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 .
Status of the Claims
Claims 1-20 are currently pending and under exam herein.
Claims 1-20 are rejected.
Priority
The instant application is a 371 of PCT/US2022/072488 which claims priority from provisional application 63/191,175 filed on 05/20/2021 and 63/196,637 filed on 06/03/2021. Thus, the earliest effective filing date of the instant application is 05/20/2021.
Drawings
The Drawings filed on 11/20/2023 were considered.
Information Disclosure Statement
The information disclosure statement (IDS) submitted on 02/02/2026 is in compliance with the provisions of 37 CFR 1.97. Accordingly, the information disclosure statement has been considered by the examiner.
Claim Rejections - 35 USC § 101
35 U.S.C. 101 reads as follows:
Whoever invents or discovers any new and useful process, machine, manufacture, or composition of matter, or any new and useful improvement thereof, may obtain a patent therefor, subject to the conditions and requirements of this title.
Claims 1-20 are rejected under 35 U.S.C. 101 because the claimed invention is directed to an abstract idea without significantly more. The 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 1-20 are directed to a method to determine RNA structure.
[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 or described in the claim.
Independent claim 1 recites the following steps which fall within the mental processes and/or mathematical concepts groupings of abstract ideas:
training a machine learning model with the experimentally determined RNA structure; (mental process and/or mathematical concept)
providing an RNA sequence to the trained machine learning model; (mental process and/or mathematical concept)
and determining an RNA structure for the RNA sequence with the trained machine learning model. (mental process and/or mathematical concept)
Dependent claim 6 recites the following steps which fall within the mental processes and/or mathematical concepts groupings of abstract ideas:
wherein the equivariant layer comprises at least one of: a radial function and an angular function (mathematical concept)
Dependent claim 7 recites the following steps which fall within the mental processes and/or mathematical concepts groupings of abstract ideas:
wherein the radial function encodes distances between atoms (mathematical concept)
Dependent claim 8 recites the following steps which fall within the mental processes and/or mathematical concepts groupings of abstract ideas:
wherein the angular function considers orientations between atoms. (mathematical concept)
Dependent claim 10 recites the following steps which fall within the mental processes and/or mathematical concepts groupings of abstract ideas:
wherein training the machine learning model comprises sampling a training set of RNA molecules. (mathematical concept)
Dependent claim 11 recites the following steps which fall within the mental processes and/or mathematical concepts groupings of abstract ideas:
wherein the training set of RNA molecules comprises three-dimensional coordinates and chemical element type of each atom in each RNA molecule in the training set of RNA molecules. (mathematical concept)
Dependent claim 12 recites the following steps which fall within the mental processes and/or mathematical concepts groupings of abstract ideas:
wherein sampling is selected from FARFAR2 and Monte Carlo sampling. (mathematical concept)
Dependent claim 13 recites the following steps which fall within the mental processes and/or mathematical concepts groupings of abstract ideas:
wherein training the machine learning model comprises optimizing the machine learning model. (mathematical concept)
Dependent claim 14 recites the following steps which fall within the mental processes and/or mathematical concepts groupings of abstract ideas:
wherein optimizing the machine learning model comprises selecting model parameters based on a lowest root mean square deviation (RMSD) between a predicted structure and its experimentally determined structure (mathematical concept)
Dependent claim 15 recites the following steps which fall within the mental processes and/or mathematical concepts groupings of abstract ideas:
wherein the training set comprises RNA molecules of 17-47 nucleotides. (mathematical concept)
Dependent claim 16 recites the following steps which fall within the mental processes and/or mathematical concepts groupings of abstract ideas:
wherein training the machine learning model further comprises benchmarking the machine learning model with a benchmarking set of RNA molecules. (mathematical concept)
Dependent claim 17 recites the following steps which fall within the mental processes and/or mathematical concepts groupings of abstract ideas:
wherein the benchmarking set comprises RNA molecules of 27-188 nucleotides. (mathematical concept)
Dependent claim 20 recites the following steps which fall within the mental processes and/or mathematical concepts groupings of abstract ideas:
wherein the determined RNA structure comprises both secondary and tertiary structures. (mathematical concept)
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 1-20 recite an abstract idea as the dependent claims will inherit the abstract ideas from the independent claims.
[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.
The additional element in independent claim 1 includes:
A method for determining RNA structure, comprising: (Claim 1)
obtaining an experimentally determined RNA structure; (Claim 1)
The additional element in dependent claim 2 includes:
wherein the machine learning model is a geometric deep learning neural network. (Claim 2)
The additional element in dependent claim 3 includes:
wherein the machine learning model is an equivariant neural network comprising an equivariant layer, (Claim 3)
The additional element in dependent claim 4 includes:
wherein the equivariant layer passes on rotational information to the next layer in the machine learning model. (Claim 4)
The additional element in dependent claim 5 includes:
wherein the equivariant layer passes on translational information to the next layer in the machine learning model. (Claim 5)
The additional element in dependent claim 9 includes:
wherein the equivariant neural network further comprises at least one of a self-interaction layer, a pointwise normalization layer, a pointwise normalization layer, and a fully connected layer. (Claim 9)
The additional element in dependent claim 18 includes:
obtaining a structure for a ligand; and (Claim 18)
docking the ligand to the determined RNA structure to identify if the ligand binds to the RNA sequence. (Claim 18)
The additional element in dependent claim 19 includes:
further comprising providing the ligand to an individual. (Claim 19)
The additional elements of obtaining an experimentally determined RNA structure; (Claim 1), obtaining a structure for a ligand; and (Claim 18) are insignificant extra-solution activity that are part of the data gathering process used in the recited judicial exceptions (see MPEP 2106.05(g)).
The additional elements of a method for determining RNA structure, comprising: (Claim 1), wherein the machine learning model is a geometric deep learning neural network. (Claim 2), wherein the machine learning model is an equivariant neural network comprising an equivariant layer (Claim 3), wherein the equivariant layer passes on rotational information to the next layer in the machine learning model. (Claim 4), wherein the equivariant layer passes on translational information to the next layer in the machine learning model. (Claim 5), wherein the equivariant neural network further comprises at least one of a self-interaction layer, a pointwise normalization layer, a pointwise normalization layer, and a fully connected layer. (Claim 9), docking the ligand to the determined RNA structure to identify if the ligand binds to the RNA sequence. (Claim 18), further comprising providing the ligand to an individual. (Claim 19) fail to integrate a judicial exception into a practical application merely reciting the words "apply it" (or an equivalent) with the judicial exception, or merely including instructions to implement an abstract idea on a computer, or merely using a computer as a tool to perform an abstract idea, as discussed in MPEP § 2106.05(f).
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 1-20 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, and therefore claims 1-20 are directed to an abstract idea (MPEP 2106.04(d)).
[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 elements recited in claims 1-20 are identified above and carried over from Step 2A: Prong Two along with their conclusions for analysis at Step 2B. Any additional element or combination of elements that was considered to be insignificant extra-solution activity at Step 2A: Prong Two was re-evaluated at Step 2B, because if such re-evaluation finds that the element is unconventional or otherwise more than what is well-understood, routine, conventional activity in the field, this finding may indicate that the additional element is no longer considered to be insignificant; and all additional elements and combination of elements were evaluated to determine whether any additional elements or combination of elements are other than what is well-understood, routine, conventional activity in the field, or simply append well-understood, routine, conventional activities previously known to the industry, specified at a high level of generality, to the judicial exception, per MPEP 2106.05(d).The additional elements of obtaining an experimentally determined RNA structure; (Claim 1), obtaining a structure for a ligand; and (Claim 18) are conventional and part of the data gathering process used in the recited judicial exceptions (see MPEP 2106.05(g)). Evidence for conventionality is shown by Calonaci et al, Li et al. and Guilbert et al. Each work obtains RNA structures and Guilbert et al. obtains RNA bound to ligands and discusses how RNA ligand interaction has been previously studied.
The additional elements of a method for determining RNA structure, comprising: (Claim 1), wherein the machine learning model is a geometric deep learning neural network. (Claim 2), wherein the machine learning model is an equivariant neural network comprising an equivariant layer (Claim 3), wherein the equivariant layer passes on rotational information to the next layer in the machine learning model. (Claim 4), wherein the equivariant layer passes on translational information to the next layer in the machine learning model. (Claim 5), wherein the equivariant neural network further comprises at least one of a self-interaction layer, a pointwise normalization layer, a pointwise normalization layer, and a fully connected layer. (Claim 9), docking the ligand to the determined RNA structure to identify if the ligand binds to the RNA sequence. (Claim 18), further comprising providing the ligand to an individual. (Claim 19) are conventional fail to integrate a judicial exception into a practical application merely reciting the words "apply it" (or an equivalent) with the judicial exception, or merely including instructions to implement an abstract idea on a computer, or merely using a computer as a tool to perform an abstract idea, as discussed in MPEP § 2106.05(f).
Therefore, when taken alone, all additional elements in claims 1-20 do not amount to significantly more than the above-identified judicial exception(s). Even when evaluated as a combination, the additional elements fail to transform the exception(s) into a patent-eligible application of that exception. Thus, claims 1-20 are deemed to not contribute an inventive concept, i.e., amount to significantly more than the judicial exception(s) (MPEP 2106.05(II)).
[Step 2B: NO]
Claim Rejections - 35 USC § 102
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 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.
(a)(2) the claimed invention was described in a patent issued under section 151, or in an application for patent published or deemed published under section 122(b), in which the patent or application, as the case may be, names another inventor and was effectively filed before the effective filing date of the claimed invention.
Claims 1 is rejected under 35 U.S.C. 102 as being anticipated over Calonaci et al. (Calonaci et al. Machine Learning a Model for RNA Structure Prediction. NAR Genomics and Bioinformatics 2020, 2 (4).). The italicized text corresponds to the instant claim limitations.
With respect to the limitations of Claim 1, Calonaci et al. teaches and builds and trains a machine learning that can be used to predict RNA structure taking as an input RNA sequence (Discussion, paragraph 1, A method for determining RNA structure, comprising: obtaining an experimentally determined RNA structure; training a machine learning model with the experimentally determined RNA structure; providing an RNA sequence to the trained machine learning model; and determining an RNA structure for the RNA sequence with the trained machine learning model (Claim 1)
Claim Rejections - 35 USC § 103
In the event the determination of the status of the application as subject to AIA 35 U.S.C. 102 and 103 (or as subject to pre-AIA 35 U.S.C. 102 and 103) is incorrect, any correction of the statutory basis (i.e., changing from AIA to pre-AIA ) for the rejection will not be considered a new ground of rejection if the prior art relied upon, and the rationale supporting the rejection, would be the same under either status.
The following is a quotation of 35 U.S.C. 103 which forms the basis for all obviousness rejections set forth in this Office action:
A patent for a claimed invention may not be obtained, notwithstanding that the claimed invention is not identically disclosed as set forth in section 102, if the differences between the claimed invention and the prior art are such that the claimed invention as a whole would have been obvious before the effective filing date of the claimed invention to a person having ordinary skill in the art to which the claimed invention pertains. Patentability shall not be negated by the manner in which the invention was made.
The factual inquiries for establishing a background for determining obviousness under 35 U.S.C. 103 are summarized as follows:
1. Determining the scope and contents of the prior art.
2. Ascertaining the differences between the prior art and the claims at issue.
3. Resolving the level of ordinary skill in the pertinent art.
4. Considering objective evidence present in the application indicating obviousness or nonobviousness.
This application currently names joint inventors. In considering patentability of the claims the examiner presumes that the subject matter of the various claims was commonly owned as of the effective filing date of the claimed invention(s) absent any evidence to the contrary. Applicant is advised of the obligation under 37 CFR 1.56 to point out the inventor and effective filing dates of each claim that was not commonly owned as of the effective filing date of the later invention in order for the examiner to consider the applicability of 35 U.S.C. 102(b)(2)(C) for any potential 35 U.S.C. 102(a)(2) prior art against the later invention.
Claims 2 are rejected under 35 U.S.C. 103 as being unpatentable over Calonaci et al. in view of Bronstein et al. (M. M. Bronstein, J. Bruna, Y. LeCun, A. Szlam and P. Vandergheynst, "Geometric Deep Learning: Going beyond Euclidean data," in IEEE Signal Processing Magazine, vol. 34, no. 4, pp. 18-42, July 2017,). The italicized text corresponds to the instant claim limitations.
The limitations of claim 1 have been taught by Calonaci et al. above.
Calonaci et al. does not explicitly teach
wherein the machine learning model is a geometric deep learning neural network (Claim 2
With respect to the limitations of Claims 2, Bronstein et al. teaches
A key problem in material and drug design is predicting the physical, chemical, or biological properties (such as solubility of toxicity) of a novel molecule from its structure. State-of-the-art methods rely on hand-crafted molecule descriptors, such as circular fingerprints . A recent work from Harvard University in Cambridge, Massachusetts proposed modeling molecules as graphs (where vertices represent atoms and edges represent chemical bonds) and employing GCNNs to learn the desired molecule properties. The authors' approach has significantly outperformed handcrafted features. This work opens a new avenue in molecule design that might revolutionize the field (Molecule Design, wherein the machine learning model is a geometric deep learning neural network (Claim 2)
A person having ordinary skill in the art would be motivated to combine Calonaci et al. with Bronstein et al. because it explicitly suggests how Geometric deep learning is useful for solving molecular structure leading to a person of ordinary skill in the art to use geometric deep learning to help solve molecular structure. There is a reasonable expectation of success because each step works independently, and applicant is just defining how their neural network is set up in dependent claims. The neural network architecture has been shown to work therefore there is a reasonable expectation of success when put together.
Claim 3-9 is/are rejected under 35 U.S.C. 103 as being unpatentable over Calonaci et al. in view of Bronstein et al. in further view of Thomas et al. (Thomas et al, Tensor field networks: Rotation- and translation-equivariant neural networks for 3D point clouds, Arxiv, May, 18, 2018). The italicized text corresponds to the instant claim limitations.
The limitations of claims 1-2 have been taught by over Calonaci et al. in view of Bronstein et al. above.
Calonaci et al. in view of Bronstein et al. does not explicitly teach
wherein the machine learning model is an equivariant neural network comprising an equivariant layer (Claim 3),
wherein the equivariant layer passes on rotational information to the next layer in the machine learning model. (Claim 4),
wherein the equivariant layer passes on translational information to the next layer in the machine learning model. (Claim 5),
wherein the equivariant layer comprises at least one of: a radial function and an angular function. (Claim 6)
wherein the radial function encodes distances between atoms. (Claim 7)
wherein the angular function considers orientations between atoms. (Claim 8),
wherein the equivariant neural network further comprises at least one of a self-interaction layer, a pointwise normalization layer, a pointwise normalization layer, and a fully connected layer (Claim 9)
However, these limitations were known in the art at the time of the effective filing date of the invention, as taught by Thomas et al. teaches
With respect to the limitations of Claims 3-9, Thomas et al. teaches introduce tensor field neural networks, which are locally equivariant to 3D rotations, translations, and permutations of points at every layer. 3D rotation equivariance removes the need for data augmentation to identify features in arbitrary orientations. Our network uses filters built from spherical harmonics; due to the mathematical consequences of this filter choice, each layer accepts as input (and guarantees as output) scalars, vectors, and higher-order tensors, in the geometric sense of these terms. We demonstrate the capabilities of tensor field networks with tasks in geometry, physics, and chemistry (abstract) Thomas et al. explains the use of equivariant neural network and how to apply them to both both Translation equivariance (section 3.2) and Translation equivariance (section 3.3) passing on the information to the next layer. Their work differs from CNN by containing radial function (Motivation). The radial function includes distance between atoms. (5.2 Physics: vectors and tensors in classical mechanics) r (vector) normalized to unit length, and r is the length of r. and (4.1 Point convolution, Figure 1) The spherical harmonics Y (l) m are functions from the points on a sphere to the complex or real numbers (where l is a non-negative integer and m is be any integer between −l and l inclu sive). The (real) spherical harmonics for l = 0 (scalars) and l = 1 (3-space vectors). To design a rotation-equivariant point convolution, we want rotation-equivariant filters. For our filters to be rotation-equivariant, we restrict them to the following form: hich contain most of the parameters within a tensor field network). Filters of this form inherit the transformation property of spherical harmonics under rotations because R(r) is a scalar in m. This choice of filter restriction is analogous to the use of circular harmonics in Worrall et al. [1] (though we do not have an analog to the phase offset because of the non-commutativity of SO(3)) (4.1.1 Spherical harmonics and filters) using point convolutions to scale feature vectors elementwise and using self-interaction layers to mix together the components of the feature vectors at each point. In general, each rotation order has different weights because there may be different numbers of channels corresponding to that rotation order. However, the same weights are used for every m for a given order; this is essential to maintain equivariance. For l = 0, we may also use biases. The D-matrices commute with the weight matrix W because W has no representation index m, so this layer is equivariant for l > 0. Equivariance for l = 0 is straightforward because D(0) = 1. (4.2 Self-interaction) wherein the machine learning model is an equivariant neural network comprising an equivariant layer (Claim 3), wherein the equivariant layer passes on rotational information to the next layer in the machine learning model. (Claim 4), wherein the equivariant layer passes on translational information to the next layer in the machine learning model. (Claim 5), wherein the equivariant layer comprises at least one of: a radial function and an angular function. (Claim 6) wherein the radial function encodes distances between atoms. (Claim 7) wherein the angular function considers orientations between atoms. (Claim 8), wherein the equivariant neural network further comprises at least one of a self-interaction layer, a pointwise normalization layer, a pointwise normalization layer, and a fully connected layer (Claim 9)
A person having ordinary skill in the art would be motivated to combine Calonaci et al. in view of Bronstein et al. with Thomas et al. discusses how their methods are ideal for determining 3d structure in space and specifically discusses using their method for generating molecular structures. Therefore, a person of ordinary skill in the art would be motivated to use this method of generating molecules in 3d space. There is a reasonable expectation of success because each step works independently, and applicant is just defining how their neural network is set up in dependent claims. The neural network architecture has been shown to work therefore there is a reasonable expectation of success when put together.
Claim 10-17, 20 is/are rejected under 35 U.S.C. 103 as being unpatentable over Calonaci et al. in view of Bronstein et al. in view of Thomas et al. in further view of Li et al. (Li, J.; Zhu, W.; Wang, J.; Li, W.; Gong, S.; Zhang, J.; Wang, W. RNA3DCNN: Local and Global Quality Assessments of RNA 3D Structures Using 3D Deep Convolutional Neural Networks. PLOS Computational Biology 2018, 14 (11), e1006514 ) The italicized text corresponds to the instant claim limitations.
The limitations of claims 1-9 have been taught by Calonaci et al. in view of Bronstein et al. in further view of Thomas et al. above.
With respect to the limitations of Claims 20, Calonaci et al. teaches the resulting model leads to secondary structure prediction that surpasses available methods when used on a validation set not seen in the training phase (Introduction paragraph 2, wherein the determined RNA structure comprises secondary structures (Claim 20)
Calonaci et al. in view of Bronstein et al. in further view of Thomas et al. does not explicitly teach
wherein training the machine learning model comprises sampling a training set of RNA molecules. (Claim 10), wherein the training set of RNA molecules comprises three-dimensional coordinates and chemical element type of each atom in each RNA molecule in the training set of RNA molecules. (Claim 11), wherein sampling is selected from FARFAR2 and Monte Carlo sampling. (Claim 12), wherein training the machine learning model comprises optimizing the machine learning model. (Claim 13), wherein optimizing the machine learning model comprises selecting model parameters based on a lowest root mean square deviation (RMSD) between a predicted structure and its experimentally determined structure (Claim 14), wherein training the machine learning model further comprises benchmarking the machine learning model with a benchmarking set of RNA molecules (Claim 16), wherein the benchmarking set comprises RNA molecules of 27-188 nucleotides (Claim 17), wherein the determined RNA structure comprises tertiary structures (Claim 20)
However, these limitations were known in the art at the time of the effective filing date of the invention, as taught by Li et al. teaches
With respect to the limitations of Claims 10-17, 20, Li teaches to construct the training dataset, first a list of 619 RNAs was downloaded with the search options “RNA Only” and “Non Redundant RNA Structures” which means that our training dataset includes RNA-only structures and the RNAs are non-redundant in both sequence and geometry. Second, the RNAs with an X-ray resolution >3.5 Å were removed from the list above. Finally, the RNAs in the test dataset were removed and the RNAs in the equivalence classes with the test dataset were also removed. “Structures that are provisionally redundant based on sequence similarity and also geometrical similarity are grouped into one equivalence class,” as Leontis et al. defined. Thus, 414 native RNAs were left to construct the training dataset. According to their length, the 414 RNAs were randomly divided into two groups, namely, 332 RNAs for training and 82 RNAs for validation in the CNN training process. Practically, the training samples were generated in two ways, namely, by MD and MC methods elaborated as follows. (Training dataset) We used a macromolecular structure modeling software Rosetta to sample RNA 3D structures based on a fragment assembly method, which is a MC process guided by a low-resolution knowledge-based energy function. The models were further refined in an all-atom potential to yield more realistic structures. For each of the 414 RNAs, we fed the sequence and secondary structural information to Rosetta. The predicted models were clustered and 300 structures were picked out for each RNA. And like in MD methods, there were 1 million training samples and 0.2 million validation samples extracted from the 332 × 300 and 82 × 300 structures randomly based on their nucleotide unfitness scores, respectively. (Training samples generated by MC methods.) Test dataset III comes from RNA-Puzzles rounds I to III a collective and blind experiment in 3D RNA structure prediction. Given the nucleotide sequences, interested groups submit their predicted structures to the RNA-Puzzles website before the experimentally determined crystallographic or NMR structures of these target sequences are published. Therefore, the dataset is produced in a real RNA modeling scenario and can reveal the real performance of the existing scoring function (Test dataset) developed a 3D CNN-based approach named RNA3DCNN. This approach uses raw atom distributions in 3D space as the input of neural networks and the output is an RMSD-based nucleotide unfitness score for each nucleotide in an RNA molecule, thus making it possible to evaluate local structural quality. Here, we tested and made comparisons with four other traditional scoring functions on three test datasets from different sources. (Author summary) (A) Surrounding environment extraction. The magenta color depicts the nucleotide under assessment. Based on the assessed nucleotide, a local Cartesian coordinate system xyz is determined and accordingly a cube with sides of length 32 Å is drawn as yellow solid lines. See the section Environment surrounding a nucleotide for details. The surrounding environment is precisely defined by the cyan-colored atoms within the box and the gray atoms beyond the box do not influence the assessment of the magenta nucleotide. (B) RNA voxelizations. The box in (A) is partitioned into 32 × 32 × 32 grid boxes drawn as yellow dashed lines, but for visual convenience, only 3 × 3 × 3 grid boxes are drawn. Each grid box resembles a voxel of three channels representing atomic occupation number, mass, and charge within the grid box. As shown in (B), an assessed nucleotide and its surrounding environment are voxelated into a 3D image of 32 × 32 × 32 voxels. The RGB channels are mapped to the three channels. The large and small spheres represent the voxels occupied by atoms and nothing, respectively, while the size difference of the spheres does not represent anything but is only for visual comparison. (C) CNN architecture. The network is arranged in the order of input layer, convolutional layer C1, convolutional layer C2, maxpooling layer M3, convolutional layer C4, convolutional layer C5, and output layer. Blue, red, and green cubes represent the input layer, convolutional layer, and maxpooling layer, respectively. The yellow stick represents the output layer. m@n3 means that there are m channels and n3 voxels. (Figure 1 Summary) A back-propagation-based mini-batch gradient descent optimization algorithm was used to optimize the parameters in the network (CNN training) Table 3 shows lengths of nucleotides including ones with length of 41 (Table 3), wherein training the machine learning model comprises sampling a training set of RNA molecules. (Claim 10), wherein the training set of RNA molecules comprises three-dimensional coordinates and chemical element type of each atom in each RNA molecule in the training set of RNA molecules. (Claim 11), wherein sampling is selected from FARFAR2 and Monte Carlo sampling. (Claim 12), wherein training the machine learning model comprises optimizing the machine learning model. (Claim 13), wherein optimizing the machine learning model comprises selecting model parameters based on a lowest root mean square deviation (RMSD) between a predicted structure and its experimentally determined structure (Claim 14), wherein training the machine learning model further comprises benchmarking the machine learning model with a benchmarking set of RNA molecules (Claim 16), wherein the benchmarking set comprises RNA molecules of 27-188 nucleotides (Claim 17), wherein the determined RNA structure comprises tertiary structures (Claim 20)
A person having ordinary skill in the art would be motivated to combine Calonaci et al. in view of Bronstein et al. in view of Thomas et al. with Li et al. because it teaches a method for designing RNA molecules in 3d space. Therefore, a person of ordinary skill in the art would be motivated to use this method of generating RNA in 3d space. There is a reasonable expectation of success because each step works independently, and applicant is just defining how their neural network is set up in dependent claims. The neural network architecture has been shown to work therefore there is a reasonable expectation of success when put together.
Claim 18, 19 is/are rejected under 35 U.S.C. 103 as being unpatentable over Calonaci et al. in view of Bronstein et al. in view of Thomas et al. in view of Li et al. in further view of Guilbert et al. () The italicized text corresponds to the instant claim limitations.
The limitations of claims 1-17, 20 have been taught by over Calonaci et al. in view of Bronstein et al. in view of Thomas et al. in further view of Li et al. above.
Calonaci et al. in view of Bronstein et al. in view of Thomas et al. in view of Li et al. does not explicitly teach
further comprising: obtaining a structure for a ligand; and docking the ligand to the determined RNA structure to identify if the ligand binds to the RNA sequence. (Claim 18),
further comprising providing the ligand to an individual (Claim 19)
With respect to the limitations of Claims 18, 19, Guilbert et al. teaches structure-based drug design is now well-established for proteins as a key first step in the lengthy process of developing new drugs. In many ways, RNA may be a better target to treat disease than a protein because it is upstream in the translation pathway, so inhibiting a single mRNA molecule could prevent the production of thousands of protein gene products. Virtual screening is often the starting point for structure-based drug design. However, computational docking of a small molecule to RNA seems to be more challenging than that to protein due to the higher intrinsic flexibility and highly charged structure of RNA. Previous attempts at docking to RNA showed the need for a new approach. We present here a novel algorithm using molecular simulation techniques to account for both nucleic acid and ligand flexibility. In this approach, with both the ligand and the receptor permitted some flexibility, they can bind one another via an induced fit, as the flexible ligand probes the surface of the receptor. A possible ligand can explore a low-energy path at the surface of the receptor by carrying out energy minimization with root-mean-square-distance constraints. Our procedure was tested on 57 RNA complexes (33 crystal and 24 NMR structures); this is the largest data set to date to reproduce experimental RNA binding poses. With our procedure, the lowest-energy conformations reproduced the experimental binding poses within an atomic root-mean-square deviation of 2.5 Å for 74% of tested complexes. (abstract, further comprising: obtaining a structure for a ligand; and docking the ligand to the determined RNA structure to identify if the ligand binds to the RNA sequence. (Claim 18), further comprising providing the ligand to an individual (Claim 19)
A person having ordinary skill in the art would be motivated to combine Calonaci et al. in view of Bronstein et al. in view of Thomas et al. in view of Li et al. with Guilbert et al. because it provides the motivation to understand ligand RNA interaction as RNA is upstream of proteins so it may be a valuable target in drug design. (abstract) Guilbert et al. simply suggests to extend the architecture to include ligand interactions. There is a reasonable expectation of success because each step works independently, and applicant is just defining how their neural network is set up in dependent claims. The neural network architecture has been shown to work therefore there is a reasonable expectation of success when put together.
Conclusion
Any inquiry concerning this communication or earlier communications from the examiner should be directed to Connor Beveridge whose telephone number is 571-272-2099. The examiner can normally be reached Monday - Thursday 9 am - 5 pm.
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.
/C.H.B./Examiner, Art Unit 1687
/Karlheinz R. Skowronek/Supervisory Patent Examiner, Art Unit 1687
Prosecution Insights