2023): 43496-43524.).
As to claim 1, Zhang teaches a method, comprising:
a) receiving, at a processor, a trained neural network (Zhang Figure 1: pretraining):
i) wherein the neural network is configured to accept a representation of one or more specified conditions as input, and to yield as output, a representation of a protein associated with the one or more specified conditions (Zhang Figure 1: the neural-network inputs in are a representation of a protein that is associated with a Diffusion Trajectory Prediction; see Zhang’s Title),
ii) wherein the neural network has at least two output heads (Zhang page 23 lines provides evidence that the output is using Multilayer Perceptron heads), including one output head which generates the output protein’s sequence and a different output head which generates the output protein’s structure (Zhang Figure 1 lower right corner of Pre-Training section indicates that the neural network outputs sequence data and structure data, see Abstract),
iii) wherein the process of training the neural network entailed a loss function computation corresponding to the sequence head output and a different loss function computation corresponding to the structure head output (Zhang page 5 lines 5-6: decomposing the loss function into a loss function corresponding to the sequence S and a loss function corresponding to the structure R),
iv) wherein some non output head weights of the neural network are shared (Zhang’s theta parameter in section 2.2 on page 3 lines 8-13 is a shared parameter used for the process on sequences and structures);
b) using the neural network to generate a representation of a protein or small molecule, given a representation of the specified condition(s) as input (Zhang’s theta parameter in section 2.2 on page 3 lines 8-13 is a shared parameter used for the process on sequences and structures),
wherein, based on the input, the sequence head generates a sequence representation of the protein and the structure head generates a structure representation of the protein, and wherein a sequence and structure representation of the generated protein is returned as output (Zhang Figure 1 lower right corner of Pre-Training section indicates that the neural network outputs sequence data and structure data, see Abstract; Zhang page 23 lines provides evidence that the output is using Multilayer Perceptron heads).
As to claim 2, Zhang teaches the method of claim 1, wherein the output protein is synthesized (Zhang abstract: generating conformers).
As to claim 3, Zhang teaches the method of claim 1, wherein the training process uses backpropagation (Zhang page 2 section title for 2.2: Diffusion model, which entails backpropogation by definition); and wherein during training, the backpropagation and weight updates proceed backwards independently from each of the heads through the ancestral nodes of the respective head (Zhang page 23 lines 43-46: the Multilayer Perceptron head is used for downstream tasks, whereas the diffusion model is upstream; see Algorithm 1 on page 18, wherein the last line of the algorithm updates theta).
As to claim 4, Zhang teaches the method of claim 3, wherein the sequence head’s final output is a probability distribution over amino acids and auxiliary tokens; and wherein the structure head’s final output is a probability distribution over possible structure parameters associated with each residue (Zhang page 3 section 2.2: decoding using a reverse generative process based on amino acids and 3D coordinates).
As to claim 21, Zhang teaches a method, comprising:
a) receiving, at a processor, a representation of a protein (Zhang Figure 1: pretraining):
i) wherein the representation of the protein was obtained using a neural network trained and configured to return a representation of a protein, given one or more specified conditions on the protein (Zhang Figure 1: the neural-network inputs in are a representation of a protein that is associated with a Diffusion Trajectory Prediction; see Zhang’s Title),
ii) wherein the neural network has at least two heads (Zhang page 23 lines provides evidence that the output is using Multilayer Perceptron heads), including one output head which generates the output protein’s sequence and a different output head which generates the output protein’s structure (Zhang Figure 1 lower right corner of Pre-Training section indicates that the neural network outputs sequence data and structure data, see Abstract),
iii) wherein during training of the neural network, there was a loss function computation corresponding to the sequence head output and a different loss function computation corresponding to the structure head output (Zhang page 5 lines 5-6: decomposing the loss function into a loss function corresponding to the sequence S and a loss function corresponding to the structure R),
iv) wherein some non output head weights of the neural network were shared (Zhang’s theta parameter in section 2.2 on page 3 lines 8-13 is a shared parameter used for the process on sequences and structures),
v) wherein, based on the input, the sequence head generated the sequence representation of the protein and the structure head generated the structure representation of the protein (Zhang Figure 1 lower right corner of Pre-Training section indicates that the neural network outputs sequence data and structure data, see Abstract; Zhang page 23 lines provides evidence that the output is using Multilayer Perceptron heads);
b) synthesizing the protein (Zhang abstract: generating conformers).
As to claim 22, Zhang teaches the method of claim 21, wherein biological properties of the protein are assessed in silico or in vitro (Zhang Page 9: section 5.2: experimental results).
As to claim 23, Zhang teaches the method of claim 21, wherein biological properties of the protein are assessed in vivo (Zhang page 3 section 2.2: decoding using a reverse generative process based on amino acids and 3D coordinates)..
As to claim 24, Zhang teaches the method of claim 21, wherein the protein is used as a diagnostic or therapeutic agent in a human, animal, or plant.(Zhang abstract: generating conformers; medical use is at once envisaged).
As to claim 25, Zhang teaches a method, comprising:
a) receiving a protein:
i) wherein the protein was synthesized from a representation obtained using a neural network trained and configured to return a representation of a protein, given one or more specified conditions on the protein (Zhang Figure 1: the neural-network inputs in are a representation of a protein that is associated with a Diffusion Trajectory Prediction; see Zhang’s Title),
ii) wherein the neural network has at least two heads, including one output head which generates the output protein’s sequence and a different output head which generates the output protein’s structure (Zhang page 23 lines provides evidence that the output is using Multilayer Perceptron heads; Zhang Figure 1 lower right corner of Pre-Training section indicates that the neural network outputs sequence data and structure data, see Abstract),
iii) wherein during training of the neural network, there was a loss function computation corresponding to the sequence head output and a different loss function computation corresponding to the structure head output (Zhang page 5 lines 5-6: decomposing the loss function into a loss function corresponding to the sequence S and a loss function corresponding to the structure R),
iv) wherein some non output head weights of the neural network were shared (Zhang’s theta parameter in section 2.2 on page 3 lines 8-13 is a shared parameter used for the process on sequences and structures),
v) wherein, based on the input, the sequence head generated the sequence representation of the protein and the structure head generated the structure representation of the protein (Zhang’s theta parameter in section 2.2 on page 3 lines 8-13 is a shared parameter used for the process on sequences and structures);
b) assessing biological properties of the protein in vitro or in vivo (Zhang Page 9: section 5.2: experimental results).
As to claim 26, Zhang teaches the method of claim 25, wherein the protein is used as a diagnostic or therapeutic agent in a human, animal, or plant (Zhang abstract: generating conformers; medical administration is at once envisaged).
As to claim 27, Zhang teaches the method of claim 25, wherein the protein is the ligand of an industrial enzyme (Zhang abstract: generating conformers; industrial presence is at once envisaged).
As to claim 28, Zhang teaches a method, comprising:
a) receiving a protein (Zhang Figure 1: pretraining):
i) wherein the protein was synthesized from a representation obtained using a neural network trained and configured to return a representation of a protein, given one or more specified conditions on the protein (Zhang Figure 1: the neural-network inputs in are a representation of a protein that is associated with a Diffusion Trajectory Prediction; see Zhang’s Title),
ii) wherein the neural network has at least two heads (Zhang page 23 lines provides evidence that the output is using Multilayer Perceptron heads), including one output head which generates the output protein’s sequence and a different output head which generates the output protein’s structure (Zhang Figure 1 lower right corner of Pre-Training section indicates that the neural network outputs sequence data and structure data, see Abstract),
iii) wherein during training of the neural network, there was a loss function computation corresponding to the sequence head output and a different loss function computation corresponding to the structure head output (Zhang page 5 lines 5-6: decomposing the loss function into a loss function corresponding to the sequence S and a loss function corresponding to the structure R),
iv) wherein some non output head weights of the neural network were shared (Zhang’s theta parameter in section 2.2 on page 3 lines 8-13 is a shared parameter used for the process on sequences and structures),
v) wherein, based on the input, the sequence head generated the sequence representation of the protein and the structure head generated the structure representation of the protein (Zhang Figure 1 lower right corner of Pre-Training section indicates that the neural network outputs sequence data and structure data, see Abstract; Zhang page 23 lines provides evidence that the output is using Multilayer Perceptron heads);
b) using the protein as a diagnostic or therapeutic agent in a human, animal, or plant (Zhang abstract: generating conformers; medical administration is at once envisaged).
As to claim 29, Zhang teaches the method of claim 28, wherein the specified condition includes a representation of an antigen, and wherein the output protein is an associated antibody (Zhang abstract: generating conformers of proteins; medical administration is at once envisaged).
As to claim 30, Zhang teaches the method of claim 28, wherein the specified condition includes a representation of a target receptor, and wherein the output protein is a peptide ligand of that receptor Zhang abstract: generating conformers; industrial presence is at once envisaged).
Double Patenting
The nonstatutory double patenting rejection is based on a judicially created doctrine grounded in public policy (a policy reflected in the statute) so as to prevent the unjustified or improper timewise extension of the “right to exclude” granted by a patent and to prevent possible harassment by multiple assignees. A nonstatutory double patenting rejection is appropriate where the conflicting claims are not identical, but at least one examined application claim is not patentably distinct from the reference claim(s) because the examined application claim is either anticipated by, or would have been obvious over, the reference claim(s). See, e.g., In re Berg, 140 F.3d 1428, 46 USPQ2d 1226 (Fed. Cir. 1998); In re Goodman, 11 F.3d 1046, 29 USPQ2d 2010 (Fed. Cir. 1993); In re Longi, 759 F.2d 887, 225 USPQ 645 (Fed. Cir. 1985); In re Van Ornum, 686 F.2d 937, 214 USPQ 761 (CCPA 1982); In re Vogel, 422 F.2d 438, 164 USPQ 619 (CCPA 1970); In re Thorington, 418 F.2d 528, 163 USPQ 644 (CCPA 1969).
A timely filed terminal disclaimer in compliance with 37 CFR 1.321(c) or 1.321(d) may be used to overcome an actual or provisional rejection based on nonstatutory double patenting provided the reference application or patent either is shown to be commonly owned with the examined application, or claims an invention made as a result of activities undertaken within the scope of a joint research agreement. See MPEP § 717.02 for applications subject to examination under the first inventor to file provisions of the AIA as explained in MPEP § 2159. See MPEP § 2146 et seq. for applications not subject to examination under the first inventor to file provisions of the AIA . A terminal disclaimer must be signed in compliance with 37 CFR 1.321(b).
The filing of a terminal disclaimer by itself is not a complete reply to a nonstatutory double patenting (NSDP) rejection. A complete reply requires that the terminal disclaimer be accompanied by a reply requesting reconsideration of the prior Office action. Even where the NSDP rejection is provisional the reply must be complete. See MPEP § 804, subsection I.B.1. For a reply to a non-final Office action, see 37 CFR 1.111(a). For a reply to final Office action, see 37 CFR 1.113(c). A request for reconsideration while not provided for in 37 CFR 1.113(c) may be filed after final for consideration. See MPEP §§ 706.07(e) and 714.13.
The USPTO Internet website contains terminal disclaimer forms which may be used. Please visit www.uspto.gov/patent/patents-forms. The actual filing date of the application in which the form is filed determines what form (e.g., PTO/SB/25, PTO/SB/26, PTO/AIA /25, or PTO/AIA /26) should be used. A web-based eTerminal Disclaimer may be filled out completely online using web-screens. An eTerminal Disclaimer that meets all requirements is auto-processed and approved immediately upon submission. For more information about eTerminal Disclaimers, refer to www.uspto.gov/patents/apply/applying-online/eterminal-disclaimer.
Claims 1-10 and 21-30 are rejected on the ground of nonstatutory double patenting as being unpatentable over claims 12-15 of U.S. Patent No. 12,437,837 in view of Zhang (Zhang, Zuobai, et al. "Pre-training protein encoder via siamese sequence-structure diffusion trajectory prediction." Advances in Neural Information Processing Systems 36 (2023): 43496-43524.). The claims are obvious variants of each other because the missing element in the reference patent is merely “…wherein some non output head weights of the neural network are shared…” However, this element is taught by Zhang (Zhang’s theta parameter in section 2.2 on page 3 lines 8-13 is a shared parameter used for the process on sequences and structures).
Reference patent and Zhang are in the same field of bioinformatics. Therefore, it would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to modify the teachings of the reference to include the teachings of Zhang because the using a joint distribution surpasses the limitations of unimodal pretraining (Zhang Introduction at page 1 last two lines). There is a reasonable expectation of success because the reference patent already separation of loss functions as part of a larger neural network architecture.
Instant Application 19/304,885
Reference U.S. Patent No. 12,437,837
1. A method, comprising:
a) receiving, at a processor, a trained neural network:
i) wherein the neural network is configured to accept a representation of one or more specified conditions as input, and to yield as output, a representation of a protein associated with the one or more specified conditions,
ii) wherein the neural network has at least two output heads, including one output head which generates the output protein's sequence and a different output head which generates the output protein's structure,
iii) wherein the process of training the neural network entailed a loss function computation corresponding to the sequence head output and a different loss function computation corresponding to the structure head output,
iv) wherein some non output head weights of the neural network are shared; wherein, based on the input, the sequence head generates a sequence representation of the protein and the structure head generates a structure representation of the protein, and
b) using the neural network to generate a representation of a protein or small molecule, given a representation of the specified condition(s) as input.
12. A method, comprising:
a) receiving, at a processor, a pre-trained mixed modality neural network: i) wherein the representation modalities are for representations of features of proteins, ii) wherein the represented features include one or more of sequence, structure, function, interactions, interactors, binding partners, attributes, and properties, iii) wherein the respective modalities of the representations include one or more of: natural language representation modality, protein sequence representation modality, protein structure representation modality, or small molecule drug representation modality,
iv) wherein the neural network is configured to accept as input data, a query consisting of one or more of the modalities, and to yield as output data, a response to the query, wherein the response also consists of one or more of the modalities,
v) wherein the neural network is autoregressive,
vi) wherein the neural network has multiple output heads,
each with its own loss function,
vii) wherein for each respective output head of the neural network, the final output is a probability distribution over a set of possible values at that head; b) using a plurality of mixed modality reason-oriented query response pairs to perform supervised fine tuning of the pre-trained neural network: i) wherein for each reason-oriented query used, the pre-trained neural network's output is scored against a corresponding chain-of-thought response, ii) wherein the representation modalities of each query and each response are for representations of features of proteins, iii) wherein the represented features include one or more of sequence, structure, function, interactions, interactors, binding partners, attributes, and properties, iv) wherein the respective modalities of the representations include one or more of: natural language representation modality, protein sequence representation modality, protein structure representation modality, or small molecule drug representation modality, v) wherein an optimization process is used to iteratively update the weights of the pre-trained neural network, vi) wherein the supervised fine tuning weight updates proceed until termination criteria are met, vii) wherein the output is a reasoning-oriented mixed modality neural network; c) using the reasoning-oriented mixed modality neural network as an output generator method for obtaining a representation of an output protein, in response to an input query specifying conditions on the protein;
d) using the output generator method to generate an output protein by randomly sampling the output probability distribution of the neural network's active head at each iteration of the autoregression;
e) running the random-sampling based generation process a plurality of times with a given input query, wherein each of the plurality of runs uses the same input query, and wherein each of the plurality of runs yields a representation of a candidate protein; f) obtaining the plurality of generated representations of proteins as output.
2. The method of claim 1, wherein the output protein is synthesized.
15. The method of claim 14, wherein the ligand is synthesized.
3. The method of claim 1, wherein the training process uses backpropagation; and wherein during training, the backpropagation and weight updates proceed backwards independently from each of the heads through the ancestral nodes of the respective head.
See the elements of claim 12 in which there are separate loss functions in an autoregressive neural network.
4. The method of claim 3, wherein the sequence head's final output is a probability distribution over amino acids and auxiliary tokens; and wherein the structure head's final output is a probability distribution over possible structure parameters associated with each residue.
See the elements of claim 12 which include sequence representation modalities and protein structure representation modalities; also see claim 12 for the elements of multiple heads each with its own loss functions.
5. The method of claim 4, wherein the sequence and structure generation is via an autoregressive procedure.
See claim 12 regarding autoregressive procedures.
6. The method of claim 5, wherein the specified condition is a target receptor and the specified condition's representation is a representation of the target receptor's sequence and structure; and wherein the output protein is a peptide ligand drug.
13. The method of claim 12, wherein the input query into the output generator specifies a target receptor and requests a peptide ligand of the receptor; and wherein the generated output is a representation of a peptide ligand of the specified target receptor.
7. The method of claim 6, for generating a representation of a peptide ligand drug's sequence and structure given a representation of a target receptor's sequence and structure, wherein the method is also for obtaining and synthesizing an effective peptide ligand drug, the method further comprising:
a) using the trained neural network to obtain a representation of a peptide ligand drug, given a representation of a target receptor:
i) wherein, during autoregression, each residue is determined by randomly sampling the output probability distribution of the sequence head,
ii) wherein, during autoregression, the structure parameters associated with each residue are determined by randomly sampling the output probability distribution of the structure head;
b) repeating the random sampling-based peptide ligand drug representation generation procedure a plurality of times, thereby generating a plurality of representations of candidate peptide ligand drugs;
c) assessing the binding interaction and efficacy of each of the generated candidate peptide ligand drug representations;
d) selecting the most effective candidate peptide ligand drug;
e) synthesizing the peptide ligand drug.
Claim 15, which depends from claim 14.
14. The method of claim 13,
wherein for each candidate ligand in the plurality of generated representations of peptide ligands, an assessment is made of its interaction, efficacy, and properties with the target receptor, and
the most effective ligand is selected.
15. The method of claim 14, wherein the ligand is synthesized.
8. The method of claim 7, wherein the specified conditions are a set of desired properties of the output protein; wherein the possible values of each property are categorical classes, each numerically encoded:
a) wherein the specified conditions are represented by a vector of length equal to the number of properties, as each entry position holds the value of the respective specified property;
b) wherein each specified property is numerically encoded categorically or continuously.
At once envisaged from claim 12, which recites an autoregressive neural network involving features.
9. The method of claim 8, wherein the output protein is a peptide ligand for a given target protein.
13. The method of claim 12, wherein the input query into the output generator specifies a target receptor and requests a peptide ligand of the receptor; and wherein the generated output is a representation of a peptide ligand of the specified target receptor.
10. The method of claim 9, wherein the given target protein is a receptor, and wherein the peptide ligand represented by the output is synthesized.
15. The method of claim 14, wherein the ligand is synthesized.
21. A method, comprising:
a) receiving, at a processor, a representation of a protein:
i) wherein the representation of the protein was obtained using a neural network trained and configured to return a representation of a protein, given one or more specified conditions on the protein,
ii) wherein the neural network has at least two heads, including one output head which generates the output protein's sequence and a different output head which generates the output protein's structure, iii) wherein during training of the neural network, there was a loss function computation corresponding to the sequence head output and a different loss function computation corresponding to the structure head output
iv) wherein some non output head weights of the neural network were shared;
v) wherein, based on the input, the sequence head generated the sequence representation of the protein and the structure head generated the structure representation of the protein;
b) synthesizing the protein.
Claim 15, which depends from claim 12-14.
12. A method, comprising:
a) receiving, at a processor, a pre-trained mixed modality neural network: i) wherein the representation modalities are for representations of features of proteins, ii) wherein the represented features include one or more of sequence, structure, function, interactions, interactors, binding partners, attributes, and properties, iii) wherein the respective modalities of the representations include one or more of: natural language representation modality, protein sequence representation modality, protein structure representation modality, or small molecule drug representation modality, iv) wherein the neural network is configured to accept as input data, a query consisting of one or more of the modalities, and to yield as output data, a response to the query, wherein the response also consists of one or more of the modalities, v) wherein the neural network is autoregressive,
vi) wherein the neural network has multiple output heads, each with its own loss function, vii) wherein for each respective output head of the neural network, the final output is a probability distribution over a set of possible values at that head; b) using a plurality of mixed modality reason-oriented query response pairs to perform supervised fine tuning of the pre-trained neural network: i) wherein for each reason-oriented query used, the pre-trained neural network's output is scored against a corresponding chain-of-thought response, ii) wherein the representation modalities of each query and each response are for representations of features of proteins, iii) wherein the represented features include one or more of sequence, structure, function, interactions, interactors, binding partners, attributes, and properties, iv) wherein the respective modalities of the representations include one or more of: natural language representation modality, protein sequence representation modality, protein structure representation modality, or small molecule drug representation modality, v) wherein an optimization process is used to iteratively update the weights of the pre-trained neural network, vi) wherein the supervised fine tuning weight updates proceed until termination criteria are met, vii) wherein the output is a reasoning-oriented mixed modality neural network; c) using the reasoning-oriented mixed modality neural network as an output generator method for obtaining a representation of an output protein, in response to an input query specifying conditions on the protein; d) using the output generator method to generate an output protein by randomly sampling the output probability distribution of the neural network's active head at each iteration of the autoregression; e) running the random-sampling based generation process a plurality of times with a given input query, wherein each of the plurality of runs uses the same input query, and wherein each of the plurality of runs yields a representation of a candidate protein; f) obtaining the plurality of generated representations of proteins as output.
15. The method of claim 14, wherein the ligand is synthesized.
22. The method of claim 21, wherein the biological properties of the protein are assessed in silico or in vitro.
14. The method of claim 13, wherein for each candidate ligand in the plurality of generated representations of peptide ligands, an assessment is made of its interaction, efficacy, and properties with the target receptor, and the most effective ligand is selected.
23. The method of claim 21, wherein the biological properties of the protein are assessed in vivo.
14. The method of claim 13, wherein for each candidate ligand in the plurality of generated representations of peptide ligands, an assessment is made of its interaction, efficacy, and properties with the target receptor, and the most effective ligand is selected.
24. The method of claim 21, wherein the protein is used as a diagnostic or therapeutic agent in a human, animal, or plant.
13. The method of claim 12, wherein the input query into the output generator specifies a target receptor and requests a peptide ligand of the receptor; and wherein the generated output is a representation of a peptide ligand of the specified target receptor.
25. A method, comprising:
a) receiving a protein:
i) wherein the protein was synthesized from a representation obtained using a neural network trained and configured to return a representation of a protein, given one or more specified conditions on the protein,
ii) wherein the neural network has at least two heads, including one output head which generates the output protein's sequence and a different output head which generates the output protein's structure,
iii) wherein during training of the neural network, there was a loss function computation corresponding to the sequence head output and a different loss function computation corresponding to the structure head output,
iv) wherein some non output head weights of the neural network were shared;
v) wherein, based on the input, the sequence head generated the sequence representation of the protein and the structure head generated the structure representation of the protein;
b) assessing biological properties of the protein in vitro or in vivo.
Claim 14, which depends from claims 12 and 13.
12. A method, comprising:
a) receiving, at a processor, a pre-trained mixed modality neural network: i) wherein the representation modalities are for representations of features of proteins, ii) wherein the represented features include one or more of sequence, structure, function, interactions, interactors, binding partners, attributes, and properties, iii) wherein the respective modalities of the representations include one or more of: natural language representation modality, protein sequence representation modality, protein structure representation modality, or small molecule drug representation modality,
iv) wherein the neural network is configured to accept as input data, a query consisting of one or more of the modalities, and to yield as output data, a response to the query, wherein the response also consists of one or more of the modalities,
v) wherein the neural network is autoregressive,
vi) wherein the neural network has multiple output heads,
each with its own loss function,
vii) wherein for each respective output head of the neural network, the final output is a probability distribution over a set of possible values at that head; b) using a plurality of mixed modality reason-oriented query response pairs to perform supervised fine tuning of the pre-trained neural network: i) wherein for each reason-oriented query used, the pre-trained neural network's output is scored against a corresponding chain-of-thought response, ii) wherein the representation modalities of each query and each response are for representations of features of proteins, iii) wherein the represented features include one or more of sequence, structure, function, interactions, interactors, binding partners, attributes, and properties, iv) wherein the respective modalities of the representations include one or more of: natural language representation modality, protein sequence representation modality, protein structure representation modality, or small molecule drug representation modality, v) wherein an optimization process is used to iteratively update the weights of the pre-trained neural network, vi) wherein the supervised fine tuning weight updates proceed until termination criteria are met, vii) wherein the output is a reasoning-oriented mixed modality neural network; c) using the reasoning-oriented mixed modality neural network as an output generator method for obtaining a representation of an output protein, in response to an input query specifying conditions on the protein;
d) using the output generator method to generate an output protein by randomly sampling the output probability distribution of the neural network's active head at each iteration of the autoregression;
e) running the random-sampling based generation process a plurality of times with a given input query, wherein each of the plurality of runs uses the same input query, and wherein each of the plurality of runs yields a representation of a candidate protein; f) obtaining the plurality of generated representations of proteins as output.
14. The method of claim 13, wherein for each candidate ligand in the plurality of generated representations of peptide ligands, an assessment is made of its interaction, efficacy, and properties with the target receptor, and the most effective ligand is selected.
26. The method of claim 25, wherein the protein is used as a diagnostic or therapeutic agent in a human, animal, or plant.
13. The method of claim 12, wherein the input query into the output generator specifies a target receptor and requests a peptide ligand of the receptor; and wherein the generated output is a representation of a peptide ligand of the specified target receptor.
27. The method of claim 25, wherein the protein is the ligand of an industrial enzyme.
13. The method of claim 12, wherein the input query into the output generator specifies a target receptor and requests a peptide ligand of the receptor; and wherein the generated output is a representation of a peptide ligand of the specified target receptor.
28. A method, comprising:
a) receiving a protein:
i) wherein the protein was synthesized from a representation obtained using a neural network trained and configured to return a representation of a protein, given one or more specified conditions on the protein,
ii) wherein the neural network has at least two heads, including one output head which generates the output protein's sequence and a different output head which generates the output protein's structure,
iii) wherein during training of the neural network, there was a loss function computation corresponding to the sequence head output and a different loss function computation corresponding to the structure head output,
iv) wherein some non output head weights of the neural network were shared;
v) wherein, based on the input, the sequence head generated the sequence representation of the protein and the structure head generated the structure representation of the protein;
b) using the protein as a diagnostic or therapeutic against in a human, animal, or plant.
Claim 13, which depends from claim 12.
12. A method, comprising:
a) receiving, at a processor, a pre-trained mixed modality neural network: i) wherein the representation modalities are for representations of features of proteins, ii) wherein the represented features include one or more of sequence, structure, function, interactions, interactors, binding partners, attributes, and properties, iii) wherein the respective modalities of the representations include one or more of: natural language representation modality, protein sequence representation modality, protein structure representation modality, or small molecule drug representation modality,
iv) wherein the neural network is configured to accept as input data, a query consisting of one or more of the modalities, and to yield as output data, a response to the query, wherein the response also consists of one or more of the modalities,
v) wherein the neural network is autoregressive,
vi) wherein the neural network has multiple output heads,
each with its own loss function,
vii) wherein for each respective output head of the neural network, the final output is a probability distribution over a set of possible values at that head; b) using a plurality of mixed modality reason-oriented query response pairs to perform supervised fine tuning of the pre-trained neural network: i) wherein for each reason-oriented query used, the pre-trained neural network's output is scored against a corresponding chain-of-thought response, ii) wherein the representation modalities of each query and each response are for representations of features of proteins, iii) wherein the represented features include one or more of sequence, structure, function, interactions, interactors, binding partners, attributes, and properties, iv) wherein the respective modalities of the representations include one or more of: natural language representation modality, protein sequence representation modality, protein structure representation modality, or small molecule drug representation modality, v) wherein an optimization process is used to iteratively update the weights of the pre-trained neural network, vi) wherein the supervised fine tuning weight updates proceed until termination criteria are met, vii) wherein the output is a reasoning-oriented mixed modality neural network; c) using the reasoning-oriented mixed modality neural network as an output generator method for obtaining a representation of an output protein, in response to an input query specifying conditions on the protein;
d) using the output generator method to generate an output protein by randomly sampling the output probability distribution of the neural network's active head at each iteration of the autoregression;
e) running the random-sampling based generation process a plurality of times with a given input query, wherein each of the plurality of runs uses the same input query, and wherein each of the plurality of runs yields a representation of a candidate protein; f) obtaining the plurality of generated representations of proteins as output.
13. The method of claim 12, wherein the input query into the output generator specifies a target receptor and requests a peptide ligand of the receptor; and wherein the generated output is a representation of a peptide ligand of the specified target receptor.
29. The method of claim 28, wherein the specified condition includes a representation of an antigen, and wherein the output protein is an associated antibody.
13. The method of claim 12, wherein the input query into the output generator specifies a target receptor and requests a peptide ligand of the receptor; and wherein the generated output is a representation of a peptide ligand of the specified target receptor.
30. The method of claim 28, wherein the specified condition includes a representation of a target receptor, and wherein the output protein is a peptide ligand of that receptor.
13. The method of claim 12, wherein the input query into the output generator specifies a target receptor and requests a peptide ligand of the receptor; and wherein the generated output is a representation of a peptide ligand of the specified target receptor.
Conclusion
The prior art made of record and not relied upon is considered pertinent to applicant's disclosure.
US-20250279161-A1: Pertinent to future potential double-patenting considerations
US-20250316329-A1: Pertinent to future potential double-patenting considerations
US-12451216-B2: Pertinent to future potential double-patenting considerations
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 Jesse P Frumkin whose telephone number is (571)270-1849. The examiner can normally be reached Monday - Saturday, 10-5 ET.
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, Olivia Wise can be reached at (571) 272-2249. 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.
/JESSE P FRUMKIN/Primary Examiner, Art Unit 1685 April 18, 2026