DETAILED ACTION
Notice of 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 .
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 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.
Withdrawal of Objections and Rejections
Applicant's response, filed 04/02/2026, has been fully considered.
In view of the amendment and remarks from 04/02/2026, the objection to the claims and the rejection of the following claims are withdrawn:
claims 1-21 under 35 USC § 112(b);
The following rejections and/or objections are either maintained or newly applied for claims 1 and 3-20. They constitute the complete set applied to the instant application. Herein, "the previous Office action" refers to the Non-Final Rejection of 01/14/2026.
Status of the Claims
Claims 2 and 21 are canceled.
Claims 1 and 3-20 are pending.
Claims 1 and 3-20 are rejected.
Priority
This US Application 17/849,269 (06/24/2022) claims priority from US Application 63/215,357 (06/25/2021), as reflected in the filing receipt mailed on 07/01/2022. The claims to the benefit of priority are acknowledged; and the effective filing date of claims 1 and 3-20 is 06/25/2021.
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 and 3-20 are rejected under 35 USC § 101 because the claimed inventions are directed to one or more Judicial Exceptions (JEs) without significantly more. Regarding JEs, "Claims directed to nothing more than abstract ideas..., natural phenomena, and laws of nature are not eligible for patent protection" (MPEP 2106.04 §I). Abstract ideas include mathematical concepts and procedures for evaluating, analyzing or organizing information, which are a type of mental process (MPEP 2106.04(a)(2)). Any newly recited portions are necessitated by claim amendment.
101 background
MPEP 2106 organizes JE analysis into Steps 1, 2A (Prong One & Prong Two), and 2B as analyzed below. MPEP 2106 and the following USPTO website provide further explanation and case law citations: uspto.gov/patent/laws-and-regulations/examination-policy/examination-guidance-and-training-materials.
Step 1: Are the claims directed to a process, machine, manufacture, or composition of matter (MPEP 2106.03)?
Step 2A, Prong One: Do the claims recite a judicially recognized exception, i.e., a law of nature, a natural phenomenon, or an abstract idea (MPEP 2106.04(a-c))?
Step 2A, Prong Two: If the claims recite a judicial exception under Prong One, then is the judicial exception integrated into a practical application by an additional element (MPEP 2106.04(d))?
Step 2B: Do the claims recite a non-conventional arrangement of elements in addition to any identified judicial exception(s) (MPEP 2106.05)?
Analysis of instant claims
Step 1: Are the claims directed to a 101 process, machine, manufacture, or composition of matter (MPEP 2106.03)?
The instant claims are directed to a method (claims 1 and 3-18), a system (claim 19), and a CRM (claim 20);each of which falls within one of the categories of statutory subject matter.
[Step 1: claims 1 and 3-20: Yes]
Step 2A, Prong One: Do the claims recite a judicially recognized exception, i.e., a law of nature, a natural phenomenon, or an abstract idea (MPEP 2106.04(a-c))?
Background
With respect to Step 2A, Prong One, the claims recite judicial exceptions in the form of abstract ideas. MPEP § 2106.04(a)(2) further explains that abstract ideas are defined as:
• mathematical concepts (mathematical formulas or equations, mathematical relationships
and mathematical calculations) (MPEP 2106.04(a)(2)(I));
• certain methods of organizing human activity (fundamental economic principles or practices, managing personal behavior or relationships or interactions between people) (MPEP 2106.04(a)(2)(II)); and/or
• mental processes (concepts practically performed in the human mind, including observations, evaluations, judgments, and opinions) (MPEP 2106.04(a)(2)(III)).
Analysis of instant claims
With respect to the instant claims, under the Step 2A, Prong One evaluation, the claims are found to recite abstract ideas that fall into the grouping of mathematical concepts (in particular mathematical relationships and formulas) and mental processes (in particular procedures for observing, analyzing and organizing information) are as follows.
Mathematical concepts (in particular mathematical relationships and formulas) include:
• "after the training, generating a plurality of conformations for at least a portion of the macromolecule that each include respective three-dimensional coordinates in the coordinate system of each of the plurality of atoms, comprising, for each conformation" (independent claims 1 and 19-20);
• "sampling a conformation latent representation from a prior distribution over conformation latent representations" (independent claims 1 and 19-20);
• "processing the sampled conformation latent representation …to generate, as an output of an output layer of the decoder neural network, a conformation output that specifies three-dimensional coordinates in the coordinate system of each of the plurality of atoms for the conformation" (independent claims 1 and 19-20);
• "generating the conformation from the conformation output" (independent claims 1 and 19-20);
• "for each of the plurality of atoms, applying the respective three-dimensional displacement vector for the atom specified by the conformation output for the atom to the base three-dimensional coordinates for the atom to generate the respective three-dimensional coordinates for the atom" (claim 3);
• "processing the image … to generate an encoder output" (claim 7);
• "sampling a set of conformation latent representations from the conformation posterior distribution in accordance with the parameters of the conformation posterior distribution in the …output" (claim 7);
• "processing each of the conformation latent representations in the set … to generate a respective decoder output for each of the conformation latent representations" (claim 7);
• "generating a respective reconstruction of the image from each of the decoder outputs using a differentiable renderer" (claim 7);
• "training … on a loss function that includes one or more loss terms that measure, for each image in the batch, an error between the image and the respective reconstructions of the image generated from the decoder output for the image" (claim 7);
• "sampling a pose latent representation from the pose posterior distribution over pose latent representations in accordance with the parameters of the pose posterior distribution" (claim 13);
• "generating the respective reconstruction of the image using the sampled pose latent representation and the differentiable renderer" (claim 13);
• "generating … three-dimensional coordinates of each of the plurality of atoms" (claim 14);
• "modifying a pose of the plurality of atoms using the sampled pose latent to generate modified three-dimensional coordinates of each of the plurality of atoms" (claim 14); and
• "applying the differentiable renderer to the modified three-dimensional coordinates of each of the plurality of atoms to generate the respective reconstruction" (claim 14).
The claims identified above read on math. The abstract ideas recited in the claims are evaluated under the Broadest Reasonable Interpretation and determined each element performed either in the mind and/or by mathematical operation. Without further detail as to the methodology involved in "executing a model comprising algorithmic rules for gene prioritization and represent temporal evolution of the state values to generate simulation results", under the BRI, one may simply, for example, use pen and paper to perform mathematical steps to arrive at the described steps. Further support for the mathematical techniques used in the claims is provided in the specification at pg. 1 para. 1, which discloses an algorithm implemented in the processing unit which generates a network map based on the plurality of datasets in the database and which allows for identifying nodes within network map based on predefined parameters; and at pg. 9 para. 3 which discloses the execution of logical operations. Thus, the recited terms correspond to verbal equivalents of mathematical concepts because they constitute actions executed by a group of mathematical steps in a form of a mathematical algorithm; thus mathematical concepts (MPEP 2106.04(a)(2)). A mathematical concept need not be expressed in mathematical symbols, because "words used in a claim operating on data to solve a problem can serve the same purpose as a formula." In re Grams, 888 F.2d 835, 837 and n.1, 12 USPQ2d 1824, 1826 and n.1 (Fed. Cir. 1989). MPEP 2106.04(a)(2) pertains.
Mental processes, defined as concepts or steps practically performed in the human mind such as steps of observations, evaluations, judgments, analysis, opinions or organizing information include:
• "determining the base conformation for the macromolecule through a single state reconstruction" (claim 4).
Under the BRI, the recited limitations are mental processes because a human mind is also sufficiently capable of determine a conformation for a molecule by drawing it by hand.
Dependent claims 5, 8, 13 and 15 recite further steps that limit the judicial exceptions in independent claim 1 and, as such, also are directed to those abstract ideas. For example, claim 5 recites further details about the delta parameter; claim 8 recites further details about the "conformation latent representation"; dependent claims 9-12 recites further details about the loss function; claim 13 recites further details about the "encoder network"; claim 15 recites further details about the "decoder network."
[Step 2A Prong One: claims 3-20: Yes ]
Step 2A, Prong Two: If the claims recite a judicial exception under Prong One, then is the judicial exception integrated into a practical application by an additional element (MPEP 2106.04(d))?
Background
MPEP 2106.04(d).I lists the following example considerations for evaluating whether a judicial exception is integrated into a practical application:
An improvement in the functioning of a computer or an improvement to other technology or another technical field, as discussed in MPEP §§ 2106.04(d)(1) and 2106.05(a);
Applying or using a judicial exception to effect a particular treatment or prophylaxis for a disease or medical condition, as discussed in MPEP § 2106.04(d)(2);
Implementing a judicial exception with, or using a judicial exception in conjunction with, a particular machine or manufacture that is integral to the claim, as discussed in MPEP § 2106.05(b);
Effecting a transformation or reduction of a particular article to a different state or thing, as discussed in MPEP § 2106.05(c); and
Applying or using the judicial exception in some other meaningful way beyond generally linking the use of the judicial exception to a particular technological environment, such that the claim as a whole is more than a drafting effort designed to monopolize the exception, as discussed in MPEP § 2106.05(e).
Analysis of instant claims
Instant claims 1, 6-7 and 19-20 recite additional elements that are not abstract ideas:
• "training a decoder neural network on the plurality of images " (independent claims 1 and 19-20);
• "one or more computers" (independent claim 1);
• "one or more computers and one or more storage devices storing instructions" (independent claim 19);
• "one or more non-transitory computer storage media storing instructions" (independent claim 20);
• "obtaining a plurality of images of a macromolecule having a plurality of atoms" (independent claims 1 and 19-20);
• "receive an input comprising a conformation latent representation of a conformation of the macromolecule and to process the input to generate a conformation output that specifies three-dimensional coordinates in a coordinate system of each of the plurality of atoms, wherein the conformation output specifies, for each of the plurality of atoms, a respective three-dimensional displacement vector for the atom relative to a three-dimensional spatial position of the atom in a base conformation of the macromolecule, wherein the base conformation is a predetermined structural template for the macromolecule" (independent claims 1 and 19-20);
• "encoder neural network" (claims 1, 7 and 19-20);
• "decoder neural network" (claims 1, 7 and 19-20);
• "obtaining a batch of one or more images from the plurality of images" (claim 7); and
• "receive an image of the macromolecule and to process the image to generate an encoder output that comprises parameters of a conformation posterior distribution over the conformation latent representations" (claim 6).
Dependent claims 16-18 recite further details about the "plurality of images" received.
Considerations under Step 2A, Prong Two
The recited limitations in claims 1, 6-7 and 19-20 are interpreted as requiring the use of a computer. Hence, the claims explicitly recite steps executed by computers and therefore can be described as computer functions or instructions to implement on a generic computer.
Further steps directed to additional non-abstract elements of a computing device/computer do not describe any specific computational steps by which the "computer parts" perform or carry out the judicial exceptions, nor do they provide any details of how specific structures of the computer are used to implement these functions. The claims state nothing more than a generic computer which performs the functions that constitute the judicial exceptions. The instant claims state nothing more than that a generic computer performs the functions that constitute the abstract idea (MPEP 2106.05(f)).
The judicial exceptions in the claims are considered to perform the claimed abstract idea with a computer, which is not sufficient to integrate an abstract idea into a practical application (see MPEP 2106.05(f)); since steps that can be performed mentally and merely performing the mental process in a computer environment do not negate the fact that something that can be carried out in the human mind. See MPEP 2106.04(a)(2).III.C.
Claims directed to "obtaining and receive" read on receiving or transmitting data over a network -Symantec, 838 F.3d at 1321 - MPEP 2106.05(a) pertains; which constitutes just necessary data gathering and therefore correspond to insignificant extra-solution activity.
With respect to claims 1, 7 and 19-20, the computer-related elements or the general purpose computer and the recited encoder/decoder neural network model does not rise to the level of significantly more than the judicial exception. The claims state nothing more than a generic computer which performs the functions that constitute the judicial exceptions. Hence, these are mere instructions to apply the judicial exceptions using a computer, which the courts have found to not provide significantly more when recited in a claim with a judicial exception (Alice Corp., 573 U.S. at225-26, 110 USPQ2d at 1984; see MPEP 2106.05(A)). The specification as published also notes that neural network systems, as example, are known and widely used examples of neural networks that may be used without limitation (pg. 16 line 14). The additional elements are set forth at such a high level of generality that they can be met by a general purpose computer. Therefore, the computer components constitute no more than a general link to a technological environment, which is insufficient to constitute an inventive concept that would render the claims significantly more than the judicial exceptions (see MPEP 2106.05(b)I-III).
Hence, these are mere instructions to apply the abstract idea using a computer and insignificant extra-solution activity and therefore the claims do not integrate that abstract idea into a practical application (see MPEP 2106.04(d) § I; 2106.05(f); and 2106.05(g)).
In Step 2A, Prong One above, claim steps and/or elements were identified as part of one or more judicial exceptions (JEs).
In this Step 2A, Prong Two immediately above claim steps and/or elements were identified as part of one or more additional elements. Additional elements are further discussed in Step 2B below.
Here in Step 2A, Prong Two, no additional step or element clearly demonstrates integration of the JE(s) into a practical application.
[Step 2A Prong Two: claims 1 and 3-20: No]
Step 2B: Do the claims recite a non-conventional arrangement of elements in addition to any identified judicial exception(s) (MPEP 2106.05)?
According to analysis so far, the additional elements described above do not provide significantly more than the judicial exception. A determination of whether additional elements provide significantly more also rests on whether the additional elements or a combination of elements represents other than what is well-understood, routine, and conventional. Conventionality is a question of fact and may be evidenced as: a citation to an express statement in the specification or to a statement made by an applicant during examination that demonstrates a well-understood, routine or conventional nature of the additional element(s); a citation to one or more of the court decisions as discussed in MPEP 2106(d)(II) as noting the well-understood, routine, conventional nature of the additional element(s); a citation to a publication that demonstrates the well-understood, routine, conventional nature of the additional element(s); and/or a statement that the examiner is taking official notice with respect to the well-understood, routine, conventional nature of the additional element(s).
Claims 1, 6-7 and 19-20 recite a computer or computer functions, interpreted as instructions to apply the abstract idea using a computer, where the computer does not impose meaningful limitations on the judicial exceptions; which can be performed without the use of a computer (MPEP 2106.04(d) § I; and MPEP 2106.05(f)).
The computer-related elements or the general purpose computer and the neural network model do not rise to the level of significantly more than the judicial exception. The claims state a generic computer which performs the functions that constitute the judicial exceptions. Hence, these are mere instructions to apply the judicial exceptions using a computer, which the courts have found to not provide significantly more when recited in a claim with a judicial exception (Alice Corp., 573 U.S. at225-26, 110 USPQ2d at 1984; see MPEP 2106.05(A)).
Further, the courts have found that receiving is well-understood, routine, and conventional functions of a computer when claimed in a generic manner or as insignificant extra-solution activity (see Symantec, 838 F.3d at 1321, 120 USPQ2d at 1362 (utilizing an intermediary computer to forward information), buySAFE, Inc. v. Google, Inc., 765 F.3d 1350, 1355, 112 USPQ2d 1093, 1096 (Fed. Cir. 2014) (computer receives and sends information over a network), Versa ta Dev. Group, Inc. v. SAP Am., Inc., 793 F.3d 1306, 1334, 115 USPQ2d 1681, 1701 (Fed. Cir. 2015), and OIP Techs., 788 F.3d at 1363, 115 USPQ2d at 1092-93, as discussed in MPEP 2106.05(d)(Il)(i)).
When the claims are considered as a whole, they do not integrate the abstract idea into a practical application; they do not confine the use of the abstract idea to a particular technology; they do not solve a problem rooted in or arising from the use of a particular technology; they do not improve a technology by allowing the technology to perform a function that it previously was not capable of performing; and they do not provide any limitations beyond generally linking the use of the abstract idea to a broad technological environment. See MPEP 2106.05(a) and 2106.05(h).
The instant claims constitute insignificant extra solution activity, and when considered individually, are insufficient to constitute inventive concepts that would render the claims significantly more than an abstract idea (see MPEP 2106.05(g)). Hence, these elements, when considered individually, are insufficient to constitute inventive concepts that would render the claims significantly more than an abstract idea (see MPEP 2106.05(d)).
[Step 2B: claims 1 and 3-20: No]
Conclusion: Instant claims are directed to non-statutory subject matter
For the reasons above, the claims in this instant application, when the limitations are considered individually and as a whole, are directed to an abstract idea and lack an inventive concept not clearly anything significantly more.
Response to applicant's remarks in regard to Claim Rejection 35 U.S.C. ~ 101
The Remarks of 04/02/2026 have been fully considered but are not persuasive for the reasons below:
Applicant asserts starting in pg. 11 para. 5:
The claimed invention is patent eligible under the reasoning provided in MPEP § 2106.04(d)(l), which explains that ''the specification should be evaluated to determine if the disclosure provides sufficient details such that one of ordinary skill in the art would recognize the claimed invention as providing an improvement in the functioning of a computer, or an improvement to other technology or a technical field... if the specification sets forth an improvement in technology or a technical field, the claim must be evaluated to ensure that the claim itself reflects the disclosed improvement, i.e., that the claim includes the components or steps of the invention that provide the improvement described in the specification. The present invention provides an improvement to computer functionality, by facilitating molecular modeling with reduced sampling requirements and computational resources … Thus, as described above, the claimed invention provides an improvement to the technical field of molecular modeling by generating conformations computationally in a way that is accurate and less computationally intensive. For example, training a neural network to determine macromolecule distributions is a difficult computational problem that often requires a prohibitively large number of samples, making training computationally expensive. The claimed invention addresses this issue by configuring the decoder neural network to generate three-dimensional atomic coordinate displacements relative to a base conformation. This provides a practical computational shortcut where the decoder neural network manipulates a useful structural template instead of rebuilding complex geometry from scratch, significantly narrowing the structural search space … information across all conformations through continuous deformation, which is a technical advantage not possible for conventional techniques, e.g., as described in Zhong, that output grids of voxels. This architecture allows the claimed invention to leverage data correlations far more efficiently to reconstruct accurate conformation distributions from a limited number of images …The claims thus improve computer functionality by facilitating molecular modeling with significantly reduced sampling requirements and computational resources.
It is respectfully submitted that this is not persuasive because, although claims are interpreted in light of the specification, examples or details from the specification are not imported into the claims. (MPEP 2173.05(b) pertains). The analysis at Step 2A, Prong 2, considers the claims as a whole, i.e., the additional elements in combination with the judicial exceptions (see MPEP 2106.05(a)), although the integration or improvement provided in the claim must flow from the additional elements and not the judicial exceptions to be considered persuasive. The limitations related to the argued molecular modeling pointed by the Applicant are all considered to recite a judicial exceptions as described above and are therefore not considered at Step 2 Prong 2. All asserted arguments referring to "improvement to computer functionality, by facilitating molecular modeling with reduced sampling requirements and computational resources" are not accompanied by any evidence that suggests that anything other than a generic computer would be needed to perform said computer functions. Further steps directed to additional non-abstract elements of a computing device/computer do not describe any specific computational steps by which the "computer parts" perform or carry out the judicial exceptions, nor do they provide any details of how specific structures of the computer are used to implement these functions. The claims state nothing more than a generic computer which performs the functions that constitute the judicial exceptions. The instant claims state nothing more than that a generic computer performs the functions that constitute the abstract idea (MPEP 2106.05(f)).
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)(l) 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, 3-9, 12-13 and 15-20 are rejected under 35 U.S.C. 102(a)(1) as being anticipated by Zhong ("CryoDRGN: reconstruction of heterogeneous structures from cryo-electron micrographs using neural networks" bioRxiv (2020)) as evidenced by Paranjothy ("Direct chemical dynamics simulations: coupling of classical and quasiclassical trajectories with electronic structure theory." Wiley Interdisciplinary Reviews: Computational Molecular Science 3(3) 296-316 (2013)),as cited on the 01/14/2026 Form PTO-892. Any newly recited portions are necessitated by claim amendment.
Claim 1 recites a method performed by one or more computers comprising steps. Claim 19 recites a system comprising one or more computers and one or more storage devices storing instructions that when executed by the one or more computers cause the one more computers to perform operations comprising said steps. Claim 20 recites one or more non-transitory computer storage media storing instructions that when executed by one or more computers cause the one or more computers to perform operations comprising said steps.
The prior art to Zhong discloses a method and a system related to an algorithm - Deep Reconstructing Generative Networks - that leverages the representation power of deep neural networks to directly reconstruct continuous distributions of 3D density maps and map per-particle heterogeneity of single-particle cryo-EM datasets (pg. 176 para. 1); wherein steps are taught via the use of a custom Python script available in the cryoDRGN software (pg. 186 col. 2 para. 9).
The steps performed by the method of claim 1, a system of claim 19, and a non-transitory computer-readable media of claim 20 comprise:
obtaining a plurality of images of a macromolecule having a plurality of atoms
• Zhong teaches an algorithm performing heterogeneous reconstruction by learning a deep generative model of 3D structure from single-particle cryo-EM images (i.e. plurality of images) (pg. 177 col. 1 para. 2).
training a decoder neural network on the plurality of images, wherein the decoder neural network is configured to receive an input comprising a conformation latent representation of a conformation of the macromolecule and to process the input to generate a conformation output that specifies three-dimensional coordinates in a coordinate system of each of the plurality of atoms, wherein the conformation output specifies, for each of the plurality of atoms, a respective three-dimensional displacement vector for the atom relative to a three-dimensional spatial position of the atom in a base conformation of the macromolecule, wherein the base conformation is a predetermined structural template for the macromolecule
• Zhong teaches a method consisting a specialized image-encoder–volume-decoder architecture, which learns an encoding of the inputted two-dimensional (2D) particle images into a continuous vector space described by the latent space (i.e. input comprising a conformation latent representation) (pg. 177 col. 1 para. 2); wherein the trained decoder network can then generate 3D density maps (pg. 177 col. 2 para. 2) using coordinate-based networks to represent 3D structure (pg. 186 col. 1 para. 1) and map per-particle heterogeneity of single-particle cryo-EM datasets (pg. 176 para. 1); wherein the model was applied to the macromolecule protein complex RAG1–RAG2 (i.e. conformation output that specifies three-dimensional coordinates of each of the plurality of atoms) (pg. 178 Fig. 2); wherein the algorithm can reconstruct 3D density maps from user-defined positions in latent space (i.e. three-dimensional spatial position of the atom) enabling visualization of structural trajectories (i.e. a respective three-dimensional displacement vector for the atom relative to a three-dimensional spatial position of the atom in a base conformation of the macromolecule) (pg. 184 col. 1para. 2).
after the training, generating a plurality of conformations for at least a portion of the macromolecule that each include respective three-dimensional coordinates in the coordinate system of each of the plurality of atoms, comprising, for each conformation: sampling a conformation latent representation from a prior distribution over conformation latent representations
• Zhong teaches as, once the model is trained, the encoder network encodes the inputted images into a latent space that allows for visualization and inspection of particle distribution from which representative structures can be generated from regions of latent space with high particle density and continuous conformational trajectories can be reconstructed by sampling points along a trajectory through latent space (i.e. sampling from a prior distribution over conformation latent representations) (pg. 177 col. 2 para. 2).
processing the sampled conformation latent representation using the decoder neural network to generate, as an output of an output layer of the decoder neural network, a conformation output that specifies three-dimensional coordinates in the coordinate system of each of the plurality of atoms for the conformation and generating the conformation from the conformation output
• Zhong teaches as, in sampling structures from the latent space (i.e. processing a respective input comprising sampled conformations), the generated density maps (i.e. output that specifies three-dimensional coordinates of each of the plurality of atoms) revealed expected spliceosome conformations from the largest cluster, poorly resolved structures as seen in Fig. 6B (i.e. generating the conformation from the conformation output) (pg. 182 col. 2 para. 2); wherein cryoDRGN volumes were generated from decoder networks with three hidden layers and 1,024 nodes per hidden layer (i.e. as an output of an output layer of the decoder neural network) (pg. 178 Fig. 2).
Claim 3 recites:
wherein generating the conformation from the conformation output comprises: for each of the plurality of atoms, applying the respective three-dimensional displacement vector for the atom specified by the conformation output for the atom to the base three-dimensional coordinates for the atom to generate the respective three-dimensional coordinates for the atom
• Zhong teaches as simulated particle images being generated by rotating the density map with a random rotation sampled uniformly and by shifting the image with an in-plane translation sampled uniformly (i.e. applying the respective delta specified by the conformation output) (pg. 186 col. 1 para. 6); wherein each coordinate is featurized with a fixed positional encoding function that follows a geometric progression from 1 (i.e. initial/base coordinate) pg. 186 col. 1 para. 1); wherein the algorithm can reconstruct 3D density maps from user-defined positions in latent space (i.e. three-dimensional spatial position of the atom) enabling visualization of structural trajectories (i.e. a respective three-dimensional displacement vector for the atom relative to a three-dimensional spatial position of the atom in a base conformation of the macromolecule) (pg. 184 col. 1para. 2); wherein structural trajectories inherently comprise the recited 3D displacement vectors to perform said trajectories calculations (i.e. as evidenced by Paranjothy pg. 300 col. 1 para. 3).
Claim 4 recites:
determining the base conformation for the macromolecule through a single state reconstruction
• Zhong teaches as homogeneous reconstruction (i.e. single state reconstruction) of the particle images constituting a large ribosomal subunit cluster which reproduced a similar, albeit lower-resolution structure, confirming the existence of this structural state in the original dataset (i.e. determining the base conformation for the macromolecule) (pg. 182 col. 2 para. 1)
Claim 5 recites:
wherein the conformation output specifies, for each of a plurality of residues that each include one or more of the plurality of atoms, a respective relative translation and relative rotation for the residue relative to a position of the residue in the base conformation
• Zhong teaches a density map generated from the atomic models (i.e. from a base conformation) (pg. 186 col. 1 para. 6); wherein datasets used in reconstruction models were produced by rotating a single dihedral angle of a hypothetical protein complex to simulate a conformational transition along a one-dimensional (1D) reaction coordinate (i.e. inherently when a protein complex rotates so does the residues that makeup the protein complex) (pg. 17 col. 2 para. 2); wherein reconstructed density maps reproduced the ground-truth continuous motion of the complex (i.e. inherent relative translation of the protein complex and its inherent residues) (pg. 179 col. 1 para. 1).
Claim 6 recites:
wherein training the decoder neural network on the plurality of images comprises: training the decoder neural network jointly with an encoder neural network that is configured to receive an image of the macromolecule and to process the image to generate an encoder output that comprises parameters of a conformation posterior distribution over the conformation latent representations
• Zhong teaches a specialized image-encoder–volume-decoder architecture in which the neural networks are jointly trained (i.e. training the decoder neural network jointly with an encoder neural network) (pg. 177 col. 1 para. 2); wherein The encoder and decoder are parameterized with fully connected neural networks with parameters ξ and θ, respectively (pg. 186 col. 1 para. 4); wherein after training, the encoder network encodes particle images (i.e. processing of images) into the continuous latent space, outputting per-particle latent encodings, zi (i.e. latent variable), describing the dataset’s heterogeneity (i.e. parameters of a conformation latent representations) which allows for visualization and inspection of particle distribution (i.e. posterior distribution over the conformation) (pg. 177 col. 2 para. 2); enabling the visualization of data-supported motions in the RAG complex (i.e. macromolecule) (pg. 184 col. 1 para. 2).
Claim 7 recites:
wherein training the decoder neural network jointly with the encoder neural network comprises:
obtaining a batch of one or more images from the plurality of images;
• Zhong teaches as training system for the image-encoder–volume-decoder architecture in which observed images are generated from projections (i.e. obtaining images from images) of a volume at a random unknown orientation (pg. 186 col. 1 para. 3).
for each image in the batch: processing the image using the encoder neural network to generate an encoder
output; sampling a set of conformation latent representations from the conformation posterior distribution in accordance with the parameters of the conformation posterior distribution in the encoder output;
• Zhong teaches as the use of a positionally encoded multilayer perceptron to model structures generated from an n-dimensional continuous latent space (pg. 186 col. 1 para. 1); wherein given an image, the variational encoder produces a mean and variance statistics that parameterize a Gaussian distribution with diagonal covariance, as the variational approximation to the true posterior (pg. 186 col. 1 para. 4).
processing each of the conformation latent representations in the set using the decoder neural network to generate a respective decoder output for each of the conformation latent representations;
• Zhong teaches as, given a sample out of the encoder and the oriented coordinates, an image can be reconstructed pixel by pixel through the decoder (pg. 186 col. 1 para. 5).
generating a respective reconstruction of the image from each of the decoder outputs using a differentiable renderer;
• Zhong teaches as the reconstructed image is translated by the image’s in-plane shift (i.e. projection) and multiplied by the contrast transfer function (CTF) before it is compared to the input image (i.e. projection followed by CTF is equivalent to the renderer's function) (pg. 186 col.1 para. 5).
training the encoder neural network and the decoder neural network on a loss function that includes one or more loss terms that measure, for each image in the batch, an error between the image and the respective reconstructions of the image generated from the decoder output for the image
• Zhong teaches as the neural the networks being trained by multiple passes through the dataset, with lower values of the objective loss function as training progressed (pg. 178 col. 1 para. 2 and Fig. 2d); wherein the model computed the negative log likelihood of a given image as the mean square error between the reconstructed image and the input image (i.e. an error between the image and the respective reconstructions of the image) (pg. 186 col. 1 para. 5).
Claim 8 recites:
wherein the set of conformation latent representations includes a plurality of conformation latent representations
• Zhong teaches as a plurality of conformations in the latent space representation for macromolecules such as the large ribosomal subunit pg. 181 Fig. 5f/g) and the RAG complex (pg. 180 Fig. 4c).
Claim 9 recites:
wherein the loss function includes one or more auxiliary loss terms that measure, for each decoder output, a deviation of a structure of the macromolecule as specified by the three-dimensional coordinates of each of the plurality of atoms from an expected structure of the macromolecule
• Zhong teaches as the optimization objective being a variational lower bound of the model evidence with the first term representing the reconstruction error estimated with one Monte Carlo sample and the second Kullback–Leibler divergence term representing a regularization term on the latent representation (i.e. auxiliary loss terms that measure deviation of a structure) (pg. 186 col. 1 para. 5).
Claim 12 recites:
wherein the loss function includes one or more terms that measure, for each encoder output, a divergence between the conformation posterior distribution and the prior distribution in accordance with the parameters specified in the encoder output
• Zhong teaches as the variational encoder producing a mean and variance statistics that parameterize a Gaussian distribution with diagonal covariance, as the variational approximation to the true posterior; wherein the prior on the latent variable is a standard normal distribution (pg. 186 col. 1 para. 4) (i.e. the Kullback–Leibler term of the objective loss function of the variational autoencoder accounts for the divergence/variance between the posterior distribution and the prior distribution in accordance with the parameters specified in the encoder output).
Claim 13 recites:
wherein: the encoder neural network is configured to process the image to generate an encoded representation of the image and to process the encoded representation of the image to generate the parameters of the conformation posterior distribution over the conformation latent representations;
• Zhong teaches as during model training, the full dataset of particle images is encoded into the latent space (pg. 177 Fig. 1); with trainable parameters (pg. 178 col. 2 para. 1), from which the decoder can directly generate 3D density maps (i.e. posterior distribution) (pg. 177 Fig. 1).
the encoder neural network is configured to process at least the encoded representation to generate parameters of a pose posterior distribution over pose latent representations;
• Zhong teaches as pose estimates being sufficiently accurate to generate meaningful latent space encodings and to produce interpretable density maps of distinct structures (pg. 184 col. 2 para. 3)
generating a respective reconstruction of the image from each of the decoder outputs using the differentiable renderer comprises: sampling a pose latent representation from the pose posterior distribution over pose latent representations in accordance with the parameters of the pose posterior distribution
• Zhong teaches a trained 10D latent variable model on the sampled images using image poses derived from a consensus reconstruction (i.e. sampling a pose from posterior distribution) reporting multiple clusters observed in the latent space encodings (i.e. latent space representation) of the dataset’s particle images (pg. 182 col. 2 para. 2).
for each decoder output, generating the respective reconstruction of the image using the sampled pose latent representation and the differentiable renderer
• Zhong teaches as reconstructed images translated by the image’s in-plane shift (i.e. projection) and multiplied by the contrast transfer function (CTF) before it is compared to the input image (i.e. projection followed by CTF is equivalent to the renderer's function) with results presented employing training that used previously determined poses from a consensus reconstruction (i.e. reconstruction of the image using the sampled pose latent representation) (pg. 186 col.1 para. 5).
Claim 15 recites:
wherein the decoder neural network is configured to process the input to generate a decoded representation of the conformation latent representation and to process the decoded representation to generate the decoder output, and wherein the encoder neural network is configured to process the encoded representation of the image and respective decoded representation of each conformation latent representation in the set to generate the parameters of the pose posterior distribution over pose latent representations
• Zhong teaches the training of encoder and decoder neural networks jointly (pg. 177 col. 1 para. 1); wherein the encoder encode each particle image into the low-dimensional latent space (pg. 177 Fig. 1) from which the decoder can directly generate density maps, representing an ensemble of 3D density maps (i.e. posterior distribution) (pg. 177 Fig. 1); wherein pose estimates are sufficiently accurate to generate meaningful latent space encodings (i.e. use of parameters over pose latent representations) and to produce interpretable density maps of distinct structures (pg. 184 col. 2 para. 3).
Claim 16 recites:
wherein the plurality of images are Cryo-electron microscopy (cryo-EM) images of the macromolecule
Claim 17 recites:
wherein the plurality of images are picked particle images of the macromolecule
Claim 18 recites:
wherein the macromolecule is a protein
• Zhong teaches the recited claims as the use of neural networks to map per-particle heterogeneity of single-particle cryo-EM datasets (i.e. as in claim 16) (pg. 176 para. 1); wherein the model was applied to the macromolecule protein complex RAG1–RAG2 (i.e. macromolecule protein as in claims 17-18) (pg. 178 Fig. 2).
Claim Rejections - 35 USC § 103
The following is a quotation of pre-AIA 35 U.S.C. 103(a) which forms the basis for all obviousness rejections set forth in this Office action:
(a) A patent may not be obtained though the invention is not identically disclosed or described as set forth in section 102, if the differences between the subject matter sought to be patented and the prior art are such that the subject matter as a whole would have been obvious at the time the invention was made to a person having ordinary skill in the art to which said subject matter 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 pre-AIA 35 U.S.C. 103(a) 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.
A. Claims 10-11 are rejected under 35 U.S.C. 103(a) as being unpatentable over Zhong as applied to claim 1 in the 102 rejection above further in view of Ramaswamy ("Deep learning protein conformational space with convolutions and latent interpolations." Physical Review X 11(1):011052 (March 2021)), as cited on the 01/14/2026 Form PTO-892. Any newly recited portions are necessitated by claim amendment.
Claim 10 recites:
wherein the auxiliary loss terms include a first auxiliary loss term that measures a deviation between (i) bond lengths along a backbone of the macromolecule in the structure specified by the three-dimensional coordinates of each of the plurality of atoms and (ii) expected bond lengths along the backbone of the macromolecule
• Zhong does not teach the recitation above. However, the prior art to Ramaswamy teaches it as a convolutional neural network that learns a continuous conformational space representation from example structures, and loss functions that ensure intermediates between examples are physically plausible (pg. 1 para. 1); wherein the physics-based terms in the loss function helped in generating structures with correct bond lengths, angles, and dihedrals (i.e. reading on expected backbone lengths, angles/dihedrals) (pg. 5 col. 2 para. 1).
Claim 11 recites:
wherein the auxiliary loss terms include a second auxiliary loss term that measures a deviation between (i) a center of mass of the structure specified by the three-dimensional coordinates of each of the plurality of atoms and (ii) an expected center of mass of the structure
• Zhong does not teach the recitation above. However, the prior art to Ramaswamy teaches it as the model tracking the center of mass of the protein domain with respect to its connection to the rest of the protein and reporting it in spherical coordinates (pg. 4 col. 2 para. 2); reporting loss functions mean and standard deviation (pg. 11 col. 1 para. 5).
Rationale for combining (MPEP §2142-2143)
Regarding claims 10-11, it would have been prima facie obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to combine, in the course of routine experimentation and with a reasonable expectation of success, the methods of Zhong in view of Ramaswamy because all references disclose methods for applying objective loss functions to sample conformation space. The motivation would have been to sample conformational space of macromolecules using loss functions to ensure intermediates between examples are physically plausible, as taught by Ramaswamy (pg. 1 para. 1 Ramaswamy).
Therefore it would have been obvious to one of ordinary skill in the art to substitute applying objective loss functions to sample conformation space of Zhong to the methods by Ramaswamy because such a substitution is no more than the simple substitution of one known element for another. One of ordinary skill in the art would be able to motivated to combine the teachings in these references with a reasonable expectation of success since the described teachings pertain to methods for applying objective loss functions to sample conformation space.
B. Claim 14 is rejected under 35 U.S.C. 103(a) as being unpatentable over Zhong as applied to claim 1 in the 102 rejection above further in view of Shugurov ("Multi-view object pose refinement with differentiable renderer." IEEE Robotics and Automation Letters 6(2):2579-2586 (April 2021)), as cited on the 01/14/2026 Form PTO-892. Any newly recited portions are necessitated by claim amendment.
Claim 14 recites:
wherein generating the respective reconstruction of the image using the sampled pose latent representation and the differentiable renderer comprises:
generating, from the decoder output, three-dimensional coordinates of each of the
plurality of atoms;
modifying a pose of the plurality of atoms using the sampled pose latent to generate modified three-dimensional coordinates of each of the plurality of atoms; and
applying the differentiable renderer to the modified three-dimensional coordinates of
each of the plurality of atoms to generate the respective reconstruction
• Zhong does not teach the recitation above. However, Shugurov teaches it as a multi-view pose refinement (i.e. modifying a pose) with differentiable renderer (pg. 1 Title) using the three-dimensional Normalized Object Coordinates Space (pg. 4 col. 1 1para. 2) (i.e. applying the differentiable renderer to the modified three-dimensional coordinates) for image reconstruction in 3D space (pg. 3 col. 2 par. 1); wherein the loss function is implemented with a differentiable renderer and is optimized iteratively (pg. 1 col. 1 para. 1).
Rationale for combining (MPEP §2142-2143)
Regarding claim 14, it would have been prima facie obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to combine, in the course of routine experimentation and with a reasonable expectation of success, the methods of Zhong in view of Shugurov because all references disclose methods for applying a differential renderer to use sampled poses for reconstruction of 3D images. The motivation would have been to produce 3D images using a loss function implemented with a differentiable renderer, as taught by Shugurov (pg. 1 col. 1 para. 1 Shugurov).
Therefore it would have been obvious to one of ordinary skill in the art to substitute applying a differential renderer to use sampled poses for reconstruction of 3D images of Zhong to the methods by Shugurov because such a substitution is no more than the simple substitution of one known element for another. One of ordinary skill in the art would be able to motivated to combine the teachings in these references with a reasonable expectation of success since the described teachings pertain to methods for applying a differential renderer to use sampled poses for reconstruction of 3D images.
Response to applicant's remarks in regard to Claim Rejection 35 U.S.C. ~ 102/103
The Remarks of 04/02/2026 have been fully considered but are not persuasive for the reasons below:
Applicant asserts starting in pg. 15 para. 2:
Zhong makes no mention of ''a conformation output that specifies three-dimensional coordinates in a coordinate system of each of the plurality of atoms." In particular, Zhong makes no mention of a conformation output that ''specifies, for each of the plurality of atoms, a respective three-dimensional displacement vector for the atom relative to a three-dimensional spatial position of the atom in a base conformation of the macromolecule," and ''wherein the base conformation is a predetermined structural template for the macromolecule." Zhong makes no mention of a ''conformation output that specifies three-dimensional coordinates in the coordinate system of each of the plurality of atoms for the conformation'' that is ''an output of an output layer of the decoder neural network
It is respectfully submitted that this is not persuasive because Zhong's teachings indeed included the argued outputted 3D coordinates that specifies respective three-dimensional displacement vector for the atom relative to a three-dimensional spatial position. Said limitation is taught as Zhong discloses a trained decoder network that can then generate 3D density maps (pg. 177 col. 2 para. 2) using coordinate-based networks to represent 3D structure (pg. 186 col. 1 para. 1); wherein the algorithm can reconstruct 3D density maps from user-defined positions in latent space (i.e. three-dimensional spatial position of the atom) enabling visualization of structural trajectories (i.e. a respective three-dimensional displacement vector for the atom relative to a three-dimensional spatial position of the atom in a base conformation of the macromolecule) (pg. 184 col. 1 para. 2); wherein structural trajectories inherently comprise the recited 3D displacement vectors based on spatial positions of atoms in said trajectories calculations (i.e. as evidenced by Paranjothy pg. 300 col. 1 para. 3). The prior art to Zhong will anticipate the instant claim limitations as described above as they would provide an equivalent product.
Conclusion
No claims are allowed.
Applicant's amendment necessitated the new ground(s) of rejection presented in this Office action. Accordingly, THIS ACTION IS MADE FINAL. See MPEP § 706.07(a). Applicant is reminded of the extension of time policy as set forth in 37 CFR 1.136(a).
A shortened statutory period for reply to this final action is set to expire THREE MONTHS from the mailing date of this action. In the event a first reply is filed within TWO MONTHS of the mailing date of this final action and the advisory action is not mailed until after the end of the 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 FRANCINI A FONSECA LOPEZ whose telephone number is (571)270-0899. The examiner can normally be reached Monday - Friday 8AM - 5PM 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.
/F.F.L./Examiner, Art Unit 1685
/OLIVIA M. WISE/Supervisory Patent Examiner, Art Unit 1685
Prosecution Insights