DETAILED ACTION
Notice of Pre-AIA or AIA Status
The present application, filed on or after March 16, 2013, is being examined under the first inventor to file provisions of the AIA .
Continued Examination Under 37 CFR 1.114
A request for continued examination under 37 CFR 1.114, including the fee set forth in
37 CFR 1.17(e), was filed in this application after final rejection. Since this application is eligible for continued examination under 37 CFR 1.114, and the fee set forth in 37 CFR 1.17(e) has been timely paid, the finality of the previous Office action has been withdrawn pursuant to 37 CFR 1.114. Applicant's submission filed on 12/01/2025 has been entered.
Claim Status
Claims 1-7, 9-15, and 17-20 are currently pending and under examination herein.
Claims 8 and 16 were canceled.
Information Disclosure Statement
The Information Disclosure Statements filed on 09/09/2025 is in compliance with the provisions of 37 CFR 1.97 and have been considered in full. A signed copy of the list of references cited from each IDS is included with this Office Action.
Withdrawn Rejections/Objections
Rejections and/or objections not reiterated from previous office actions are hereby
withdrawn in view of the amendments filed 12/01/2025.
The 35 U.S.C. 112(a) rejections to claims 1-7, 9-15, and 17-20 in the office action filed 06/17/2025 has been withdrawn in view of amendments filed 12/01/2025 specifically by clarifying that the “identifying a drug target” refers to identifying, for example, a protein or region of a protein and pointing to specific section of instant disclosure for further clarification.
The 35 U.S.C. 112(f) rejections in the office action filed 06/17/2025 has been withdrawn in view of amendments received 12/01/2025 specifically by acknowledging that data anomaly detection module, state detection module, and data separation modules are interpreted as software.
The following rejections and/or objections are either maintained or newly applied. They constitute the complete set presently being applied to the instant application.
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 1-7, 9-15, and 17-20 are rejected under 35 U.S.C. 101 because the claimed invention is directed to an abstract idea without significantly more.
In accordance with MPEP § 2106, claims found to recite statutory subject matter (Step 1: YES) are then analyzed to determine if the claims recite any concepts that equate to an abstract idea, law of nature or natural phenomenon (Step 2A, Prong 1). In the instant application, the claims recite the following limitations that equate to an abstract idea:
Claims 1, 11, and 19 recite determining a current layer of data from the input molecular
dynamics simulation data; separating abnormal data from the current layer of data; extracting a targeted state using the abnormal data; and separating targeted state data from the current layer of data using the targeted state; detecting abnormal data by calculating and ranking an absolute value of probabilistic z-scores of latent variables and ranking the probabilistic z-scores; Identifying a drug target based on the targeted state data (the limitation detecting abnormal data by calculating and ranking an absolute value of a score are considered mathematical calculation and mathematical relationship, and as such, falls within mathematical concept groupings of abstract ideas; the limitations determining a current layer of data, separating abnormal data, extracting a targeted state, separating targeted state data can be particularly performed in the human mind because the human mind is able to separate data); the limitation Identifying a drug given the plain meaning of “identifying” can be particularly performed in human mind (mental process) since human mind is capable of identifying/ detecting data based on the result of an analysis.
Claims 1, 11, and 19 further recites storing untargeted data identified, based on a user-defined threshold, as statistically irrelevant as targeted data for a next iteration; the limitation storing given the plain meaning of “storing” can be practically be performed in human mind, since human mind is capable of storing data, for example, by using pen and paper.
Claims 1, 11, and 19 further recite a first clustering finding targeted samples among abnormal samples separated from the current layer of data, the target samples exemplifying the targeted state (the limitation finding targeted samples among abnormal samples and separating from current layer of data the limitation sampling data can be particularly performed in the human mind because the human mind is able to find and separate data).
Claim 2 and 12 recite iteration through a plurality of layers of data, wherein at each iteration the method processes a next layer comprising untargeted data from a prior layer (the limitations iteration through a plurality of layers of data and processing the next layer from the prior layer can be particularly performed in the human mind because the human mind is able to iterate through data).
Claim 3 and 13 recite the input molecular dynamics simulation data is the current layer of data for a first iteration and the targeted state defines the current layer of data for a subsequent iteration (the limitations determining the simulation data as the current layer for a first iteration and targeted data for a subsequent iteration can be particularly performed in the human mind because the human mind is able to put data into different categories).
Claim 5 and 14 recite that the method ends upon determining that a ratio of untargeted data to total data is greater than a threshold (the limitation determining a ratio can be particularly performed in the human mind because the human mind is able to calculating ratios).
Claims 6 and 16 recite determining the current layer of data from the input molecular dynamics simulation data comprises sampling the input molecular dynamics simulation data to reduce a size of the current layer of data in a first iteration (the limitation sampling data can be particularly performed in the human mind because the human mind is able to sample data).
Claims 7 and 15 recite separating the abnormal data from the current layer of data by an autoencoder (the limitation separating data by autoencoder equates to carrying out mathematical calculations with mathematical functions, to transform and recreate the input data which falls under the “Mathematical concepts” grouping of abstract idea).
Claims 9 and 17 recite a second clustering separating the targeted state data from the current layer of data using the targeted state (the limitation separating the targeted state data from the current layer of data can be particularly performed in the human mind because the human mind is able to separate data based on other data).
Claims 10 and 18 require calculating the distance from the center of the cluster for the second clustering (the limitations distance measurements of the second clustering equate to carrying out mathematical calculations with mathematical functions, which falls under the “Mathematical concepts” grouping of abstract idea).
Therefore, the above limitations fall under the “Mental process” and “Mathematical concepts” groupings of abstract ideas. While claims 11-15 and 17-20 recites performing these steps with a processor and memory, there are no additional limitations that indicate that the processor and memory require anything other than carrying out the recited mental process or mathematical concept in a generic computer environment. Merely reciting that a mental process is being performed in a generic computer environment does not preclude the steps from being performed practically in the human mind or with pen and paper as claimed (MPEP 2106.04(a)(2)). If a claim limitation, under its broadest reasonable interpretation, covers performance of the limitation in the mind but for the recitation of generic computer components, then if falls within the “Mental processes” grouping of abstract ideas. As such, claims 1-7, 9-15, and 17-20 recite an abstract idea (Step 2A, Prong 1: YES).
Claims found to recite a judicial exception under Step 2A, Prong 1 are then further
analyzed to determine if the claims as a whole integrate the recited judicial exception into a
practical application or not (Step 2A, Prong 2). This judicial exception is not integrated into a
practical application because the claims do not recite an additional element that reflects an
improvement to technology or applies or uses the recited judicial exception in some other
meaningful way. Rather, the instant claims recite additional elements that amount to mere
instructions to implement the abstract idea in a generic computing environment or insignificant
extra-solution activity. Specifically, the claims recite the following additional elements:
Claims 1, 11, and 19 recites anomaly detection module, state detection module, and data separation module; using a machine learning model.
Claim 1 recites receiving input molecular dynamics simulation data.
Claim 4 recites that the method outputs the targeted state and the targeted state data from each iteration.
Claim 11 recites a non-transitory computer readable medium comprising computer executable instruction which when executed by a computer system cause the computer to perform the method for finding an unknown molecular dynamics state comprising: receiving input molecular dynamics simulation data.
Claim 14 recites outputting the targeted state and targeted state data.
Claim 19 recites a system comprising a communication interface, a processor, and a memory.
Claim 20 recites a display controlled by the processor.
There are no limitations that indicate that having a computer readable medium, communication interface, a processor, a memory, anomaly detection module, state detection module, and data separation module, or a display require anything other than generic computing systems. As such, these limitations equate to mere instructions to implement the abstract idea on a generic computer that the courts have stated does not integrate a recited judicial exception into a practical application in Alice Corp., 573 U.S. at 223, 110 USPQ2d at 1983. See also 573 U.S. at 224, 110 USPQ2d at 1984.
The limitations for receiving input molecular dynamics simulation data, outputting the targeted state and targeted state data, all equate to mere data gathering and storage activity because they serve merely to provide and store data that is analyzed by the judicial exception and these limitations do not integrate the recited exception into a practical application (see MPEP 2106.04(d)(2)). The courts have identified limitations that merely gather data or stores data as insignificant extra-solution activity that does not integrate the abstract idea into a practical application (see MPEP 2106.05(g)).
Furthermore, the limitation of using a machine learning model provide nothing more than mere instructions to implement an abstract idea on a generic computer. See MPEP 2106.05(f). Additionally, the recitation of “using a machine learning model” merely indicates a field of use or technological environment in which the judicial exception is performed. See MPEP 2106.05(h).
As such, claims 1-7, 9-15, and 17-20 are directed to an abstract idea (Step 2A, Prong 2: NO).
Claims found to be directed to a judicial exception are then further evaluated to determine if the claims recite an inventive concept that provides significantly more than the judicial
exception itself (Step 2B). An inventive concept cannot be furnished by an abstract idea itself (MPEP 2106.05). The claims do not include additional elements that are sufficient to amount to
significantly more than the judicial exception because the claims recite additional elements that
equate to mere instructions to apply the recited exception in a generic computing environment or well-understood, routine and conventional activity. The instant claims recite the following
additional elements:
Claims 1, 11, and 19 recites anomaly detection module, state detection module, and data separation module; using a machine learning model.
Claim 1 recites receiving input molecular dynamics simulation data.
Claim 4 recites that the method outputs the targeted state and the targeted state data from each iteration.
Claim 11 recites a non-transitory computer readable medium comprising computer executable instruction which when executed by a computer system cause the computer to perform the method for finding an unknown molecular dynamics state comprising: receiving input molecular dynamics simulation data.
Claim 14 recites outputting the targeted state and targeted state data.
Claim 19 recites a system comprising a communication interface, a processor, and a memory.
Claim 20 recites a display controlled by the processor.
As discussed above, there are no additional limitations to indicate that the claimed computer readable medium, computer, processor, memory, anomaly detection module, state detection module, and data separation module, or display require anything other than generic computer components in order to carry out the recited abstract idea in the claims. Claims that amount to nothing more than an instruction to apply the abstract idea using a generic computer do not render an abstract idea eligible. Alice Corp., 573 U.S. at 223, 110 USPQ2d at 1983. See also 573 U.S. at 224, 110 USPQ2d at 1984.
The limitations for receiving input molecular dynamics simulation data, outputting the targeted data and the targeted state data all equate to well-understood, routine and conventional activities. The courts have identified that receiving or transmitting data over a network are well-understood, routine and conventional computer functions in Symantec, 838 F.3d at 1321, 120 USPQ2d at 1362 (utilizing an intermediary computer to forward information); TLI Communications LLC v. AV Auto. LLC, 823 F.3d 607, 610, 118 USPQ2d 1744, 1745 (Fed. Cir. 2016) (using a telephone for image transmission); OIP Techs., Inc., v. Amazon.com, Inc., 788 F.3d 1359, 1363, 115 USPQ2d 1090, 1093 (Fed. Cir. 2015) (sending messages over a network); buySAFE, Inc. v. Google, Inc., 765 F.3d 1350, 1355, 112 USPQ2d 1093, 1096 (Fed. Cir. 2014); and Versata Dev. Group, Inc. v. SAP Am., Inc., 793 F.3d 1306, 1334, 115 USPQ2d 1681,
1701 (Fed. Cir. 2015).
Furthermore, the limitation of using a machine learning model provide nothing more than mere instructions to implement an abstract idea on a generic computer. See MPEP 2106.05 I. A. Additionally, the recitation of “using a machine learning model” merely indicates a field of use or technological environment in which the judicial exception is performed. See MPEP 2106.05(h).
The additional elements do not comprise an inventive concept when considered individually or as an ordered combination that transforms the claimed judicial exception into a patent-eligible application of the judicial exception. Therefore, the claims do not amount to significantly more than the judicial exception itself (Step 2B: No). As such, claims 1-7, 9-15, and 17-20 are not patent eligible.
Response to Applicant’s Arguments
Applicant's arguments filed 12/01/2025 have been fully considered but they are not persuasive. Applicant states:
The limitations of the independent claims are indicative of integration of the respective improvements into a practical application at least because the limitations are directed to improvements in the relevant technical field and because the limitations are patentable subject matter under the 2-part rubric of MPEP 2106.04(d)(1). The improvements are set forth in the specification, as outlined above (see paragraph [0008] of the original application). Furthermore, the independent claims include the components or steps of the invention that provide, for example, the improvements to the technological process of finding unknown molecular dynamics states and corresponding samples
In particular, Applicant notes that the claimed operation of "detecting abnormal data by calculating and ranking an absolute value of probabilistic z-scores of latent variables" is
implemented using machine learning. Such improvements to computerized machine learning are patent eligible under Section 101, as recently recognized by the PTAB in Ex parte Desjardins, Appeal 2024-000567, Decision on Request for Rehearing, BPAI September 26, 2025.
It is respectfully submitted that the above statements are not persuasive. The Applicant remarks are directed to Step 2A Prong Two of 101 analysis, specifically whether the additional elements integrate the recited judicial exception into a practical application of the exception.
With regards to Applicant pointing to specification [0008] for improvement in the relevant technical field, Examiner submits that iteratively finding one or more states by collecting and manipulating data are abstract ideas and therefore, not patent eligible. See MPEP 2106.04(a)(2) III. It is important to note, the judicial exception alone cannot provide the improvement (See MPEP 2106.04(d) III). The improvement must be provided by one or more additional elements. See the discussion of Diamond v. Diehr, 450 U.S. 175, 187 and 191-92, 209 USPQ 1, 10 (1981)) in subsection II, below. In addition, the improvement can be provided by the additional element(s) in combination with the recited judicial exception. See MPEP § 2106.04(d) (discussing Finjan, Inc. v. Blue Coat Sys., Inc., 879 F.3d 1299, 1303-04, 125 USPQ2d 1282, 1285-87 (Fed. Cir. 2018)).
With regards to Applicant using a machine learning model to detect abnormal data and referring to Desjardin stating “improvements to computerized machine learning are patent eligible”, Examiner submits that, as stated above, using a machine learning model provide nothing more than mere instructions to implement an abstract idea on a generic computer and merely indicates a field of use or technological environment in which the judicial exception is performed. Such limitations do not integrate judicial exception(s) into a practical application in Step 2A Prong 2 nor do they amount to significantly more in Step 2B. See MPEP 2106.05(f) and MPEP 2106.05 I.
Furthermore, with regards to applicant referring to Desjardin, Examiner submits that in Desjardin the improvement was to how the machine learning model itself operates (improvement to the machine learning architecture), and not, for example, mathematical calculations. In contrast, instant claims use machine learning to solve the problem of finding anomaly and identifying a state (judicial exception(s)) and merely indicates a field of use or technological environment in which the judicial exception is performed. Therefore, the recitation of “using a machine learning model” does not integrate the judicial exception(s) into a practical application and does not amount to significantly more.
Applicant further states:
the operation of "identifying a drug target based on the targeted state data" applies or uses the alleged judicial exception in a meaningful way.
It is respectfully submitted that this is not persuasive. The limitation of “identifying a drug”, given the plain meaning of “identifying”, encompasses observation, evaluation, judgment, and opinion (See MPEP 2106.04(a)(2), subsection III.) performable by human mind (mental process), since human mind is capable of identifying based on known information/result of an analysis.
Applicant further states:
Applicant directs the Examiner to paragraphs [0029]-[0030] of the original application. Paragraph [0029], for example, describes that the ADM can treat the entirety of the input molecular dynamics simulation data as the current layer of data, or can sample the input molecular dynamics simulation data to reduce a size of the data to be processed. The reduction in the size of the data improves the efficiency of performing the described method, reduces the time required to perform the described method and reduces the resources required to perform the described method. Paragraph [0030] mentions enabling processing of large-scale data, not previously possible, for the identification of unknown states. The cited reduction in the size of the data to be processed and the enablement of large-scale data are clearly improvements to the relevant existing technology.
It is respectfully submitted that the above statement is not persuasive. The Applicant remarks are directed to Step 2A Prong Two of 101 analysis, specifically whether the additional elements integrate the recited judicial exception into a practical application of the exception.
The reduction of size of data, as noted by the Applicant, is considered an abstract idea of reducing the size of data by, for example, sampling, clustering, filtering (see specification [0027] [0039]), therefore, not patent eligible.
In Step 2A Prong 2 of 101 analyses, claims are evaluated for additional elements that integrate the judicial exception into a practical application. The additional elements of a computer readable medium, communication interface, a processor, a memory, anomaly detection module, state detection module, and data separation module, or a display are generic computer components and/or processes and not integrate a judicial exception into a practical application. See MPEP 2106.05(f). Furthermore, the limitations of receiving data and outputting data amount to necessary data gathering and outputting, and as such, considered insignificant extra-solution activity, which do not integrate the judicial exception into a practical application. See MPEP 2106.05(g). Furthermore, the additional element of using a machine learning model provide nothing more than mere instructions to implement an abstract idea on a generic computer. See MPEP 2106.05 I. A. Additionally, the recitation of “using a machine learning model” merely indicates a field of use or technological environment in which the judicial exception is performed. See MPEP 2106.05(h).
Therefore, the instantly rejected claims are not drawn to eligible subject matter as they are directed to an abstract idea (and/or natural correlation) without significantly more. See MPEP § 2106.
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.
Claims 1-5, 6, 9, 11-14, 17, and 19-20 are rejected under 35 U.S.C. 103 as being unpatentable over Nakamura et al. (PUB No. US20170039268A1), in view of Harada et al. (Enhanced Conformational Sampling Method Based on Anomaly Detection Parallel Cascade Selection Molecular Dynamics: ad-PaCS MD, Journal of Chemical Theory and Computation, September 14, 2020, 16,6716−6725), in view of Likić et al. (A statistical approach to the interpretation of molecular dynamics simulations of calmodulin equilibrium dynamics, Protein Sci. 2005 Dec; 14(12): 2955–2963), and further in view of Xu et al. (US20140066321A1).
Regarding claims 1, 11, and 19:
“A method of finding an unknown molecular dynamics state comprising: receiving input molecular dynamics simulation data;”
Nakamura discloses that the storage unit 11 stores a plurality of structures of substances whose structure changes (substance structures 11a-1, 11a-2,...) and a dimension group 11b, which is a dimension group representing a plurality of dimensions of the structure of the substance as an index for structural analysis (i.e. disclosing molecular dynamic simulation data). Using OFLOOD, time series data of atomic coordinates or "trajectories" produced by MD simulations are clustered ([0036] FIG. 1).
Nakamura et al. discloses the storage that stores a plurality of structures of a substance/index dimensions (molecular dynamics simulation data) (p. 0036) that are atomic coordinated or trajectories which are generated by MD simulation and are to be used for clustering and further MD simulations (p. 0006). Also, Nakamura et al. discloses an index dimension extracting method that performs extraction for each of candidate dimensions, which are not included in a dimension group which is a group of index dimensions that are indices for structural analysis of a substance whose structure changes, out of a plurality of dimensions that express a structure of the substance (molecular dynamic simulation data) (page 7, claim 5).
“Determining a current layer of data from the input molecular dynamics simulation data;”
Nakamura et al. discloses out of the plurality of dimensions that express the structure of a substance; the computing unit sets (separates) a plurality of dimensions that are not included in the dimension group (current layer of data) as candidate dimensions, for example d3, d4, and d5 (abnormal data). (p. 0037). Nakamura further discloses program and an application program executed by the processor [0047].
Nakamura et al. further explains that in the d3, d4 and d5 dimensions are candidates for the hidden dimension (current layer of data) [0085].
“Detecting, abnormal data”
Nakamura discloses that the computing unit 12 performs analysis according to OFLOOD (outlier FLOODing: iterating outlier detection and resampling). When executing OFLOOD, the computing unit 12 carries out clustering for a plurality of structures in a multidimensional space that has the index dimensions included in the dimension group 11 b as coordinate axes and performs a MD simulation with an outlier structure that is not included in any of the clusters as the initial structure. Nakamura discloses that when the metastable structures are set as C1 and C2, clusters are generated for each metastable structure ([0084], FIG. 9)
“Separating, using an anomaly detection module, abnormal data from the current layer of data;”
Nakamura et al. discloses out of the plurality of dimensions that express the structure of a substance; the computing unit sets (separates) a plurality of dimensions that are not included in the dimension group (current layer of data) as candidate dimensions, for example d3, d4, and d5 (abnormal data). (p. 0037). Nakamura further discloses program and an application program executed by the processor [0047]. Nakamura further discloses clustering in a three-dimensional space having dimensions "dl, d2, and d3", performs clustering in a three-dimensional space having dimensions "dl, d2, and d4", and performs clustering in a three-dimensional space having dimensions "dl, d2, and d5" [0085].
“Extracting, using a state detection module, a targeted state using the abnormal data;
Nakamura et al. discloses that for each dimension in the plurality of candidate dimensions (abnormal data), the computing unit carries out clustering of substance structures that has all or the index dimensions included in the dimension group (normal data) and also the candidate dimension (abnormal data) as coordinate axes. (p. 0037). Then the computing unit determines a hidden dimension (targeted state) (p. 0038). Therefore, using abnormal data to extract a targeted state by means of clustering. Nakamura further discloses program and an application program executed by the processor [0047].
and separating, using a data separation module, targeted state data from the current layer of data using the targeted state.”
Nakamura et al. discloses that after determining the hidden dimension (targeted state) by the computing unit, it adds the hidden dimension (targeted state) to the dimension group 11 b (step S2) (p. 0038). Figure 1, also confirms the separation of hidden dimension d3 (targeted state) in step S2. Nakamura further discloses program and an application program executed by the processor [0047].
“Storing untargeted data identified, based on a user-defined threshold, as statistically irrelevant as targeted data for a next iteration”
Nakamura discloses that when executing OFLOOD, the computing unit 12 carries out clustering for a plurality of structures in a multidimensional space that has the index dimensions included in the dimension group 11 b as coordinate axes and performs a MD simulation with an outlier structure that is not included in any of the clusters as the initial structure.
Nakamura further discloses that when a MD simulation has been performed, the computing unit 12 stores the structure of the substance generated by the MD simulation in the storage unit 11. The computing unit 12 then repeatedly executes the processing in steps S1 to 33 above every time the structure of a substance is stored it the storage unit 11 [0040].
“And identifying a drug target based on the targeted state data” wherein the extraction of the targeted state further comprises a first clustering finding targeted samples among abnormal samples separated from the current layer of data, the target samples exemplifying the targeted state.
Nakamura et al. discloses using the present embodiments in drug design (p. 0106). Nakamura further discloses that the computing unit carries out clustering of substance structures that has all or the index dimensions included in the dimension group (normal data) and also the candidate dimension (abnormal data) as coordinate axes. (p. 0037). Then the computing unit determines a hidden dimension (targeted state) (p. 0038). Therefore, using abnormal data to extract a targeted state by means of clustering.
Further regarding claim 19:
“A system configured to perform an iterative method of finding unknown molecular dynamics states and corresponding samples, the system comprising: a communication interface configured to receive molecular dynamics data, the molecular dynamics data simulating movement of particles;”
Nakamura et al. discloses a connecting interface, which is a communication interface that allows peripheral devices such as memory device to be connected to the computer (“the system”) (p. 0053). And that the memory is configured to store structures of a substance (simulation data) whose structure changes (movement of particles) and a dimension group (p. 0020).
“A processor configured to determine a current layer of data from the molecular dynamics data, separate abnormal data from the current layer of data, extract a targeted state using the abnormal data, and separate targeted state data from the current layer of data using the targeted state extracted using the abnormal data;”
Nakamura et al. discloses a processor configured to perform a procedure including: performing, for each of a plurality of candidate dimensions, which are not included in the dimension group, out of the plurality of dimensions, clustering of the plurality of structures in a multidimensional space that has every index dimension included in the dimension group and a candidate dimension as coordinate axes; and adding a specified candidate dimension for which it is possible to generate, a largest number of clusters to the dimension group as an index dimension. (p. 0020).
“And a memory configured to store the targeted state and its data derived from the molecular dynamics data.”
Nakamura et al. discloses a memory configured to store a plurality of structures of a substance whose structure changes and a dimension group, which is a group of index dimensions that are indices for structural analysis of the substance, out of a plurality of dimensions that express a structure of the substance (p.0020). Figure. 1 and 3 disclose the storage unit 11 that stores structure of the substance/significant dimension information (simulation data) and hidden dimension information (targeted state).
Further regarding claim 11:
“A non-transitory computer readable medium
comprising computer executable instructions which when executed by a computer system cause the computer to perform the method for finding an unknown molecular dynamics state comprising”
Nakamura et al. discloses a non-transitory computer-readable storage medium storing a computer program, the computer program that causes a computer to perform a procedure comprising: performing, for each of a plurality of candidate dimensions, which are not included in a dimension group which is a group of index dimensions that are indices for structural analysis of a substance whose structure changes, out of a plurality of dimensions that express a structure of the substance, clustering of a plurality of structures of the substance in a multidimensional space that has every index dimension included in the dimension group and a candidate dimension as coordinate axes; and adding a specified candidate dimension for which it is possible to generate a largest number of clusters to the dimension group as an index dimension (page 7, claim 6).
Further regarding claims 1, 11, and 19:
Nakamura discloses storing, in the memory, structure of the substance generated during the structural analysis and repeatedly executing the clustering, the adding, and the structural analysis every time the structure is stored in the memory (claim 3 [0040]).
Harada discloses a rare-event sampling method called anomaly detection parallel cascade selection molecular dynamics (ad-PaCS-MD) using Machine learning. Harada further discloses rarely occurring but essential states (configurations) of proteins for the transitions are identified based on the degrees of an anomaly. In more detail, ad-PaCS-MD adopts an algorithm called an anomaly detection generative adversarial network (anoGAN) as a measure for detecting rarely occurring states to be resampled. Here, the essential configurations with higher degrees of the anomaly are selected with anoGAN and intensively resampled by restarting short-time MD simulations from the selected configurations. By repeating the detections, ranking, and resampling of configurations with the higher degrees of the anomaly, ad-PaCS-MD automatically and efficiently promotes the rare events (abstract). Harada further discloses mapping noise in latent space, defining a norm, first, a set of normal data of a given protein (reactant) is prepared, specifying a distance matrix, normal sampling by generating by learning the normal distance matrices. (see also, section: 2.2. ad-PaCS-MD, pg. 6719; Figure 1).
With regards to limitations of detecting abnormal data by calculating and ranking an absolute value of probabilistic z-scores of latent variables and ranking the probabilistic z-scores Nakamura discloses detecting abnormal data (when the metastable structures are set as C1 and C2, clusters are generated for each metastable structure ([0084], FIG. 9)).
Likić et al. analyses large amount of data generated by MD simulation using statistical methods to extract meaningful information about protein structure. Likić discloses use of analogs of z-score that uses median absolute deviation to detect outliers in Molecular Dynamic simulation data (p. 2961, col. 2, para. 1). Additionally, Xu discloses performing Z-testing, i.e., calculating an absolute Z score (see Example 1). Z-testing is typically utilized to identify significant differences between a sample mean and a population mean… a Z-score with an absolute value greater than 1.96 indicates non-randomness. For a 99% confidence interval, if the absolute Z is greater than 2.58, it means that p<0.01, and the difference is even more significant—the null hypothesis can be rejected with greater confidence (for example, threshold ranking). .. an absolute Z-score of 1.96, 2, 2.58, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20 or more, including all decimal points in between (e.g., 10.1, 10.6, 11.2, etc.), may provide a strong measure of statistical significance.
Regarding claims 2 and 12:
“The method iterates through a plurality of layers of data, wherein at each iteration the method processes a next layer comprising untargeted data from a prior layer.”
Nakamura et al. discloses that the computing unit repeatedly executes (iterates through) the processing in steps S1 to S3 [beginning to end of each iteration] every time the structure of a substance is stored it the storage unit (p 0040). Nakamura et al. further discloses that at the end of phase S2 the hidden dimension (targeted state) is determined and is added to the next dimension group [d1 and d2] (untargeted state) to be used in the next iteration (p. 0037). Therefore, each subsequent layer includes the next dimension group (untargeted data) from a prior layer.
Regarding claims 3 and 13:
“The input molecular dynamics simulation data is the current layer of data for a first iteration and the targeted state defines the current layer of data for a subsequent iteration.”
Nakamura et al. discloses that for the first iteration the computing unit uses the MD simulation stored in the storage unit (p.0036) and that after completing one iteration [step S1-S3] it includes the hidden dimension (targeted state) for subsequent iterations (p. 0037-0039).
Regarding claims 4, 5, and 14:
“The method outputs the targeted state and the targeted state data from each iteration and wherein the method ends upon determining that a ratio of untargeted data to total data is greater than a threshold”
Nakamura et al. discloses that The OFLOOD unit stores the trajectory generated by OFLOOD which includes hidden dimensions (targeted state) in the storage unit (p. 0103). Figure 1, S111 also shows that trajectories will be stores (outputted) at the end of each iteration. Nakamura further discloses that if the density of protein structures in the cell is at or above an upper limit, such cell is determined to be a dense cell. If the density of protein structures in a cell is below an upper limit but equal to or above a lower limit, such cell is determined to be a medium cell. If the density of the protein structures in a cell is below a lower limit, the cell is determined to be a sparse cell. When a next lower layer after an upper layer is generated, only the middle cells out of the cells on the (p. 0077).
Regarding claim 6:
“Determining the current layer of data from the input molecular dynamics simulation data comprises sampling the input molecular dynamics simulation data to reduce a size of the current layer of data in a first iteration.”
Nakamura et al. discloses the FlexDice clustering algorithm of OFLOOD sampling unit performs clustering on a trajectory outputted as a result of a MD simulation that has a protein structure determined by experimentations as an initial structure (p. 0093). Clustering comprises sampling of protein structures (MD simulation data) according to their density into different cells (sampling) (p. 0077). It is inherent that sampling data reduces the size of the data.
Regarding claims 9 and 17:
“Separating the targeted state data from the current layer of data comprises a second clustering, the second clustering separating the targeted state data from the current layer of data using the targeted state.”
Nakamura et al. discloses that only medium cells among the cells of the higher-ranked layer are each divided into two in each axis direction (divided into four in total) (second clustering) (p. 0074).
Nakamura et al. discloses a second clustering, which is performed by FlexDice of OFLOOD unit, where clustering is performed after adding the hidden dimension (targeted state) to dimension group 11b step S2. When executing OFLOOD, the computing unit carries out clustering (second clustering) for structures that has the index dimensions/hidden dimensions (targeted state) included in the dimension group (p. 0039).
Regarding claim 20:
“The system of Claim 19, further comprising a display controlled by the processor to display the targeted state data.”
Nakamura et al. discloses a monitor (display) that is connected to a graphical processing device connected to the processor to display the instructions from the processor (p. 0050) Nakamura et al. also discloses that the OFLOOD results are visualized by displaying, on the monitor (p. 0062).
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 have used a machine learning model, such as one disclosed by Harada, to perform the iterative method of anomaly detection of Nakamura (repeatedly executes the processing in steps S1 to 33 [0040], OFLOOD [0039]). Further it would have been obvious to have applied the known statistical technique of calculating and ranking an absolute value of probabilistic z-scores, as shown by Likić (p. 2961, col. 2, para. 1) and Xu (pg. 4, last para.), to the known method of Nakamura and Harada to detect abnormal data, because the absolute deviation performs well in detecting outliers in data offering a more efficient, automated, and more accurate anomaly detection. There would be a reasonable expectation of success in applying the technique of Likić and Xu to the method of Nakamura and Harada because they all use statistical methods to analyze anomaly in data.
Response to Applicant’s Arguments
Applicant's arguments filed 12/01/2025 have been fully considered but they are not persuasive. Claim amendments necessitated a new round of art rejections. As such, combination of Nakamura, Harada, Likić, and Xu discloses above-mentioned claims.
Claims 7 and 15 are rejected under 35 U.S.C 103 as being unpatentable over Nakamura et al. (PUB No. US20170039268A1), in view of Harada et al. (Enhanced Conformational Sampling Method Based on Anomaly Detection Parallel Cascade Selection Molecular Dynamics: ad-PaCS MD, Journal of Chemical Theory and Computation, September 14, 2020, 16,6716−6725), in view of Likić et al. (A statistical approach to the interpretation of molecular dynamics simulations of calmodulin equilibrium dynamics, Protein Sci. 2005 Dec; 14(12): 2955–2963), and further in view of Xu et al. (US20140066321A1), as applied to claims 1-5, 6, 9, 11-14, 17, and 19-20 above, and further in view of Han et al. (Research on ensemble model of anomaly detection based on autoencoder, 11 December 2020, Publisher: IEEE, Published in: 2020 IEEE 20th International Conference on Software Quality, Reliability and Security (QRS)).
The limitations of claims 1 and 11, from which claims 7 and 15 depend, have been taught above.
Regarding claims 7 and 15:
“The abnormal data is separated from the current layer of data by an autoencoder.”
Nakamura et al. discloses out of the plurality of dimensions that express the structure of a substance; the computing unit sets (separates) a plurality of dimensions that are not included in the dimension group (current layer of data) as candidate dimensions (abnormal data). (p. 0037).
Nakamura, Harada, Likić , and Xu do not expressly teach separating the abnormal data from the current layer of data by an autoencoder.
However, Han et al. teaches separating abnormal data from current layer of data using an autoencoder (Section D The Proposed Ensemble Autoencoder (EAE) for anomaly detection (figure 3) teaches an autoencoder that extracts/separates abnormal/anomaly data).
Therefore, it would have been prima facie obvious to someone of ordinary skill in the art before the effective filling date of the claimed invention to separate abnormal data from current layer of data using an autoencoder. Doing so would make the abnormal data extraction from the original input data more robust as taught by Han et al. The person of ordinary skill would have had a reasonable expectation of success in selecting this combination because it would have done anomaly detection/separation of abnormal data.
Response to Applicant’s Arguments
Applicant's arguments filed 12/01/2025 have been fully considered but they are not persuasive. Claim amendments necessitated a new round of art rejections. As such, combination of Nakamura, Harada, Likić, and Xu discloses above-mentioned claims.
Claims 10 and 18 are rejected under 35 U.S.C 103 as being unpatentable over Nakamura et al. (PUB No. US20170039268A1), in view of Harada et al. (Enhanced Conformational Sampling Method Based on Anomaly Detection Parallel Cascade Selection Molecular Dynamics: ad-PaCS MD, Journal of Chemical Theory and Computation, September 14, 2020, 16,6716−6725), in view of Likić et al. (A statistical approach to the interpretation of molecular dynamics simulations of calmodulin equilibrium dynamics, Protein Sci. 2005 Dec; 14(12): 2955–2963), and further in view of Xu et al. (US20140066321A1), as applied to claims 1-5, 6, 9, 11-14, 17, and 19-20 above, and further in view of Wu et al. (Identify High-Quality Protein Structural Models by Enhanced K-Means, March 22, 2017, Publisher: Biomed Research International).
The limitations of claims 1, 9 and 19, from which claims 10 and 18 depend, have been taught above.
Regarding claims 10 and 18:
“The second clustering uses a measure of distance from a center of a cluster of the current layer of data and a threshold for the measure of distance.”
Nakamura et al. teaches a clustering algorithm, FlexDice. Nakamura, Harada, Likić , and Xu do not expressly teach that second clustering uses a measure of distance from a center of a cluster of the current layer of data and a threshold for the measure of distance. However, Wu et al. teach a K-mean clustering algorithm which is distance-based algorithm that uses a measure of distance from a center of cluster of data and a threshold for the measure of the distance (section 2.2. Classical K-Means Algorithm and 3D Distance Metrix, subsection 2.2.1 Classical K-Means Algorithm).
Therefore, it would have been prima facie obvious to someone of ordinary skill in the art before the effective filling date of the claimed invention to use a measure of distance from a center of a cluster of the current layer of data and a threshold for the measure of distance for the second clustering. Doing so would result in high quality models and efficient processing of large data sets as taught by Wu et al. (Conclusion Section). The person of ordinary skill would have had a reasonable expectation of success in selecting this combination because either of the clustering methods could be used for clustering of biological data.
Response to Applicant’s Arguments
Applicant's arguments filed 12/01/2025 have been fully considered but they are not persuasive. Claim amendments necessitated a new round of art rejections. As such, combination of Nakamura, Harada, Likić, and Xu discloses above-mentioned claims.
Conclusion
No claims are allowed.
The prior art made of record and not relied upon is considered pertinent to applicant's disclosure.
Nielsen et al. (Quality and bias of protein disorder predictors, 26 March 2019, Scientific Reports | (2019) 9:5137 |, pages 1-11).
Nielson discloses a method of predicting disorder from protein sequences using molecular structure data (abstract). Neilson further discloses benchmarking the performance of disorder predictors using statistical methods such as calculating and ranking absolute value of probabilistic z-score (pg. 4, last para.; Figure 3). Nielson further discloses determining the Pearson correlation coefficient, RP, of agreement (see Fig. 3). This number is ideal for ranking the predictors from the best (largest absolute value) to the worst. As Z-scores increase with order while p is a measure of disorder, –1 indicates a perfect correlation and 0 expresses a complete lack of correlation.
Any inquiry concerning this communication or earlier communications from the examiner should be directed to GHAZAL SABOUR whose telephone number is (703)756-1289. The examiner can normally be reached M-F 7:30-5:00.
Examiner interviews are available via telephone, in-person, and video conferencing using a USPTO supplied web-based collaboration tool. To schedule an interview, applicant is encouraged to use the USPTO Automated Interview Request (AIR) at http://www.uspto.gov/interviewpractice.
If attempts to reach the examiner by telephone are unsuccessful, the examiner’s supervisor, Larry D. Riggs can be reached at (571) 270-3062. The fax phone number for the organization where this application or proceeding is assigned is 571-273-8300.
Information regarding the status of published or unpublished applications may be obtained from Patent Center. Unpublished application information in Patent Center is available to registered users. To file and manage patent submissions in Patent Center, visit: https://patentcenter.uspto.gov. Visit https://www.uspto.gov/patents/apply/patent-center for more information about Patent Center and https://www.uspto.gov/patents/docx for information about filing in DOCX format. For additional questions, contact the Electronic Business Center (EBC) at 866-217-9197 (toll-free). If you would like assistance from a USPTO Customer Service Representative, call 800-786-9199 (IN USA OR CANADA) or 571-272-1000.
/G.S./ Examiner, Art Unit 1686
/LARRY D RIGGS II/ Supervisory Patent Examiner, Art Unit 1686
Prosecution Insights