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 .
Status of Claims
Claims 1-10 are pending. All have been examined on the merits.
Information Disclosure Statement
The information disclosure statement filed 07/30/2021 is acknowledged. A signed copy of the corresponding 1449 form has been included with this Office action.
Priority
Priority As detailed on the 07/30/2021 filing receipt, this application claims priority to as early as 06/28/2020. At this point in examination, all claims have been interpreted as being accorded this priority date.
Claim Rejections - 35 USC § 112
The following is a quotation of 35 U.S.C. 112(b):
(b) CONCLUSION. —The specification shall conclude with one or more claims particularly pointing out and distinctly claiming the subject matter which the inventor or a joint inventor regards as the invention.
The following is a quotation of 35 U.S.C. 112 (pre-AIA ), second paragraph:
The specification shall conclude with one or more claims particularly pointing out and distinctly claiming the subject matter which the applicant regards as his invention.
Claim 1-10 rejected under 35 U.S.C. 112(b) or 35 U.S.C. 112 (pre-AIA ), second paragraph, as being indefinite for failing to particularly point out and distinctly claim the subject matter which the inventor or a joint inventor (or for applications subject to pre-AIA 35 U.S.C. 112, the applicant), regards as the invention.
Claim 1 recites the limitation “batch of candidate compounds that meet the requirements”, the limitation “requirements” lacks antecedent basis in the claim. It is unclear what “requirements” refer to as it is not previously introduced or defined in a manner that provides sufficient context of understanding.
Claim 1 recites “appropriate weights” the use of the relative term “appropriate” render the limitation vague and do not specify a concrete or reproducible method. The specification discloses “The user can assign appropriate weights to each important feature according to the important features of the drug's activity on the interface, and finally integrate it into a pharmacophore evaluation module” (P.0030). This non-limiting definition does not provide the metes and bounds of the limitation. Therefore, the limitation is indefinite.
Claim 5 recites “existing patent literature” in line 4. The specification discloses “Preferably, the filtering conditions include the number of heavy atoms of the compound,
the number of hydrogen bond donors, the number of hydrogen bond acceptors, scaffold structure, false positives, and compounds that have been reported in existing patent documents.” (P.0021). The claims and specifications do not define or list the compounds “existing patent literature.” The metes and bounds of the limitation are unclear rendering the claim indefinite.
Claim 7 recites “extracts feature that enhance the affinity of the drug to the protein” in Step A, lines 4-5. The metes and bounds of the limitation are unclear. One skilled in the art would not recognize what encompasses “enhance the affinity.” The specification discloses “extract the hydrogen bond interaction, hydrophobic interaction and other features that may enhance the affinity of the drug to the protein” (P0030). The use of may render the extracted affinity vague and do not specify a concrete or reproducible method.
Claim 7 recites the limitation “non-compliant compounds”. The metes and bounds of the limitation are unclear, a person of ordinary skill in the art would not recognize the non-compliant compounds. The specification discloses “then the user inputs filter conditions to eliminate non-compliant compounds, and the remaining compounds form a compound library” (P.0027). The claims and specifications provide This non-limiting definition; therefore, the limitation is indefinite.
Claim 7 recites “deletes unreasonable and repetitive compounds”, the limitation “unreasonable” is a relative term. A person of ordinary skill in the art would not recognize what are the unreasonable compounds. The specification discloses “input7 million sampling quantity parameters in the large-scale sampling subsystem, perform large-scale sampling of the Al model, produce more than 7 million compounds, delete unreasonable and repetitive compounds, and finally get more than 800,000 compounds” (P.0057). Considering the claims and specifications, the metes and bounds of the limitation are unclear, rendering the claim indefinite.
Claim 8 recites “suitable values”, the limitation suitable is a relative term. A person of ordinary skill in the art would not recognize what are the suitable values. The specification discloses “after the Al model training is completed, the Al model parameters are also optimized to suitable values.” (P.0034). The claim and specifications do not provide a definition for the limitation suitable. The metes and bounds of the limitation are unclear rendering the claim indefinite.
Claim 9 recites “screen out qualified compounds” in the penultimate line. The metes and bounds of the limitation “qualified” are unclear. One person in the art will not recognize the criteria based on which the qualified compounds are chosen unless specified. The specification discloses “screen out qualified compounds from the compound library according to the simulation results.” (P.0039). The claim and specifications do not provide a definition for the limitation qualified. The metes and bounds of the limitation are unclear rendering the claim indefinite.
Claim 9 recites “protein pdb file” and “pdb library”. The claim recites an acronym “pdb” without providing a definition or context for this term, rendering the scoop of the claim indefinite. (MPEP 2173.05). Any acronym used in claims must be spelled out before it.
Claim 9 recites “delete irrelevant ligands” in line 4-5. The limitation “irrelevant” is a relative term. The metes and bounds of the limitation are unclear. One skilled in the art would not recognize what encompasses “irrelevant ligands.” The specification discloses “delete irrelevant ligands, and define the pretreatment of the site that needs to be docked;” (P.036). the specifications and the claim do not provide a clear definition of the limitation rendering the claim indefinite.
Claim 9 recites “needs to be docked” in line 5. The metes and bounds of “needs” are unclear. One skilled in the art will not recognize what compounds will need to be docked. The specification discloses “delete irrelevant ligands, and define the pretreatment of the site that needs to be docked.” (P.0036) The claim and specification do not provide clear definition of the compounds that needs to be docked, rendering the claim indefinite.
Claim 9 recites “lowest energy” in line 8. The metes and bounds of the limitation “lowest” are unclear. A person of ordinary skill in the art will not recognize what are the values of the lowest energy needed to conform the compounds. The specification discloses “use the genetic algorithm to search for the conformation with the lowest energy of the compound” (P.0037). The claim and specification do not provide clear definition of the lowest emery needed, rendering the claim indefinite.
Claim 10 provides equations for arithmetic and geometric weighted average. The parameters of the equations or functions are not defined in neither the claims or the specification (P.0040). The metes and bounds of the claim are unclear rendering the claim indefinite.
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 1-6 are drawn to a” A virtual drug screening system” comprising of sub systems and instructions for modeling a system. The instant specification discloses “This application is described with reference to the method of embodiments of this invention and flowcharts and/or block diagrams of devices (systems), and computer program products. It should be understood that each process and/or block in the flowchart and/or block diagram, and the combination of processes and/or blocks in the flowchart and/or block diagram can be realized by computer program instructions.” (P.80). In view of the claims and specifications claims 1-6 are interpretated as using a software, reading on a signal. The computer program product/computer readable media is not limited to a physical embodiment and may read on carrier waves and other nonstatutory media. See, e.g., In re Nuitjen, Docket no. 2006-1371 (Fed. Cir. Sept. 20, 2007) (slip. op. at 18) (“A transitory, propagating signal like Nuitjen' s is not a process, machine, manufacture, or composition of matter.' … Thus, such a signal cannot be patentable subject matter.”).
Claims 1-10 rejected under 35 U.S.C. 101 because the claimed invention is directed to a judicial exception (i.e., a law of nature, a natural phenomenon, or an abstract idea) without significantly more.
The claims are found herein to recite abstract ideas to fall into the grouping of mathematical concepts.
Framework Analysis as pertains to the Instant Claims 1-10:
Step 1: Are the claims directed to a category of statutory subject matter (a process, machine, manufacture, or composition of matter? (See MPEP § 2106.03)
Claims 1-6 are properly directed to a 101 statutory category, specifically: a “System”.
Claims 7-10 are properly directed to a 101 statutory category, specifically: a “Method”.
[Step 1: claims 1-10: YES]
Step 2A, Prong One: Do the claims recite a judicially recognized exception, i.e., an abstract idea, a law of nature, or a natural phenomenon? (See MPEP § 2106.04(a))
Claims 1-10 recite abstract ideas. The MPEP at 2106.04(a)(2) further explains that abstract ideas are defined as:
• mathematical concepts, (mathematical formulas or equations, mathematical
relationships and mathematical calculations);
• certain methods of organizing human activity (fundamental economic practices
or principles, managing personal behavior or relationships or interactions between people); and/or
• mental processes (procedures for observing, evaluating, analyzing/ judging and
organizing information).
The MPEP at 2106.04(b) defines natural law/ natural phenomena as:
• naturally occurring principles/ relations that are naturally occurring or that do not have
markedly different characteristics compared to what occurs in nature.
With respect to the instant claims, under Step 2A, Prong One evaluation, the claims are found to recite abstract ideas that fall into the grouping of mathematical concepts and mental processes.
Regarding claim 1
The visualization subsystem is used to view the binding position of a ligand of a protein in the crystal complex, analyze a binding mode of the ligand and the protein, and extract features that enhance the affinity of the drug to the protein. Under the BRI, the extracted features are hydrogen bonding and/or hydrophobic interaction. (0017)
Analyzing the ligand and protein is a mental process and is an abstract idea
Extracting features is a mental process and is an abstract idea.
The evaluation tool box subsystem encapsulates a plurality of compound evaluation modules, and is used to design an evaluation function by selecting the plurality of compound evaluation modules and assigning appropriate weights; Under the BRI, the evaluation function id defined by 2 equations representing the aromatic and geometric means. (0040)
Designing an evaluation function and assigning weights is a mental process and is an abstract idea.
Using the evaluation functions is a calculation and is an abstract idea.
Regarding claim 4
The Al model training, and the update of the Al model parameter; wherein the Al model is a neural network system for generating the compounds; wherein the Al model parameter is the parameter of the neural network system; and the Al model itself can generate the compounds randomly.
Training an Ai model to generate compounds is a mathematical concept of a calculation, and is an abstract idea.
Updating the model parameters is a mental process and is an abstract idea.
Regarding claim 7
Define binding characteristics of the ligand in the crystal complex through an analysis of the visualization subsystem,
Defining the binding site is a mental process and is an abstract idea.
Analyzes the binding mode of the ligand and the protein, and extracts the features that enhance the affinity of the drug to the protein;
Analyzing and extracting features are mental processes and are abstract ideas.
Combine the visualization subsystem with the evaluation tool box subsystem to form a complete evaluation pipeline;
Forming data by combining information and evaluating it, is a mental process and is an abstract idea.
Start the Al model through the Al model management subsystem and start training.
Using an AI model is a mathematical concept of a calculation and is an abstract idea.
Samples the trained Al model, generates a specified number of compounds, deletes unreasonable and repetitive compounds.
Using the AI model to generate compounds or delete compounds is a mathematical concept of a calculation and is an abstract idea.
The user inputs filter conditions to eliminate non- compliant compounds, and the remaining compounds form a compound library.
Filtering data using filter conditions is a mental process and is an abstract idea.
Each compound evaluation module in the evaluation tool box subsystem will output a score, which is then integrated into a comprehensive score through the evaluation function;
Under the BRI, the evaluation function is calculating the arithmetic or geometric mean. (0018,0040)
To generate a score using an equation is a mathematical concept and is an abstract idea.
Regarding claim 8
The Al model outputs the compounds generated by the Al model to the evaluation pipeline through interaction, and collects scores of the compounds output by the evaluation pipeline, the Al model parameters are automatically updated. The Al model parameters are also optimized to suitable values.
Updating and optimizing the parameters of an AI model is a mental process and is an abstract idea.
Regarding claim 9
Delete water molecules, hydrogenate, delete irrelevant ligands, and define the pretreatment of a site that needs to be docked; conformation optimization: carry out a conformation optimization operation for the compounds, after generating a 3D conformation of the compound.
Deleting water molecules, irrelevant ligands, pretreatment of a site and optimizing operations are mental processes and are abstract ideas.
Use a genetic algorithm to search for the conformation of the compounds in the lowest energy.
The use of an algorithm is a mathematical concept of a calculation and is an abstract idea.
[Step 2A, Prong One, abstract idea: claims 1-10: 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? (See MPEP § 2106.04(d))
A claim can be said to integrate a judicial exception into a practical application when it applies, relies on, or uses the judicial exception in a manner that imposes a meaningful limit on the judicial exception.
Regarding Claims 1-6 and 7-8: Additional elements have been added to the claims. Yet the claims fail to integrate the judicial exception into a practical application [see MPEP § 2106.04(d)(III)].
Regarding claims 1-6 additional element a “System”
The system does not apply the mathematical and mental processes into any practical application, rather it is simply a system for carrying out the mathematical processes, as disclosed in the specification “These computer program instructions can also be loaded on a computer or other programmable data processing equipment, so that a series of operation steps are executed on the computer” (0082, page12).
The use of a computer or other machinery in its ordinary capacity for economic or other tasks (e.g., to receive, store, or transmit data) or simply adding a general-purpose computer or computer components after the fact to an abstract idea (e.g., a fundamental economic practice or mathematical equation) does not integrate a judicial exception into a practical application or provide significantly more.
Regarding claims 7-8 additional element “Input” and ”output”.
Processing and manipulating data, as an input and output, are not considered
abstract ideas, but perform functions of inputting, collecting, and outputting the data needed to
carry out the abstract idea. These steps are considered insignificant extra-solution activity, and
are not sufficient to integrate an abstract idea into a practical application as they do not impose
any meaningful limitation on the abstract idea or how it is performed. To integrate a judicial
exception into a practical application, the additional limitations must not be mere instructions to
apply the judicial exception [see MPEP § 2106.04(d) and MPEP § 2106.05(g)].
[Step 2A, 2nd prong: claims 1-6, 7 and 8: NO]
Step 2B: If the claims do not integrate the judicial exception, do the claims provide an inventive concept? (See MPEP § 2106.05)
Step 2B, which evaluates whether the additional elements, individually and in combination, amount to significantly more than the judicial exception itself by providing an inventive concept. An inventive concept is furnished by an element or combination of elements that is recited in the claim in addition to the judicial exception, and is sufficient to ensure that the claim, as a whole, amounts to significantly more than the judicial exception itself (see MPEP § 2106.05).
Explaining the Berkheimer v. HP, Inc., 881 F.3d 1360, 125 USPQ2d 1649 (Fed. Cir. 2018) as an example of using the computer as a tool to perform a mental process. The MPEP stated “The patentee claimed methods for parsing and evaluating data using a computer processing system. The Federal Circuit determined that these claims were directed to mental processes of parsing and comparing data, because the steps were recited at a high level of generality and merely used computers as a tool to perform the processes. 881 F.3d at 1366, 125 USPQ2d at 1652-53”
As indicated in the summary of the Berkheimer v. HP ruling above and in view of the specifications, the claims do not include additional elements that are sufficient to amount to significantly more than the judicial exceptions because the claims recite additional elements that are generic, conventional, or nonspecific. Those additional elements are as follows:
Claims 7 and 8 additional element is “Input” and “output” of data, the additional elements of data processing and manipulation, do not cause the claims to rise to the level of significantly more than the judicial exception. The courts have recognized receiving or transmitting data over a network; storing and retrieving information in memory and virtual screening of molecules or crystal complexes [see MPEP§2106.05(d)(II)], as well-understood, routine, conventional activity when they are claimed in a merely generic manner. Durrant in 2016 used neural networks to identify estrogen receptor level using a library of molecules.
Therefore, the claims, when the limitations are considered individually and as a whole, are rejected under 35 U.S.C. § 101 as being directed to non-statutory subject matter
Claim Rejections - 35 USC § 103
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, 4, 7 and 8 are rejected under 35 U.S.C. 103(a) as being unpatentable over Durrant (PMC: 2016 Mar 7), in view of Sander (Journal of Chemical Information and Modeling: 2015 Jan 5) and Castilo (Curr Neuropharmacol: 2017 NOV 15) as evidence by Alto (TDS Archive: 2019 Jul 5).
Regarding the recited in claims 1
Regarding claim 1,
Regarding the use of a visualization subsystem.
Regarding the recited in claims 1, the visualization subsystem is used to view the binding position of a ligand of a protein in the crystal complex, analyze a binding mode of the ligand and the protein, and extract features that enhance the affinity of the drug to the protein.
Durrant displayed graphical representation of the crystal complex of compound and targets, which reads on the use of visualization system, and displaying different crystallographic binding poses, represented hydrogen bonds as black lines in figure 4, which showed the binding characteristics of the crystal complex.
Regarding the use of an evaluation subsystem
Regarding the recited in claims 1, the evaluation tool box subsystem encapsulates a plurality of compound evaluation modules, and is used to design an evaluation function by selecting the plurality of compound evaluation modules and assigning appropriate weights.
Durrant teaches an evaluation step introduced after docking, “Structure-based virtual screening is a two-step process in which a molecule is first docked (i.e., positioned) into a receptor pocket and then evaluated using a scoring function to predict activity. “ (Introduction, third paragraph).
Durrant did not assign weights.
Castillo teaches assigning weights to the docking scores “In addition to the raw docking scores, the scoring value of each compound was weighted by the number of heavy atoms on it.” (2.5. Molecular Docking Post-processing, Lines 8-9)
Regarding the use of an AI system
Regarding the recited in claim 1, Al model is a neural network system for generating compounds
Durrant used and trained neural network model in more than one virtual screen. Durrant trained neural networks to predict small-molecule/receptor binding by first generating numeric “descriptors” of thousands of crystallographic binding poses. (2nd paragraph, Back ground Neural network)
Durrant teaches “In six of these virtual screens, over ~75% of the known ligands were contained in the set of top-ranking compounds large enough to include 5% of the presumed decoys” (8th paragraph, Results, and discussion section)
Regarding the recited in claim 1, the AI model parameter is a parameter of the neural network,
Durrat teaches adjusting the connection strengths of the neural network which is adjusting the weights between neurons. Durrant teaches “Neural networks are trained by gradually adjusting the connection strengths until the networks can reliably predict the correct output from a given input. “(Paragraph 1, Background: Neural Networks section, lines 6-8)
Regarding the use of large-scale sampling system
Regarding the recited in claims 1 the large-scale sampling subsystem is used to sample and screen the trained Al model to obtain a compound library composed of the corresponding compounds;
The large-scale sampling subsystem is used to sample and screen the trained Almodel to obtain a compound library composed of the corresponding compounds. Under the BRI “Large scale sampling is interpreted as “accepts a sampling quantity parameter input by the user” (0027, Page 4). Durrant teaches the use of small molecule library which is a large-scale sample. “In the previous retrospective virtual-screening study, we used a small-molecule library consisting of known ERα ligands and presumed decoys (e.g., molecules presumed to be non-binders for testing purposes” (2nd paragraph, Background: NNScore Performance against the Estrogen Receptor section)
Regarding the use of the virtual screening system.
Regarding the recited in claim 1, the virtual screening subsystem is used for further screening of the compounds in the compound library.
The use of the limitation “library” is interpreted as evaluating large collection of compounds. Durrant teaches the use of a small-molecule library, which is equivalent to a large collection of compounds.
Durrant teaches “In the previous retrospective virtual-screening study, we used a small-molecule library consisting of known ERα ligands and presumed decoys (e.g., molecules presumed to be non-binders for testing purposes, though without experimental conformation)” (2nd paragraph, Background: NNScore Performance against the Estrogen Receptor section)
Regarding the use of data log storage
Regarding the recited in claims 1, the data log storage subsystem is used to establish and store a user's log information file; the log information file is used to record user operations and generate corresponding data.
To be able to record and save actions for each user, inherently means the use of a user account and log information. Durrant does not teach a user interface, Sander teaches a user interface from which the user can record, modify, execute, and save action sequences as editable and reusable macros.
Sander teaches “It creates the menu and has implementations for all actions that can be launched from the main menu or the popup menus of the DataWarrior user interface components. It provides dialogues for more complex actions, which need to be configured before being started and it contains some of the algorithms needed to execute these actions. It also contains the functionality to record, modify, execute, and save action sequences as editable and reusable macros.” (Architecture, DataWarrior sections)
Regarding claim 4
Al model management subsystem is used for Al model, Al model training, and update of Al model parameter;
Regarding updating the AI parameters
A neural network constantly updates its parameters, specifically the weights and biases as evidence by
Alto “Then, depending on how accurate our predictions are, the algorithm updates itself through the so-called ‘backpropagation’ phase, according to a given optimization strategy.” (Lines 6-8). Durrant teaches “Neural networks are trained by gradually adjusting the connection strengths until the networks can reliably predict the correct output from a given input.” “(Paragraph 1, Background: Neural Networks section, lines 6-8)
Regarding the recited in claim 7
Regarding the recited in step 7A, step A: define binding characteristics of the ligand in the crystal complex through an analysis of the visualization subsystem, wherein the user downloads a target of the crystal complex structure from a protein crystal structure database.
Durrant displayed graphical representation of the crystal complex of compound and targets and displaying different crystallographic binding poses, represented hydrogen bonds as black lines in figure 4, which showed the binding characteristics of the crystal complex. Durrent teaches in figure 1, the protocol used in is study, which involves the use of diverse targets which suggests downloading different targets from a source.
Regarding the recited in claim 7, step A, visualizing and analyzing a binding position of the ligand and protein, defining binding characteristics. Considering the claims and specification the BRI of the limitation “binding characteristics” is interpreted as the binding characteristics of the crystal complexes that have been reported in the relevant literature, or the binding characteristics of the ligands that have been reported in the literature and the analysis of the visualization subsystem. (0031, page 5)
Durrant teaches the binding characteristics through the use of BINANA and PoseView algorithms, that can identify and display graphically potential receptor-ligand interactions between all the crystallographic poses. (Binding poses section), Durrant identifies the three highest affinities novel ERα ligands and analyzed them showing similar docked poses, the back bone interactions and the hydrogen bonds formed shown in figure 3 and 4.
Regarding the recited in claim 7 step B, the evaluation tool box will output a score, which is then integrated into a comprehensive score through the evaluation function.
Durrant evaluated the structure using a NNScore scoring function (paragraph 1, Background: Neural Networks section) which is based on artificial neural networks. Durrant did not calculate a final or comprehensive score. Castillo generated a score for each confirmation, based on which rescoring was performed. Castillo teaches “For every compound the best scored conformation was selected for further rescoring using six scoring functions implemented in DOCK.” ( 2.5. Molecular Docking Post-processing, Line 2-3)
Regarding the recited in claims 7 step C, combine the visualization subsystem with the evaluation tool box subsystem to form a complete evaluation pipeline and start AI model.
Durrant visualized structures as shown in the (Graphical abstract) and also shown in figure3. and Figure4. Thus, teaching a visualization system. In addition, Durrant evaluated and ranked the compounds “When a list of docked compounds is ordered by this score, the set of top-ranked molecules is often enriched for true ligands.” (Background: Neural Networks, 4th paragraph)
Under the BRI, the combination of the visualization and evaluation sub systems to start the AI modeling, is combining the information generated from these subsystems to start the AI model. Which reads on generating training features or descriptors to start AI modeling. Durrant generated numeric descriptors for thousands of compounds. Durrant teaches “In previous studies, we trained neural networks to predict small-molecule/receptor binding by first generating numeric “descriptors” of thousands of crystallographic binding poses.” (2nd paragraph, Background: Neural Networks section)
Regarding the recited in 7 step C, start the Al model through the Al model management subsystem and start training. Durrant trained neural networks to predict the strength of binding using the descriptors. Durrant teaches “Neural networks were trained to predict the strength of binding from these descriptors by fitting against experimentally measured binding affinities. Specifically, NNScore 1.0 was trained to categorize ligands by potency (high-affinity vs. low-affinity binder).” (3nd paragraph, Background: Neural Networks section)
Regarding the recited in claim 7 step D, samples the trained Al model, generates a specified number of compounds, deletes unreasonable and repetitive compounds, and then the user inputs filter conditions to eliminate non- compliant compounds, and the remaining compounds form a compound library.
Durrant generated a list of compounds “List of docked compounds is ordered by this score; the set of top-ranked molecules is often enriched for true ligands.” (4th paragraph, Background: Neural Networks section), which reads on generates a specified number of compounds.
Durrant used a filtering process and ranked the compounds, which reads on then the user inputs filter conditions to eliminate non- compliant compounds) and Durrant also teaches detect problematic compounds, which reads on delete un reasonable compounds and repeated ones. Durrant teaches “This filtering process had a substantial impact on three of the six high-performing ERα virtual screens. Of the top-ranked compounds from these screens, 14/15 or 15/15 were judged problematic” (3rd paragraph, Ignoring Potentially Promiscuous Top-Ranked Compounds section).
In addition, Sander teaches a compound table form which compounds can be modified or deleted, Sander teaches “The table model allows rows to be added, changed, or deleted.” Compound Table section, line 3).
Regarding the recited in claim 7 step D, user inputs filter conditions, Durant does not teach an interactive user interface. Durrant does not teach user interface. Sander teaches a user interface.
Sander teaches a compound table form which compounds can be modified or deleted, Sander teaches “The table model allows rows to be added, changed, or deleted.” Compound Table section, line 3).
Regarding the recited in claim 7 step E, the virtual screening subsystem further screens the compounds in the compound library.
Durrant teaches the use of a small-molecule library. Durrant teaches “In the previous retrospective virtual-screening study, we used a small-molecule library consisting of known ERα ligands and presumed decoys (e.g., molecules presumed to be non-binders for testing purposes, though without experimental conformation)” (2nd paragraph, Background: NNScore Performance against the Estrogen Receptor section).
Regarding the recited in claim 7 F, the data log storage subsystem creates and stores the user's log information file when the user uses the subsystem to design drugs.
To be able to record and save actions for each user, inherently means the use of a user account and log information. Durrant does not teach a user interface, Sander teaches a user interface from which the user can record, modify, execute, and save action sequences as editable and reusable macros.
Sander teaches “It creates the menu and has implementations for all actions that can be launched from the main menu or the popup menus of the DataWarrior user interface components. It provides dialogues for more complex actions, which need to be configured before being started and it contains some of the algorithms needed to execute these actions. It also contains the functionality to record, modify, execute, and save action sequences as editable and reusable macros.” (Architecture, DataWarrior sections)
Regarding claim 8
Regarding the recited in claim 8, Al model outputs the compounds generated by the Al model to the evaluation pipeline through interaction, and collects scores of the compounds output by the evaluation pipeline.
Durrant teaches a list of compounds is generated, “When a list of docked compounds is ordered by this score, the set of top-ranked molecules is often enriched for true ligands.” (4th paragraph, Background: Neural Networks) which reads on the model out puts compounds.
Durrant teaches a scoring function based on which compounds are ranked, which reads on collecting scores. Durrant teaches “Structure-based virtual screening is a two-step process in which a molecule is first docked (i.e., positioned) into a receptor pocket and then evaluated using a scoring function to predict activity. Reliable scoring functions are required to effectively enrich a set of top-predicted binders with potential hits” (3rd paragraph, Introduction section).
Regarding the recited in claim 8, the Al model parameters are automatically updated; after repeating the process for a number of time, the compounds generated by the Al model will get a higher score in the evaluation pipeline; after the Al model training is completed, the Al model parameters are also optimized to suitable values.
Durrant trained neural network, Durrant teaches “Neural networks were trained to predict the strength of binding from these descriptors by fitting against experimentally measured binding affinities.” (3rd paragraph, Background: Neural Networks section), which reads on Al model training is completed.
A neural network constantly updates its parameters until the network can reliably predict the correct output, specifically the weights and biases as evidence by Alto “Then, depending on how accurate our predictions are, the algorithm updates itself through the so-called ‘backpropagation’ phase, according to a given optimization strategy.” (Lines 6-8). Durrant teaches “Neural networks are trained by gradually adjusting the connection strengths until the networks can reliably predict the correct output from a given input.” “(Paragraph 1, Background: Neural Networks section, lines 6-8), which reads on, the Al model parameters are also optimized to suitable values.
Regarding the recited in claim 8, repeating the process for a number of time, the compounds generated by the Al model will get a higher score. Durrant implied iterative or repetitive training until the model achieves reliable predictions with guided optimized parameters. Durrant teaches “Neural networks are trained by gradually adjusting the connection strengths until the networks can reliably predict the correct output from a given input.” (1rst paragraph, Background: Neural Networks, Lines 6-7).
A person of ordinary skill of the art would find it obvious to combine Durrant, catillo and Sander to create a more functional and user-friendly platforms allowing controlled data management, more accurate predictions, and workflow optimization. There is a likelihood of success since the combination is a logical and expected improvement to enhance usability and functionality.
Claim 2 is rejected under 35 U.S.C. 103(a) as being unpatentable over Durrant (PMC: 2016), as applied to claims 1, 4, 7 and 8 above, in view of Laskowski (Journal of Chemical Information and Modeling: 2011 Sept 15).
Durrant is applied to claims 1, 4, 7 and 8 above.
Regarding claim 2
Regarding the recited in claim 2, the features that enhance the affinity of the drug to the protein is hydrogen bonding and/or hydrophobic interaction.
Durrant displayed graphical representation of the crystal complex of compound and targets and displaying different crystallographic binding poses, represented hydrogen bonds as black lines in figure 4. Durrant did not display hydrophobic interactions.
Laskowski teaches a graphical system that can generate multiple 2D diagrams. The diagrams can portray hydrogen and hydrophobic interactions.
Laskowski teaches “We describe a graphical system for automatically generating multiple 2D diagrams of ligand–protein interactions from 3D coordinates. The diagrams portray the hydrogen-bond interaction patterns and hydrophobic contacts between the ligand(s) and the main-chain or side-chain elements of the protein.” (Abstract)
It would have been obvious to a person ordinary skill in the art to combine Laskowski teaches and Durrant, to add an interactive structure analysis tool to enhance interpretability and refinement of ligand binding predictions. There is a likelihood of success since the combination would improve workflow efficiency and aid in molecular optimization. Both teachings are well known in the art and have been used in different research before the filling date of the application.
Claims 3 and 10 are rejected under 35 U.S.C. 103(a) as being unpatentable over Durrant (PMC: 2016), as applied to claims 1, 4, 7 and 8 above, in view of Castillo (Curr Neuropharmacol. 2017).
Durrant is applied to claims 1, 4, 7 and 8 above.
Durrant is applied to claim 2 above.
Regarding claim 3
Regarding the recited in claim 3, the evaluation function is a weighted arithmetic mean, a weighted geometric mean, or a user-defined function.
Durrant did not teach evaluation functions to be weighted arithmetic mean, a weighted geometric mean. Castillo used a combination of scoring functions; a fused rank was computed as either the arithmetic or geometric mean.
Castillo teaches “Then, for a specific combination of scoring functions, a fused rank was computed as either the arithmetic or geometric mean of the compound’s rank in the individual models.” (2nd paragraph, 2.5. Molecular Docking Post-processing)
It would have been obvious to a person of ordinary skill in the art to combine Durrant and Castillo to improve virtual screening accuracy to enhance ligand selection. There is a likelihood of success since both methods are well established in computational drug discovery and complement each other in optimizing ligand identification and evaluation.
Regarding the recited in claim 10
Regarding the recited, characterized in that, in the evaluation function, a
weight is set for each of the score: w1, w2, W3, ...... Wn to form the evaluation function, and the
evaluation function is an arithmetic weighted average or a geometric weighted average represented by 2 equations.
Durrant did not use arithmetic weighted average or a geometric weighted average score.
Castillo used a combination of scoring functions; a fused rank was computed as either the arithmetic or geometric mean and described the steps of calculating each, which reads on the equations provided in the claim.
Castillo teaches “As mentioned before, we tested the arithmetic and geometric means as fusion operators. FS1arithcontains three aggregation steps: the aggregation of A 2A AR scoring functions, the aggregation of MAO-B scoring functions and the aggregation of the rankings obtained for both targets. In this case all possible combinations of both fusion operators were tested. That is, scoring functions were first aggregated using the same fusion operator, either arithmetic or geometric mean, for each target separately.” (4th paragraph, Page 4)
It would have been obvious to a person of ordinary skill in the art to combine Durrant and Castillo to improve virtual screening accuracy to enhance ligand selection. There is a likelihood of success since both methods are well established in computational drug discovery and complement each other in optimizing ligand identification and evaluation.
Claim 5 is rejected under 35 U.S.C. 103(a) as being unpatentable over Durrant (PMC: 2016 Mar 7), as applied to claims 1, 4, 7 and 8 above, in view of Salentin (Nucleic Acids Res: 2015 Apr 14) as evidence by Pike (pubmed 2016, title an overview of heavy-atom derivatization of protein crystals)
Durrant is applied to claims 1, 4, 7 and 8 above.
Durrant is applied to claim 2 above.
Durrant is applied to claims 3 and 10 above.
Regarding the recited in claim 5, filter condition of the screening includes a number of heavy atoms of the compound, a number of hydrogen bond donors, a number of hydrogen bond acceptors, scaffold structure, false positives, and the compounds that have been reported in existing patent literature.
Durrant Identified false positives in the neural network predictions. Durrant teaches “The six high-performing ERα virtual screens described above identified a number of potentially promiscuous and/or false-positive compounds.” (2nd paragraph, Ignoring Potentially Promiscuous Top-Ranked Compounds section)
Durrant teaches classifying compounds based on scaffolds, Durrant teaches “Compound diversity and uniqueness can be assessed by classifying compounds according to molecular scaffolds (e.g., molecular graphs)”( Comparison of Docking Methods, 4th paragraph), Classifying reads on applying a filter.
Durrant did not teach filter conditions to be number of heavy atoms of the compound, a number of hydrogen bond donors, a number of hydrogen bond acceptors, scaffold structure and did not teach filter conditions can be entered by the user.
Reactant atoms are considered "heavy atoms" when they include atoms with a relatively high atomic number such as carbon, nitrogen, oxygen, or Halogen bonds, as evidence by (Pike: pubmed 2016, title an overview of heavy-atom derivatization of protein crystals)
Salentin teaches PLIP: fully automated protein–ligand interaction profiler, Salentin is able to detect halogen bonds which are heavy atoms, hydrogen bonds, hydrophobic contacts, and detection of hydrophobic atoms as well as acceptors/donors for hydrogen and halogen bonds.
Salentin teaches “It returns a list of detected interactions on single atom level, covering seven interaction types (hydrogen bonds, hydrophobic contacts, pi-stacking, pi-cation interactions, salt bridges, water bridges and halogen bonds).” (Abstract)
Salentin teaches “In order to find interacting groups (Figure3B), the binding partners need to be functionally characterized first. This includes detection of hydrophobic atoms as well as acceptors/donors for hydrogen and halogen bonds. Furthermore, PLIP searches for aromatic rings and charge centers in protein and ligand. The latter functionalities are a precondition for formation of π-stacking, π-cation interactions, or salt bridges. In the case of DegV with palmitic acid, charges can be assigned to two amino acids as well as the ligand carboxyl group” (Page 9, Lines 3-9)
It would have been obvious to a person of ordinary skill in the art to combine the teaching of Durrant and Salentin, these tools would improve the ligand selection by adding a filtering mechanism that will ensure an accurate prediction. There is a likelihood of success since all tools combined are well known in the art, all used methods are main domains of computational drug discovery systems and integrating them into a single workflow would be a routine application of known techniques.
Claim 6 is rejected under 35 U.S.C. 103(a) as being unpatentable over Durrant (PMC: 2016 Mar 7), as applied to Claims 1, 4, 7 and 8 above, in view of Sander (Journal of Chemical Information and Modeling: 2015 Jan 5)
Durrant is applied to claims 1, 4, 7 and 8 above.
Durrant is applied to claim 2 above.
Durrant is applied to claims 3 and 10 above.
Durrant is applied to claim 5 above.
Regarding claim 6
Regarding the recited in claim 6, the data log storage subsystem further includes a function of standardizing user permissions. Considering the claims and specification, the limitation “standardizing” is considered as disclosed “The system groups users according to different R&D pipelines, and each user has different permissions for data and logs of various projects.” (0042, page 6)
Under the BRI, the limitation is interpreted as accessibility of tools depend on type of project and data used.
Sander teaches the interface components can be shared among multiple projects.
Sander teaches “User interface components shared among multiple projects, e.g., a chemical editor, molecule and reaction panels, an editable molecule list, clipboard and drag and drop support, molecule and reaction table cell renderers, a dockable view framework, Windows metafile support, and custom user interface components.” (DD_GUI section)
Claim 9 is rejected under 35 U.S.C. 103(a) as being unpatentable over Durrant (PMC: 2016 Mar 7), as applied to Claims 1-4, 6-8 and 10 above, in view of Morris (PMC: 2010 Dec 1).
The claims are drawn to a virtual screening system for crystal complexes. The system consists of 6 subsystems used to view the binding position of the ligand in the protein in the crystal complex, analyze the binding mode of the ligand and the protein, and extract features that enhance the affinity of the drug to the protein, based on these steps and information the compounds are evaluated.
Durrant is applied to claims 1, 4, 7 and 8 above.
Durrant is applied to claim 2 above.
Durrant is applied to claims 3 and 10 above.
Durrant is applied to claim 5 above.
Durrant is applied to claim 6 above.
Regarding claim 9
Regarding the recited in claim 9 comprises following steps of: protein pretreatment: download a protein pdb file of the compounds from a pdb library, perform protein pretreatment operations, delete water molecules, hydrogenate, delete irrelevant ligands, and define the pretreatment of a site that needs to be docked; conformation optimization: carry out a conformation optimization operation for the compounds, after generating a 3D conformation of the compounds, use a genetic algorithm to search for the conformation of the compounds in the lowest energy; molecular docking: perform a molecular docking, sort in descending order according to a score of the molecular docking, and select the top 5%-15% of the compounds; molecular dynamics simulation: perform molecular dynamics simulation on the selected compounds, and screen out qualified compounds from the compound library based on a result of the simulation.
Pre-docking preparations typically involves calculating and displaying secondary structure, adding, or deleting hydrogens, calculating charges and molecular surfaces, and many others.
Durrant did not teach pre-docking preparations. Morris teaches Automated Docking with Selective Receptor Flexibility, were Pdb library was used, perform commands specific to the preparation, launching and analysis of AutoDock calculations such as delete water molecules, hydrogenate, delete irrelevant ligands, optimization operation for the compounds and perform a molecular docking.
Morris teaches “AutoDockTools consists of a set of commands dynamically extending PMV with commands specific to the preparation, launching and analysis of AutoDock calculations. Hence, all PMV commands (such as reading/writing files, calculating and displaying secondary structure, adding, or deleting hydrogens, calculating charges and molecular surfaces, and many others) are also naturally available in AutoDockTools.” (3rd paragraph, Overview of AutoDockTools section)
Morris teaches “This sidechain-ligand structure is then treated as flexible during the docking simulation, searching torsional degrees of freedom to optimize the interaction with the rest of the protein. (Methods for Covalent Docking section)
Morris teaches “Docking experiments were performed with AutoDock4 and compared with docking experiments with AutoDock3.” (Docking Experiments section)
Regarding the use of pdb library.
The PDBbind database is a comprehensive collection of experimentally measured binding affinity data for protein-ligand complexes found in the Protein Data Bank (PDB).
Morris teaches “A set of 188 diverse protein-ligand complexes were taken from the Ligand-Protein Database and a set of 87 HIV protease complexes were taken from the PDB Bind database.” (Validation Data Sets Files section)
It would have been obvious to a person of ordinary skill in the art to combine Durrant and Morris teachings to create a more effective workflow for virtual screening and drug discovery. There is a likelihood of success since both techniques are well known in the art and it is a predictable enhancement without requiring invented effort.
Conclusion
No claim is allowed
Any inquiry concerning this communication or earlier communications from the examiner should be directed to MARWY ELATTAR whose telephone number is (571)272-1182. The examiner can normally be reached full time.
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 on 5712722248. 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.
/M.E./Examiner, Art Unit 1685
/LARRY D RIGGS II/Supervisory Patent Examiner, Art Unit 1686
Prosecution Insights