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 .
Applicant’s amendment and response, filed 1/26/2026 has been entered and carefully considered.
Claims 1-6, 8-11, 17, 19-21, 23-25 are under examination to the extent they read on the elected species of “AI” modeling.
The rejection under 35 USC 101 has been withdrawn, as any abstract idea is specifically and practically applied utilizing particular user interface and display elements particularly connected to and driven by the method, achieving patent-eligibility under step 2A-2.
The rejections under 35 USC 112(a) and (b) are withdrawn in view of Applicant’s amendments and arguments.
The rejections under 35 USC 103(a) are withdrawn, however new grounds of rejection are set forth below.
As set forth previously, this application was filed under 35 USC 371, as a National Stage application of PCT/KR2024/000313 which claims priority to two KR priority documents. One Priority document has been retrieved from the IB. A certified translation of KR 10-2023-
0197930, filed 12/28/2023 has been received. No copy or translation appears to be of record for KR 10-2023-0097539, filed 7/26/2023. Therefore, the effective filing date for the claims is 12/28/2023.
Claim Rejections - 35 USC § 102
In the event the determination of the status of the application as subject to AIA 35 U.S.C. 102 and 103 (or as subject to pre-AIA 35 U.S.C. 102 and 103) is incorrect, any correction of the statutory basis (i.e., changing from AIA to pre-AIA ) for the rejection will not be considered a new ground of rejection if the prior art relied upon, and the rationale supporting the rejection, would be the same under either status.
The following is a quotation of the appropriate paragraphs of 35 U.S.C. 102 that form the basis for the rejections under this section made in this Office action:
A person shall be entitled to a patent unless –
(a)(1) the claimed invention was patented, described in a printed publication, or in public use, on sale, or otherwise available to the public before the effective filing date of the claimed invention.
(a)(2) the claimed invention was described in a patent issued under section 151, or in an application for patent published or deemed published under section 122(b), in which the patent or application, as the case may be, names another inventor and was effectively filed before the effective filing date of the claimed invention.
Claim(s) 1-3, 5, 6 and 11, 19-21, 23-25 is/are rejected under 35 U.S.C. 102a1 as being anticipated by Prat (2022) in light of RCSB PDB Help (2025).
Prat, A. et al. System and method for molecular reconstruction from molecular probability distributions. US 2022/0198286 A1, published 6/23/2022, with a filing date as early as 12/1/2021. Matured to US Patent 12,223,435 B2.
RCSB PDB Help (2025) Search Examples, The Protein Database Help Documentation, downloaded 2/2026, 6 pages. www (dot ) rcsb.org/docs/search-and-browse/advanced-search/search-examples.
Prat provides cloud-based platforms [0224] for use in computer-implemented methods, in a web-based service [0223] to provide structure-based candidate discovery. The data platform and user interface of Prat is illustrated in Fig 1 and discussed at least at [0092-0096].
[0092]: “The EDA interface 112 is a user interface through which pharmaceutical research may be performed by making queries and receiving responses. The queries are sent to a data analysis engine 113 which uses the knowledge graph 111 to determine a response, which is then provided to the user through the EDA interface 112. In some embodiments, the data analysis engine 113 comprises one or more graph-based neural networks (graph neural networks, or GNNs) to process the information contained in the knowledge graph 111 to determine a response to the user's query. As an example, the user may submit a query for identification of molecules likely to have similar bioactivity to a molecule with known bioactivity. The data analysis engine 113 may process the knowledge graph 111 through a GNN to identify such molecules based on the information and relationships in the knowledge graph 111.”
The user interface of Prat is conceptually shown in Fig 2, with nodes and directed arrows illustrating tasks, order of execution, and databases. Element 250 of Fig 2 appears to meet the BRI of the simulation setup area, module selection area, and canvas area, as further illustrated at Fig 27.
Prat creates a docking task node for a target protein and one or more ligands, to generate pose data and binding energy data. Docking is defined at [0081]. Docking procedures, and training the docking module is disclosed beginning at [0099, 0136], including the determination of more than one pose, and differing types of binding energy calculations. Pose scoring is disclosed at [0136]. The docking simulator in the simulation space is disclosed in Fig 26, and described beginning at [0145]. Figure 28 illustrates the calculation of energy states, as discussed beginning at [0146]. Fig 27 illustrates the architecture of a 3dCNN, and docking simulations, discussed at [0200]. Figure 29 illustrates the directed workflow, described beginning at [0207].
Prat determines an affinity prediction task, predicting binding affinity between the protein and the ligand as shown in: [0145], the calculations at [0201-202], Fig 29, and at [0207]. The binding energy value is related to affinity, in that the lower the energy state of the docked pose, the stronger the binding affinity.
Prat stores workflow metadata in memory, in the “data curation platform” discussed at [0146, 0161-0162, 0178-0179, 0198, 0207-0209, 0216, 0220 et al.]
Prat provides families of ligands to be tested and docked with the target protein, for example at [0101]:
“At the analysis stage, a query in the form of a target ligand 244 and a target protein 245 are entered using an exploratory drug analysis (EDA) interface 250. The target ligand 244 is processed through the trained graph-based machine learning algorithm 241 which, based on its training, produces an output comprising a vector representation of the likelihood of interaction of the target ligand 244 with certain proteins and the likelihood of the bioactivity resulting from the interactions. Similarly, the target protein 245 is processed through the trained sequence-based machine learning algorithm 242 which, based on its training, produces an output comprising a vector representation of the likelihood of interaction of the target protein 245 with certain ligands and the likelihood of the bioactivity resulting from the interactions. The results may be concatenated 243 to strengthen the likelihood information from each of the separate trained machine learning algorithms 241, 242.” (col 12-13). Figure 3 illustrates a set of ligand molecules in a database, as well as a protein database. Fig 14 illustrates obtaining ligand information for known ligands, as described at col 17. Cheminformatic libraries of ligand information can be obtained for scoring and docking to a target protein, as set forth at col 22. “According to one aspect, various cheminformatics libraries may be used as a learned force-field for docking simulations, which perform gradient descent of the ligand atomic coordinates with respect to the binding affinity 1806 and pose score 1805 (the model outputs).”
The user-controlled settings include the type of model, whether docking is correct or incorrect, whether the complex is predicted to be active or inactive, threshold setting, bond type, structure type, volume etc.
Prat displays ligand identifiers and predicted binding energy values in Fig 28, discussed beginning at [0146].
Once a ligand is selected, the binding affinity value can be calculated for the selected ligand and the target protein, as in [0145, 0169 0200-0202, 0207]. Ultimately the selected ligand information can be displayed, along with calculated values.
With respect to the display of binding energy values in Kcal/mol and binding affinity in molar units, these are routine value units for these calculations, as illustrated by RCSB PDB Help. Micromolar binding affinity concentration values are provided (µM), and free energy binding values are provided in kcal/mol or kJ/mol.
The routine use of kcal/mol, and micromolar concentrations meet claims 2-3.
With respect to the system, this is met throughout, meeting claim 11.
With respect to claim 5, and claim 19, the AI model is a 3D-CNN model, trained with dissociation constants, inhibition constants and/or IC50 data as shown in [0207].
“According to aspects of various embodiments, simple force-field based optimization of a ligand pose in a binding pocket can substitute for docked poses at reduced computational expense in a binding affinity prediction task without a significant decrease in accuracy. Force-field optimization considers at least one of the constant terms selected from the list of dissociation, inhibition, and half-concentration (IC50) in order to capture the molecular interactions, e.g., hydrogen bonds, hydrophobic bonds, etc. Many databases known in the art may be used to get this information such as the Protein Data Bank (PDB) as one example. In simple terms, docking guides the machine learning (3D-CNN) to realize what poses to keep and to realize what the molecule likely looks like in the pocket.”
With respect to claim 6, claim 19, and claim 21, experimental data, for example from the PDB for protein and ligand can be used, as in [0207].
With respect to claim 8, and claim 23, the CNN has convolutional layers as required. [0127] “According to an embodiment of active example generation, a 3-dimensional convolutional neural network (3D CNN) is used in which atom-type densities are reconstructed using a sequence of 3D convolutional layers and dense layers.”
With respect to claim 9, and claim 24 the CNN has dense layers as required. [0122, 0127, 0140, 0142, 0144, 0147].
“According to an embodiment of active example generation, a 3-dimensional convolutional neural network (3D CNN) is used in which atom-type densities are reconstructed using a sequence of 3D convolutional layers and dense layers.” The dense layers have nodes or neurons, as required. (Fig 21, Fig 23). Fig 25 illustrates a CNN which comprises three dense layers. [0144] “FIG. 25 is a block diagram of a model architecture of an autoregressive decoder 2500 for de novo drug discovery according to one embodiment. Latent vectors of size dimension z are inputs 2501 to the autoregression decoder 2500 and subsequently calculated into dense layers 2502 where their dimensions may be expanded. A concatenation function 2503 precedes a second dense layer 2504 where pre-LSTM feature extraction occurs. After the LSTM cell function 2505, which corresponds to the LSTM recurrence operation, another concatenation occurs 2506 before a third dense layer 2507 extracts nonlinear features. The loop between the third dense layer 2507 and the first concatenation has no atoms. The fourth dense layer 2508 processes atom node features for the stack 2409 to begin node reconstruction…”
With respect to claim 10, and claim 25 the AI model is a CNN, throughout.
With respect to claim 20, node input/output dependencies are saved as set forth at [0118].
“A second approach begins with augmenting the transformer model with a hyper-graph approach. Starting with an initial node of the graph as the query molecule and recursively: the molecule with highest upper-bound confidence (UCB) score is selected (specifically, the UCB is adapted to trees generation UCT), the node is expanded (if this node is not terminal), and expansions from that node are simulated to recover a reward. Rewards are backpropagated along the deque of selected nodes, and the process is repeated until convergence. Here UCB is used as a form of balancing exploration-exploitation, where X is the reward, n is the number of times the parent node has been visited, j denotes the child node index, and C.sub.p (>0) is an exploration constant. In one embodiment, the model may be constrained to a rewarding a node when its children are accessible, wherein other embodiments may use rewards such as molecular synthesis score, Log P, synthesis cost, or others known in the art.” Node/edge dependencies can be created and save through Message passing within the model. See also Fig 10, 11A-B, 12, 21, 23.
Claim(s) 1-6, 8, 11, 17, 19, 20 and 23 is/are rejected under 35 U.S.C. 102a1 as being anticipated by Zavoronkovs (2021).
Zavoronkovs et al. Workflow for generating compounds with biological activity against a specific biological target. US 2021/0057050 A1, published 2/5/2021, matured to US 12,293,809 B2.
Zavoronkovs provides cloud-based platforms [0079] for use in computer-implemented methods, in a web-based service [0079] to provide structure-based candidate discovery. The data platform and user interface of Zavoronkovs is illustrated in Fig 10 and discussed at least at [0080-0110].
The user interface of Zavoronkovs is conceptually shown in Fig 1A-1B, Fig 10A-10B, with nodes and directed arrows illustrating tasks, order of execution, and databases.
“[0063] In an example generative model, the GENTRL model operates on a AI-based platform that combines ML and deep learning (DL) models (GANs, Autoencoders, RNN-based language models, Genetic algorithms, combinatorial approaches, Ensembles). These methods are combined with the RL optimization and integrated into an end-to-end pipeline (FIGS. 1A-1B). In practice, once the user has entered all required information, the generative protocol begins with up to 30 models running in parallel. The average duration for a standard generation experiment is around 72 hours. The interface allows following the progress of the generation process together with the performance and convergence rate of each model in real time. All molecules which are progressively being generated can be compared in terms of various metrics using the interactive capabilities of the user interface.” See also [0080, 0229].
Zavoronkovs creates a docking task node for a target protein and one or more ligands, to generate pose data and binding energy data. Docking is disclosed at [0050, 0075, 0090, 0134]. Docking procedures, and training the docking module is disclosed beginning at [0179], including the determination of more than one pose, and differing types of binding energy calculations. Pose scoring is disclosed at [0179]. The docking simulator in the simulation space is disclosed in Figs 1, 10, and described beginning at [0134, 0179]. The calculation of energy states is discussed beginning at [0179], with values displayed in kcal/mol.
Zavoronkovs uses an AI generative-based tensorial reinforcement learning model, GENTRL, (Abstract, [0038, 0041-0050, Fig 14, 0061, 0063, 0087, 0111 et al.]) trained with experimental data [0008, 0042, 0049-0050, 0066, 0111, 0139-0140, 0146-0162 et al], and values of dissociation, inhibition, IC50, et al. [0050, 0051, 0060, 0117, 0128, Table 2, etc].
Zavoronkovs determines an affinity prediction task, predicting binding affinity between the biological target and the ligand as shown in [0006, 0065, 0075, 0082]. The binding energy value is related to affinity, in that the lower the energy state of the docked pose, the stronger the binding affinity.
Zavoronkovs stores workflow metadata in memory, using the MOSES module, as in [0077, 0221, 0229-0231].
Zavoronkovs provides families of ligands to be tested and docked with the target protein, for example the reference compounds used for training, and groups of generated compounds [0008, 0009, 0049, 0060 et al].
“[0009] In some embodiments, a computer-implemented method can include: receiving input of a biological target or ligand for any biological target (e.g., the biological target or other biological target); receiving input of properties of a generated compound; receiving at least one generative model trained with reference compounds; generating structures of generated compounds with each generative model, wherein the generated compounds are designed to interact with the biological target and/or correlate with structural features of the ligand; prioritizing structures of the generated compounds of each generative model based on at least one reward criteria; processing prioritized chemical structures of the generated compounds through a Sammon mapping protocol to obtain hit structures; and providing chemical structures of the hit structures.”
The user-controlled settings include the type of model, whether docking is correct or incorrect, whether the complex is predicted to be active or inactive, threshold setting, bond type, structure type, volume etc. [0077, 0086 et al]
Zavoronkovs displays ligand identifiers [0049] and predicted binding energy values as set forth at [0006, 0065, 0075, 0082 et al.]
Once a ligand is selected, the binding affinity value can be calculated for the selected ligand and the target protein, as in [0145, 0169 0200-0202, 0207]. Ultimately the selected ligand information can be displayed, along with calculated values.
With respect to the display of binding energy values in Kcal/mol and binding affinity in molar units (i.e. micromolar, nanomolar), these are routine value units for these calculations, and are utilized by Zavoronkovs. [0117-0118, 0122, 0127-0128, 0180, et al.]
With respect to claims 2-3, these are met at [0117-0118, 0122, 0127-0128, 0180, et al.].
With respect to claim 4, and claim 17, sorting is set forth at [0051, 0109, 0111, 0142, 0176.]
With respect to claim 5, and claim 19, Zavoronkovs uses an AI generative-based tensorial reinforcement learning model, GENTRL, (Abstract, [0038, 0041-0050, Fig 14, 0061, 0063, 0087, 0111 et al.]) trained with experimental data [0008, 0042, 0049-0050, 0066, 0111, 0139-0140, 0146-0162 et al], and values of dissociation, inhibition, IC50, et al. [0050, 0051, 0060, 0117, 0128, Table 2, etc].
With respect to claim 6, the training data is experimental data and structural data of biological target and ligand/ compound. [0008, 0042, 0049-0050, 0066, 0111, 0139-0140, 0146-0162 et al].
With respect to claim 8, and claim 23 the encoder of GENTRL has at least 3 layers with 128 channels/filters [0189]
With respect to claim 20, Zavoronkovs stores workflow metadata in memory, using the MOSES module, as in [0077, 0221, 0229-0231].
Double Patenting
The nonstatutory double patenting rejection is based on a judicially created doctrine grounded in public policy (a policy reflected in the statute) so as to prevent the unjustified or improper timewise extension of the “right to exclude” granted by a patent and to prevent possible harassment by multiple assignees. A nonstatutory double patenting rejection is appropriate where the conflicting claims are not identical, but at least one examined application claim is not patentably distinct from the reference claim(s) because the examined application claim is either anticipated by, or would have been obvious over, the reference claim(s). See, e.g., In re Berg, 140 F.3d 1428, 46 USPQ2d 1226 (Fed. Cir. 1998); In re Goodman, 11 F.3d 1046, 29 USPQ2d 2010 (Fed. Cir. 1993); In re Longi, 759 F.2d 887, 225 USPQ 645 (Fed. Cir. 1985); In re Van Ornum, 686 F.2d 937, 214 USPQ 761 (CCPA 1982); In re Vogel, 422 F.2d 438, 164 USPQ 619 (CCPA 1970); In re Thorington, 418 F.2d 528, 163 USPQ 644 (CCPA 1969).
A timely filed terminal disclaimer in compliance with 37 CFR 1.321(c) or 1.321(d) may be used to overcome an actual or provisional rejection based on nonstatutory double patenting provided the reference application or patent either is shown to be commonly owned with the examined application, or claims an invention made as a result of activities undertaken within the scope of a joint research agreement. See MPEP § 717.02 for applications subject to examination under the first inventor to file provisions of the AIA as explained in MPEP § 2159. See MPEP § 2146 et seq. for applications not subject to examination under the first inventor to file provisions of the AIA . A terminal disclaimer must be signed in compliance with 37 CFR 1.321(b).
The filing of a terminal disclaimer by itself is not a complete reply to a nonstatutory double patenting (NSDP) rejection. A complete reply requires that the terminal disclaimer be accompanied by a reply requesting reconsideration of the prior Office action. Even where the NSDP rejection is provisional the reply must be complete. See MPEP § 804, subsection I.B.1. For a reply to a non-final Office action, see 37 CFR 1.111(a). For a reply to final Office action, see 37 CFR 1.113(c). A request for reconsideration while not provided for in 37 CFR 1.113(c) may be filed after final for consideration. See MPEP §§ 706.07(e) and 714.13.
The USPTO Internet website contains terminal disclaimer forms which may be used. Please visit www.uspto.gov/patent/patents-forms. The actual filing date of the application in which the form is filed determines what form (e.g., PTO/SB/25, PTO/SB/26, PTO/AIA /25, or PTO/AIA /26) should be used. A web-based eTerminal Disclaimer may be filled out completely online using web-screens. An eTerminal Disclaimer that meets all requirements is auto-processed and approved immediately upon submission. For more information about eTerminal Disclaimers, refer to www.uspto.gov/patents/apply/applying-online/eterminal-disclaimer.
Claims 1-6, 8-11, 17, 19-21, 23-25 are rejected on the ground of nonstatutory double patenting as being unpatentable over claims 1-18 of U.S. Patent No. 12,505,901. Although the claims at issue are not identical, they are not patentably distinct from each other because they are both drawn to ligand candidate discovery systems, comprising workflow areas, simulation areas, data management areas, task execution, canvas areas, docking, structural free energy predictions, binding free energy predictions, directed task flow, etc. The patent claims are to generating “drug compounds” which fall within the category of ligands of the instant application.
Conclusion
Applicant's amendment necessitated the new ground(s) of rejection presented in this Office action. Accordingly, THIS ACTION IS MADE FINAL. See MPEP § 706.07(a). Applicant is reminded of the extension of time policy as set forth in 37 CFR 1.136(a).
A shortened statutory period for reply to this final action is set to expire THREE MONTHS from the mailing date of this action. In the event a first reply is filed within TWO MONTHS of the mailing date of this final action and the advisory action is not mailed until after the end of the THREE-MONTH shortened statutory period, then the shortened statutory period will expire on the date the advisory action is mailed, and any nonprovisional extension fee (37 CFR 1.17(a)) pursuant to 37 CFR 1.136(a) will be calculated from the mailing date of the advisory action. In no event, however, will the statutory period for reply expire later than SIX MONTHS from the mailing date of this final action.
Any inquiry concerning this communication or earlier communications from the examiner should be directed to MARY K ZEMAN whose telephone number is 5712720723. The examiner can normally be reached on 8am-2pm M-F. Email may be sent to mary.zeman@uspto.gov if the appropriate permissions have been filed.
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 Riggs can be reached on 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 an application may be obtained from the Patent Application Information Retrieval (PAIR) system. Status information for published applications may be obtained from either Private PAIR or Public PAIR. Status information for unpublished applications is available through Private PAIR only. For more information about the PAIR system, see http://pair-direct.uspto.gov. Should you have questions on access to the Private PAIR system, contact the Electronic Business Center (EBC) at 866-217-9197 (toll-free). If you would like assistance from a USPTO Customer Service Representative or access to the automated information system, call 800-786-9199 (IN USA OR CANADA) or 571-272-1000.
/MARY K ZEMAN/ Primary Examiner, Art Unit 1686