DETAILED ACTION
Notice of Pre-AIA or AIA Status
The present application, filed on or after March 16, 2013, is being examined under the first inventor to file provisions of the AIA .
Claim Status
Claims 1-19 are cancelled.
Claims 20-30 are currently pending and examined on the merits.
Priority
The instant application is a DIV of U.S. Application 17/512,116 filed on 10/27/2021, which is a 371 of PCT/US2020/031084 filed on 5/1/2020, which claims priority to U.S. Provisional Application 62/841,906 filed on 5/2/2019. At this point in examination, the effective filing date of claims 20-30 is 5/2/2019.
Information Disclosure Statement
The information disclosure statements (IDS) submitted on 5/31/2023, 4/4/2024, and 11/4/2024 are in compliance with the provisions of 37 CFR 1.97. A signed copy of the corresponding 1449 form has been included with this Office Action.
Claim Objections
Claims 29 objected to because of the following informalities:
In claim 29, lines 6-7, "acid residues" should read "amino acid residues".
In claim 30, line 3, a period “.” should be included after “methods”.
There is a typographical error. Appropriate correction is required.
Claim Rejections - 35 USC § 112(a)
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 30 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 30, lines 1-2 recites the synthesizing step is carried out by a computing device. para. [0065]-[0066] in the instant specification states that elements of the systems and methods discussed may be executed on any acceptable computing platform, including but not limited to a server, a cloud instance, a workstation, a thin client, a mobile device, an embedded microcontroller, a television, or any other suitable computing device. While para. [0160] states that a synthesized protein may be generated by a computing device executing a neural network or by another computing device in communication with the computing device executing a neural network, the specification does not disclose the steps for how a protein can be synthesized. The claim does not provide what the “computing device” is nor the steps used to synthesize a protein. Therefore, the specification fails to provide a written description that shows the inventor possessed the invention as recited in claim 30.
Claim Rejections - 35 USC § 112(b)
The following is a quotation of 35 U.S.C. 112(b):
(b) CONCLUSION.—The specification shall conclude with one or more claims particularly pointing out and distinctly claiming the subject matter which the inventor or a joint inventor regards as the invention.
The following is a quotation of 35 U.S.C. 112 (pre-AIA ), second paragraph:
The specification shall conclude with one or more claims particularly pointing out and distinctly claiming the subject matter which the applicant regards as his invention.
Claims 23 and 28 are rejected under 35 U.S.C. 112(b) or 35 U.S.C. 112 (pre-AIA ), second paragraph, as being indefinite for failing to particularly point out and distinctly claim the subject matter which the inventor or a joint inventor (or for applications subject to pre-AIA 35 U.S.C. 112, the applicant), regards as the invention.
The term “novel” in claim 23, line 3 is a relative term which renders the claim indefinite. The term “novel” is not defined by the claim, the specification does not provide a standard for ascertaining the requisite degree, and one of ordinary skill in the art would not be reasonably apprised of the scope of the invention. It is unclear what makes a stabilizing mutation "novel". The specification is also silent as to what "novel" means. One skilled in the art would not recognize what is considered a "novel" stabilizing mutation. Therefore, claim 23 is rendered indefinite and rejected under 35 U.S.C. 112(b).
Claim 28 recites the limitation "the measured residue" in line 3. There is insufficient antecedent basis for this limitation in the claim. The rejection might be overcome by amending the claim to introduce clear antecedent basis for “the measured residue”. For compact prosecution, it is assumed that the preceding suggested will be implemented, and the difference between the predicted candidate residue and amino acid residue and the difference between the measured residue and amino acid residue as recited in lines 2-4 are interpreted as referring to the same difference.
Claim Rejections - 35 USC § 101
35 U.S.C. 101 reads as follows:
Whoever invents or discovers any new and useful process, machine, manufacture, or composition of matter, or any new and useful improvement thereof, may obtain a patent therefor, subject to the conditions and requirements of this title.
Claims 20-30 are rejected under 35 U.S.C. 101 because the claimed invention is directed to an abstract idea without significantly more. The claims recite: (a) mathematical concepts, (e.g., mathematical relationships, formulas or equations, mathematical calculations); and (b) mental processes, i.e., concepts performed in the human mind, (e.g., observation, evaluation, judgement, opinion).
Subject matter eligibility evaluation in accordance with MPEP 2106:
Eligibility Step 1: Claims 20-28 are directed to a method (process) of improving one or more characteristics of a target protein. Claims 29-30 are directed to a method (process) of synthesizing an amino acid sequence. Therefore, these claims are encompassed by the categories of statutory subject matter, and thus satisfy the subject matter eligibility requirements under Step 1.
[Step 1: YES]
Eligibility Step 2A: First, it is determined in Prong One whether a claim recites a judicial exception, and if so, then it is determined in Prong Two whether the recited judicial exception is integrated into a practical application of that exception.
Eligibility Step 2A, Prong One: In determining whether a claim is directed to a judicial exception, examination is performed that analyzes whether the claim recites a judicial exception, i.e., whether a law of nature, natural phenomenon, or abstract idea is set forth described in the claim.
Claims 20-22 and 24-29 recite the following steps which fall within the mental processes and/or mathematical concepts groups of abstract ideas, as noted below.
Independent claim 20 further recites:
analyzing an amino acid sequence of a target protein using a trained neural network to identify one or more amino acid residues, at certain positions of the amino acid sequence, as candidate residues for mutation (i.e., mental processes);
identifying, with the neural network, one or more predicted amino acid residues for use as substitutes for at least one of the candidate residues (i.e., mental processes).
Dependent claim 21 further recites:
identifying, with the neural network, one or more predicted amino acid residues for use as substitutes for at least one other of the candidate residues (i.e., mental processes).
Dependent claim 22 further recites:
identifying, with the neural network, one or more predicted amino acid residues for use as substitutes for each of the candidate residues (i.e., mental processes).
Dependent claim 24 further recites:
the neural network is trained by: (a) generating a multi-dimensional array representative of a folded protein having a given sequence of amino acid residues, said folded protein exhibiting one or more attributes associated with a microenvironment of each amino acid residue (i.e., mental processes, mathematical concepts);
the neural network is trained by: (b) pre-processing the multi-dimensional array into a vector (i.e., mental processes);
the neural network is trained by: (c) calculating, via the neural network from the pre-processed vector, a predicted amino acid residue at a center of a microenvironment associated with the folded protein (i.e., mental processes, mathematical concepts);
the neural network is trained by: (d) determining a difference between the predicted amino acid residue and the amino acid residue associated with the microenvironment (i.e., mental processes);
the neural network is trained by: (e) responsive to the determined difference exceeding a threshold, iteratively repeating steps (a)-(d) for a different folded protein (i.e., mental processes, mathematical concepts).
Dependent claim 25 further recites:
generating the multi-dimensional array from a sample of one or more amino acids from the amino acid sequence of the target protein (i.e., mental processes, mathematical concepts).
Dependent claim 26 further recites:
wherein generating the multi-dimensional array further comprises mapping a three-dimensional model of the folded protein to a voxelized matrix (i.e., mental processes, mathematical concepts).
Dependent claim 27 further recites:
wherein pre-processing the multi-dimensional array further comprises: for each of one or more convolutional layers of the neural network, extracting a feature from a subset of the multi-dimensional array and down-sampling the extracted feature to generate a feature-specific map (i.e., mental processes, mathematical concepts);
wherein pre-processing the multi-dimensional array further comprises: combining the feature-specific maps into a one-dimensional vector (i.e., mental processes).
Dependent claim 28 further recites:
wherein step (e) further comprises modifying one or more neuron weights of the neural network, responsive to the difference between the predicted candidate residue and amino acid residue and the measured residue and amino acid residue (i.e., mental processes, mathematical concepts).
Independent claim 29 further recites:
identifying, from a series of amino acids of a protein by a trained neural network executed by a computing device, one or more amino acid residues at certain positions of the amino acid sequence as candidate residues for mutation (i.e., mental processes);
selecting, by the neural network from a second one or more acid residues, a first substitute residue for substitution of a first candidate residue, responsive to a prediction by the neural network that substitution of the first candidate residue with the first substitute residue will cause the protein to exhibit at least one improved characteristic (i.e., mental processes).
The abstract ideas recited in the claims are evaluated under the broadest reasonable interpretation (BRI) of the claim limitations when read in light of and consistent with the specification. As noted in the foregoing section, the claims are determined to contain limitations that can practically be performed in the human mind with the aid of a pencil and paper, and therefore recite judicial exceptions from the mental process grouping of abstract ideas. Additionally, the recited limitations that are identified as judicial exceptions from the mathematical concepts grouping of abstract ideas are abstract ideas irrespective of whether or not the limitations are practical to perform in the human mind.
Therefore, claims 20-22 and 24-29 recite an abstract idea.
[Step 2A, Prong One: YES]
Eligibility Step 2A, Prong Two: In determining whether a claim is directed to a judicial exception, further examination is performed that analyzes if the claim recites additional elements that, when examined as a whole, integrates the judicial exception(s) into a practical application (MPEP 2106.04(d)). A claim that integrates a judicial exception into a practical application will apply, rely on, or use the judicial exception in a manner that imposes a meaningful limit on the judicial exception. The claimed additional elements are analyzed to determine if the abstract idea is integrated into a practical application (MPEP 2106.04(d)(I); MPEP 2106.05(a-h)). If the claim contains no additional elements beyond the abstract idea, the claim fails to integrate the abstract idea into a practical application (MPEP 2106.04(d)(III)).
The judicial exceptions identified in Eligibility Step 2A, Prong One are not integrated into a practical application because of the reasons noted below.
Claim 20 recites analyzing an amino acid sequence of a target protein using a trained neural network. Claims 20-22 recite identifying one or more predicted amino acid residues for use as substitutes with a neural network. Claim 24 recites (c) calculating a predicted amino acid residue at a center of a microenvironment via a neural network, which is iteratively repeated for a different folded protein in step (e). Claim 29 recites identifying one or more amino acid residues at certain positions of an amino acid sequence from a series of amino acids of a protein by a trained neural network and selecting a first substitute residue for substitution of a first candidate residue from a second one or more acid residues by the neural network. These limitations recite using a neural network, which provide nothing more than mere instructions to implement an abstract idea on a generic computer. See MPEP 2106.05(f). Therefore, the claimed additional element does not integrate the abstract ideas into a practical application.
Claim 23 recites synthesizing a mutant protein by making one or more substitutions, in which the mutated protein includes a novel stabilizing mutation and exhibits one or more improved characteristics over those of the target protein. Claim 29 recites synthesizing the protein with the first substitute residue in place of the first candidate residue responsive to the selection. Claim 30 recites the synthesizing step is carried out by a computing device, protein synthesis, or protein expression using recombinant methods. Synthesizing a protein is considered a well-understood, routine, and conventional activity. Data gathering steps are extra-solution activity as they collect the data needed to carry out the JE. It does not impose any meaningful limitation on the JE or how the JE is performed (MPEP 2106.04/.05, citing Intellectual Ventures LLC v. Symantee Corp, McRO, TLI communications, OIP Techs. Inc. v. Amason.com Inc., Electric Power Group LLC v. Alstrom S.A.). Therefore, the claimed additional elements do not integrate the abstract ideas into a practical application.
Thus, the additionally recited elements merely invoke a computer as a tool, and/or amount to insignificant extra-solution data gathering activity, and as such, when all limitations in claims 20-30 have been considered as a whole, the claims are deemed to not recite any additional elements that would integrate a judicial exception into a practical application. Claims 20-24 and 29-30 contain additional elements that would not integrate a judicial exception into a practical application and are further probed for inventive concept in Step 2B.
[Step 2A, Prong Two: NO]
Eligibility Step 2B: Because the claims recite an abstract idea, and do not integrate that abstract idea into a practical application, the claims are probed for a specific inventive concept. The judicial exception alone cannot provide that inventive concept or practical application (MPEP 2106.05). Identifying whether the additional elements beyond the abstract idea amount to such an inventive concept requires considering the additional elements individually and in combination to determine if they amount to significantly more than the judicial exception (MPEP 2106.05A i-vi).
The claims do not include any additional elements that are sufficient to amount to significantly more than the judicial exception(s) because of the reasons noted below.
The additional element of using a neural network to analyze an amino acid sequence of a target protein (claim 20), identify predicted amino acid residues for use as substitutes (claims 20-22), calculate a predicted amino acid residue at a center of a microenvironment (claim 24), identify amino acid residues at certain positions of an amino acid sequence and select a substitute residue for substitution of a candidate residue (claim 29) is conventional. Evidence for conventionality is shown by Wang et al. (Scientific Reports, 2018, 8(6349), 1-9). Wang et al. reviews “In this study, we applied the deep learning neural network approach to computational protein design for predicting the probability of 20 natural amino acids on each residue in a protein” (pg. 1, Abstract, lines 6-8). Also, further reviews “using a sliding window method to predict the residue identity of each position one by one while considering the target residue and its neighboring residues in three-dimensional spaces, with the assumption that the identity of the target residue should be compatible with its surrounding residues.” (pg. 2, para. 2, lines 3-6). This shows that a neural network could be used to perform amino acid residue analysis in protein sequences, which makes it a conventional element in the art.
The additional element of synthesizing a mutant protein by making substitutions (claim 23), synthesizing a protein with a substitute residue in place of a candidate residue (claim 29), and the synthesizing step being carried out by a computing device, protein synthesis, or protein expression using recombinant methods (claim 30) is conventional. Evidence for conventionality is shown by Morrison et al. (Current Opinion in Chemical Biology, 2001, 5(3), 302-307). Morrison et al. reviews “Alanine-scanning mutagenesis, a method of systematic alanine substitution, has been particularly useful for the identification of functional epitopes.” (pg. 303, col. 1, para. 3, lines 1-3). Figure 1 also depicts combinatorial libraries of alanine substitutions introduced into a wild-type protein in place of sidechain functional groups (pg. 302, Figure 1). This shows that alanine-scanning mutagenesis is a protein synthesis method that synthesizes a mutant protein by substituting an amino acid residue in place of another amino acid residue, which makes it a conventional element in the art.
[Step 2B: NO]
Therefore, claims 1-20 are patent ineligible under 35 U.S.C. § 101.
Claim Rejections - 35 USC § 103
In the event the determination of the status of the application as subject to AIA 35 U.S.C. 102 and 103 (or as subject to pre-AIA 35 U.S.C. 102 and 103) is incorrect, any correction of the statutory basis (i.e., changing from AIA to pre-AIA ) for the rejection will not be considered a new ground of rejection if the prior art relied upon, and the rationale supporting the rejection, would be the same under either status.
The following is a quotation of 35 U.S.C. 103 which forms the basis for all obviousness rejections set forth in this Office action:
A patent for a claimed invention may not be obtained, notwithstanding that the claimed invention is not identically disclosed as set forth in section 102, if the differences between the claimed invention and the prior art are such that the claimed invention as a whole would have been obvious before the effective filing date of the claimed invention to a person having ordinary skill in the art to which the claimed invention pertains. Patentability shall not be negated by the manner in which the invention was made.
The factual inquiries for establishing a background for determining obviousness under 35 U.S.C. 103 are summarized as follows:
1. Determining the scope and contents of the prior art.
2. Ascertaining the differences between the prior art and the claims at issue.
3. Resolving the level of ordinary skill in the pertinent art.
4. Considering objective evidence present in the application indicating obviousness or nonobviousness.
Claims 20-22 and 24-28 are rejected under 35 U.S.C. 103 as being unpatentable over Torng et al. (BMC Bioinformatics, 2017, 18(302), 1-23), as provided in the IDS filed 5/31/2023.
With respect to claim 20:
Regarding the recited analyzing an amino acid sequence of a target protein using a trained neural network to identify one or more amino acid residues, at certain positions of the amino acid sequence, as candidate residues for mutation, Torng et al. discloses T4 lysozyme mutant structures obtained from the SCOPe2.6 database that were fed into a trained 3DCNN to determine candidate residues for mutation at corresponding variant sites (pg. 10, col. 2, para. 2, lines 1-3; pg. 10, col. 2, para. 3, lines 11-15; pg. 13, col. 1-2, para. 4, lines 1-11). As shown in Figure 4, the 3DCNN identifies a candidate residue to mutate in a target (pg. 8, Figure 4). Table 6 also depicts the mutant true column as the candidate residues for mutation at variant sites (pg. 17, Table 6). This teaches using a neural network to identify amino acid residues in certain positions of an amino acid sequence as candidate residues for mutation.
Regarding the recited identifying, with the neural network, one or more predicted amino acid residues for use as substitutes for at least one of the candidate residues, Torng et al. discloses predicting optimal residue types for both wild type and mutant structures at corresponding variant sites, which results for the 40 sites are summarized in Table 6 (pg. 13, col. 1-2, para. 4, lines 1-11). Figure 4 also depicts a predicted amino acid residue at the end of the 3DCNN workflow (pg. 8, Figure 4). This teaches using a neural network to identify a predicted amino acid residue to be substituted for a candidate residue.
With respect to claim 21:
Regarding the recited identifying, with the neural network, one or more predicted amino acid residues for use as substitutes for at least one other of the candidate residues, Torng et al. discloses predicting optimal residue types for both wild type and mutant structures at corresponding variant sites, which results for the 40 sites are summarized in Table 6 (pg. 13, col. 1-2, para. 4, lines 1-11). Figure 4 also depicts a predicted amino acid residue at the end of the 3DCNN workflow (pg. 8, Figure 4). This teaches using a neural network to identify a predicted amino acid residue to be substituted for a candidate residue.
With respect to claim 22:
Regarding the recited identifying, with the neural network, one or more predicted amino acid residues for use as substitutes for each of the candidate residues, Torng et al. discloses predicting optimal residue types for both wild type and mutant structures at corresponding variant sites, which results for the 40 sites are summarized in Table 6 (pg. 13, col. 1-2, para. 4, lines 1-11). Figure 4 also depicts a predicted amino acid residue at the end of the 3DCNN workflow (pg. 8, Figure 4). This teaches using a neural network to identify predicted amino acid residues to be substituted for candidate residues.
With respect to claim 24:
Regarding the recited the neural network is trained by: (a) generating a multi-dimensional array representative of a folded protein having a given sequence of amino acid residues, said folded protein exhibiting one or more attributes associated with a microenvironment of each amino acid residue, Torng et al. discloses a local box extraction and featurization procedure, where protein microenvironments are represented as four atom “channels” (analogous to red, green, and blue channels in images) in a 20 Å box around a central location within a protein microenvironment (pg. 2-3, col. 2, para. 3, lines 9-18). Each local 20 Å box is further divided into 1-Å 3D voxels, within which the presence of atom of the corresponding atom type is recorded in their respective channels (pg. 4, col. 2, para. 1, lines 1-4). The resulting numerical 3D matrices of each atom type channel are then stacked together as different input channels, resulting in a (4, 20, 20, 20) 4D-tensor, which serves as an input to the 3DCNN (pg. 5, Figure 2, lines 3-5). Figures 1 and 2 depict the local box extraction and featurization process in detail. This teaches generating a multi-dimensional array representative of a folded protein exhibiting attributes associated with microenvironments of each amino acid residue.
Regarding the recited the neural network is trained by: (b) pre-processing the multi-dimensional array into a vector, Torng et al. discloses processing the 4D-tensor produced from local box featurization into the 3DCNN to generate feature vectors (pg. 8, Figure 4a). This teaches pre-processing a multi-dimensional array into a vector.
Regarding the recited the neural network is trained by: (c) calculating, via the neural network from the pre-processed vector, a predicted amino acid residue at a center of a microenvironment associated with the folded protein, Torng et al. discloses using the feature vectors generated from the first half of the 3DCNN workflow to calculate for a predicted amino acid residue in the Softmax classifier layer (pg. 8, Figure 4a). Feature vectors were produced from pre-processing the 4D-tensor, which represents stacked atom type channels extracted from a 20 Å box around a central location within a protein microenvironment (pg. 2-3, col. 2, para. 3, lines 9-18; pg. 4, col. 2, para. 1, lines 1-4; pg. 5, Figure 2, lines 3-5). The predicted amino acid residues are also further depicted in the predicted mutants column of Table 6 for T4 lysozyme mutant structures (pg. 17, Table 6). Therefore, a predicted amino acid residue is calculated using a neural network on a pre-processed vector at a center of a microenvironment associated with the folded protein.
Regarding the recited the neural network is trained by: (d) determining a difference between the predicted amino acid residue and the amino acid residue associated with the microenvironment, Torng et al. discloses a prediction error between the predicted amino acid residue and the target amino acid residue associated with a microenvironment (pg. 8, Figure 4a). This teaches determining a difference between amino acid residues.
Regarding the recited the neural network is trained by: (e) responsive to the determined difference exceeding a threshold, iteratively repeating steps (a)-(d) for a different folded protein, Torng et al. discloses training the 3DCNN and MLP model on a final dataset containing 722,000 training, 38,000 validation, and 36,000 test examples, each comprising an approximately equal number of examples from all 20 amino acid microenvironment types (pg. 9, col. 2, para. 1, lines 11-14; pg. 12, col. 1, para. 1, lines 6-9). Figure 4 also further depicts backpropagation from the determined difference to drive parameter updates and learn best features for optimal performances (pg. 8, Figure 4). This teaches iteratively repeating training steps on a large dataset when a determined difference between amino acid residues is too big or exceeding a threshold, in order to minimize the error.
With respect to claim 25:
Regarding the recited generating the multi-dimensional array from a sample of one or more amino acids from the amino acid sequence of the target protein, Torng et al. discloses a local box extraction and featurization procedure, where protein microenvironments are represented as four atom “channels” (analogous to red, green, and blue channels in images) in a 20 Å box around a central location within a protein microenvironment (pg. 2-3, col. 2, para. 3, lines 9-18). Each local 20 Å box is further divided into 1-Å 3D voxels, within which the presence of atom of the corresponding atom type is recorded in their respective channels (pg. 4, col. 2, para. 1, lines 1-4). The resulting numerical 3D matrices of each atom type channel are then stacked together as different input channels, resulting in a (4, 20, 20, 20) 4D-tensor, which serves as an input to the 3DCNN (pg. 5, Figure 2, lines 3-5). Figures 1 and 2 depict the local box extraction and featurization process in detail, which shows that amino acid side chains are used for the array. This teaches generating a multi-dimensional array from one or more amino acids from an amino acid sequence of a target protein.
With respect to claim 26:
Regarding the recited wherein generating the multi-dimensional array further comprises mapping a three-dimensional model of the folded protein to a voxelized matrix, Torng et al. discloses a local box featurization procedure dividing each local 20 Å box into 1-Å 3D voxels (pg. 4, col. 2, para. 1, lines 1-4). Also, further discloses that the success of CNNs at extracting features from 2D images suggests that the convolution concept can be extended to 3D and applied to proteins represented as 3D “images” (pg. 2, col. 2, para. 2, lines 22-25). This teaches mapping a three-dimensional image of a folded protein to grid voxels.
With respect to claim 27:
Regarding the recited wherein pre-processing the multi-dimensional array further comprises: for each of one or more convolutional layers of the neural network, extracting a feature from a subset of the multi-dimensional array and down-sampling the extracted feature to generate a feature-specific map, Torng et al. discloses 3D filters in the 3D convolutional layers search for recurrent spatial patterns that best capture the local biochemical features to separate the 20 amino acid microenvironments before going through Max Pooling layers, which perform down-sampling to the input to increase translational invariances of the network. By following the 3DCNN and 3D Max-Pooling layers with fully connected layers, the pooled filter responses of all filters across all positions in the protein box can be integrated (pg. 8, Figure 4a, lines 1-5). This teaches convolutional layers of the 3DCNN extracting features from the input 4D-tensor and down-sampling the features to output feature maps from the layers.
Regarding the recited wherein pre-processing the multi-dimensional array further comprises: combining the feature-specific maps into a one-dimensional vector, Torng et al. discloses the resulting pooled filter responses of all filters across all positions becomes 3DCNN feature vectors (pg. 8, Figure 4a). This teaches combining feature maps outputted from the convolutional layers into one-dimensional vectors.
With respect to claim 28:
Regarding the recited wherein step (e) further comprises modifying one or more neuron weights of the neural network, responsive to the difference between the predicted candidate residue and amino acid residue and the measured residue and amino acid residue, Torng et al. discloses training their 3DCNN using stochastic gradient descent with the back-propagation algorithm (pg. 9, col. 1-2, para. 4, lines 1-3). Also, further discloses that during the training process, the weights of each of the 3D convolutional filters are optimized to detect local spatial patterns that best capture the local biochemical features to separate the 20 amino acid microenvironments (pg. 8, col. 1, para. 1, lines 2-6). Figure 4 depicts back propagation based on the difference between the predicted amino acid residue and the target residue, which does weight updates to the convolutional layers. This teaches modifying neuron weights of the neural network in response to the difference between a predicted or measured amino acid residue and another amino acid residue.
Therefore, the differences in the prior art were encompassed in known variations or in principle known in the prior art. The rationale would have been the predictable use of prior art elements according to their established functions. KSR 550 U.S. at 417.
For these reasons, the instant claims do not recite any new element or new function or unpredictable result, and the examiner invites the applicant to provide evidence demonstrating the novel or unobvious difference between the claimed limitations and those used in the prior art, as mere argument cannot take the place of evidence lacking in the record. Estee Lauder Inc. v. L’Oreal, S.A., 129 F .3d 588, 595 (Fed. Cir. 1997).
Claims 23 and 29-30 are rejected under 35 U.S.C. 103 as being unpatentable over Torng et al. (BMC Bioinformatics, 2017, 18(302), 1-23) as applied to claims 20-22 and 24-28 above, in view of Frenz (PROTEINS: Structure, Function, and Bioinformatics, 2005, 59(2), 147-151) and Estell et al. (Journal of Biological Chemistry, 1985, 260(11), 6518-6521).
Torng et al. is applied to claims 20-22 and 24-28 above.
With respect to claim 23:
Torng et al. does not disclose synthesizing a mutant protein by making one or more substitutions, in which the mutated protein includes a novel stabilizing mutation and exhibits one or more improved characteristics over those of the target protein.
However, Frenz discloses a neural network that predicts stabilizing mutation combinations from learning which mutant residue sequence positions are most crucial to stability along with the effects of mutation severity at each position (pg. 149, col. 1, para. 2, lines 1-9; pg. 150, col. 2, para. 2, lines 1-4; pg. 151, col. 1, para. 1, lines 10-13). Also, further discloses that increased mutant stability is critical to increasing lifespan and effectiveness of protein-based therapeutics in the body (pg. 147, col. 1, Abstract, lines 1-10; pg. 147, col. 2, para. 1, lines 1-8). This teaches a mutant protein that includes a new stabilizing mutation combination, where stability is an improved characteristic for prolonged effectiveness of a target protein.
Frenz does not disclose synthesizing a mutant protein by making one or more substitutions.
Estell et al. discloses preparing 19 amino acid substitutions at codon 222 in the cloned subtilisin gene using site-directed mutagenesis methods, which permit the replacement of methionine 222 with any amino acid (pg. 6518, col. 1, Abstract, lines 1-10; pg. 6518, col. 1, para. 2, lines 2-5; pg. 6518, col. 2, para. 3, lines 1-3). This teaches synthesizing a protein by making substitutions.
It would have been prima facie obvious to one of ordinary skill in the art to combine the amino acid mutation prediction method disclosed by Torng et al. with the mutated protein including a novel stabilizing mutation disclosed by Frenz and site-directed mutagenesis disclosed by Estell et al. One would be motivated to make this combination because an advantage of the neural network disclosed by Frenz is that the time taken to train the network took an average of 3 minutes on a 600 MHz Pentium III processor and unknown predictions were almost instantaneous (pg. 151, col. 1, para. 1, lines 29-36). This means that the overall prediction method with this modification would be less computationally intensive. The site-directed mutagenesis method disclosed by Estell et al. demonstrates that oxidative stability in proteins can be improved by replacement of oxidatively sensitive residues which are activity critical (pg. 6520, col. 2, para. 1, lines 4-9). Therefore, it would be obvious to synthesize a protein with site-directed mutagenesis to see if a predicted mutation exhibits a stabilizing characteristic because stability can be improved with site-directed mutagenesis. There is a likelihood of success, since amino acid mutation predictions using neural networks and protein synthesis are well known methods in the field of biochemistry. With respect to claim 29:
Frenz and Estell et al. do not disclose identifying, from a series of amino acids of a protein by a trained neural network executed by a computing device, one or more amino acid residues at certain positions of the amino acid sequence as candidate residues for mutation.
However, Torng et al. discloses T4 lysozyme mutant structures obtained from the SCOPe2.6 database that were fed into a trained 3DCNN to determine candidate residues for mutation at corresponding variant sites (pg. 10, col. 2, para. 2, lines 1-3; pg. 10, col. 2, para. 3, lines 11-15; pg. 13, col. 1-2, para. 4, lines 1-11). As shown in Figure 4, the 3DCNN identifies a candidate residue to mutate in a target (pg. 8, Figure 4). Table 6 also depicts the mutant true column as the candidate residues for mutation at variant sites (pg. 17, Table 6). This teaches using a neural network to identify amino acid residues in certain positions of an amino acid sequence as candidate residues for mutation.
Frenz and Estell et al. do not disclose selecting, by the neural network from a second one or more acid residues, a first substitute residue for substitution of a first candidate residue.
However, Torng et al. discloses predicting optimal residue types for both wild type and mutant structures at corresponding variant sites, which results for the 40 sites are summarized in Table 6 (pg. 13, col. 1-2, para. 4, lines 1-11). Figure 4 also depicts a predicted amino acid residue at the end of the 3DCNN workflow (pg. 8, Figure 4). This teaches using a neural network to select an amino acid residue predicted from one or more amino acid residues to be substituted for a candidate residue.
Torng et al. and Estell et al. do not disclose responsive to a prediction by the neural network that substitution of the first candidate residue with the first substitute residue will cause the protein to exhibit at least one improved characteristic.
However, Frenz discloses a neural network that predicts stabilizing mutation combinations from learning which mutant residue sequence positions are most crucial to stability along with the effects of mutation severity at each position (pg. 149, col. 1, para. 2, lines 1-9; pg. 150, col. 2, para. 2, lines 1-4; pg. 151, col. 1, para. 1, lines 10-13). Also, further discloses that increased mutant stability is critical to increasing lifespan and effectiveness of protein-based therapeutics in the body (pg. 147, col. 1, Abstract, lines 1-10; pg. 147, col. 2, para. 1, lines 1-8). This teaches a prediction by a neural network that a mutation combination of amino acid residues is stabilized, where stability is an improved characteristic for prolonged effectiveness of a target protein. Therefore, the combination of a selected amino acid residue substituted in place of a candidate amino acid residue can be predicted to exhibit an improved characteristic of stability.
Torng et al. and Frenz do not disclose synthesizing the protein with the first substitute residue in place of the first candidate residue responsive to the selection.
However, Estell et al. discloses preparing 19 amino acid substitutions at codon 222 in the cloned subtilisin gene using site-directed mutagenesis methods, which permit the replacement of methionine 222 with any amino acid (pg. 6518, col. 1, Abstract, lines 1-10; pg. 6518, col. 1, para. 2, lines 2-5; pg. 6518, col. 2, para. 3, lines 1-3). This teaches synthesizing a protein by substituting an amino acid residue in place of a candidate amino acid residue.
With respect to claim 30:
Torng et al. and Frenz do not disclose the synthesizing step is carried out by a computing device, protein synthesis, or protein expression using recombinant methods.
However, Estell et al. discloses preparing 19 amino acid substitutions at codon 222 in the cloned subtilisin gene using site-directed mutagenesis methods, which permit the replacement of methionine 222 with any amino acid (pg. 6518, col. 1, Abstract, lines 1-10; pg. 6518, col. 1, para. 2, lines 2-5; pg. 6518, col. 2, para. 3, lines 1-3). This teaches the synthesizing step being carried out by site-directed mutagenesis, which is a protein synthesis method.
Conclusion
No claims are allowed.
Any inquiry concerning this communication or earlier communications from the examiner should be directed to Jammy Luo whose telephone number is (571)272-2358. The examiner can normally be reached Monday - Friday, 9:00 AM - 5:00 PM EST.
Examiner interviews are available via telephone, in-person, and video conferencing using a USPTO supplied web-based collaboration tool. To schedule an interview, applicant is encouraged to use the USPTO Automated Interview Request (AIR) at http://www.uspto.gov/interviewpractice.
If attempts to reach the examiner by telephone are unsuccessful, the examiner’s supervisor, Larry D Riggs can be reached at (571)270-3062. The fax phone number for the organization where this application or proceeding is assigned is 571-273-8300.
Information regarding the status of published or unpublished applications may be obtained from Patent Center. Unpublished application information in Patent Center is available to registered users. To file and manage patent submissions in Patent Center, visit: https://patentcenter.uspto.gov. Visit https://www.uspto.gov/patents/apply/patent-center for more information about Patent Center and https://www.uspto.gov/patents/docx for information about filing in DOCX format. For additional questions, contact the Electronic Business Center (EBC) at 866-217-9197 (toll-free). If you would like assistance from a USPTO Customer Service Representative, call 800-786-9199 (IN USA OR CANADA) or 571-272-1000.
/J.N.L./Examiner, Art Unit 1686
/LARRY D RIGGS II/Supervisory Patent Examiner, Art Unit 1686
Prosecution Insights