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and the fee set forth in 37 CFR 1.17(e) has been timely paid, the finality of the previous Office action has been withdrawn pursuant to 37 CFR 1.114.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 3/17/2026 has been entered.
Priority
Applicant’s claim for the benefit of a prior-filed application under 35 U.S.C. 119(e) or under 35 U.S.C. 120, 121, or 365(c) is acknowledged. Priority of foreign application CHINA 202110011160.6 filed 01/06/2021 is acknowledged.
Status of Claims
Claims 2-3, 10-11, and 14-15 are cancelled.
Claims 1, 4-9, 12-13 and 16 are pending and are examined on the merits.
Withdrawn Rejections/Objections
The objections to claims 1, 9 and 13 in the Office action posted 2/11/2026 are withdrawn in view of claim amendments filed 3/17/2026. However, new objections are applied.
The rejections to claims 4, 12 and 16 under 35 U.S.C. 112(d) in the Office action posted 2/11/2026 are withdrawn in view of claim amendments filed 3/17/2026.
Claim Objections
Claims 1, 5, 9 and 13 are objected to because of the following informalities: claim 1, 5, 9 and 13 all recite “a plurality pieces of test data” in their first “wherein” phrases. The phrase is grammatically defective; it should likely be “a plurality of pieces of test data.”
Claims 1, 9 and 13 all recite “wherein the training the affinity prediction model” in the 4th “wherein phrase (par 5). It does not sound like a grammar-correct phrase.
Claims 7 and 8 are materially duplicative except for dependency, which may create redundancy.
Appropriate correction is required.
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.
Claims 1, 4-9, 12-13 and 16 are 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 training compound” at line 12. There is insufficient antecedent basis for this limitation in the claim. The method itself concerns “a compound of the training drug” and screened drug(s). This creates possible antecedent-basis and consistency problems.
Claims 1, 5, 9 and 13 all recites “the affinity prediction model is used to predict the input to obtain an output” at line 13 (the first “wherein” phrase). This limitation is unclear because “the input” are the training target, training drug and test data being input into the prediction model. Reciting that the model is predicting “the input” is unclear because “the input” is used to train the model and therefore should not be predicted by the model, unless this is referring to another input for an already trained model.
Claim 5 recites the limitation “training target” and “training drug” in clauses defining the input of the affinity prediction model. There is insufficient antecedent basis for this limitation in the claim. The method itself concerns a “preset target” and screened drug(s). This creates possible antecedent-basis and consistency problems.
Claim Rejections - 35 USC § 101
This rejection is maintained from the previous Office Action. Modification is necessitated by claim amendments.
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, 4-9, 12-13 and 16 are rejected under 35 U.S.C. 101 because the claimed invention is directed to non-statutory subject matter.
Step 1: Process, Machine, Manufacture or Composition
Claims 1 and 4 are drawn to a method, so a process.
Claims 5-8 are drawn to another method, so another process.
Claims 9 and 12 are drawn to a device, comprising at least one processor, so a machine or a manufacture.
Claims 13 and 16 are drawn to a non-transitory computer readable medium, so a machine or a manufacture.
Step 2A Prong One: Identification of an Abstract Idea
The claim(s) recite(s):
Training the affinity prediction model using the plurality of training samples (claims 1, 9 and 13).
--Under a broadest reasonable interpretation (BRI), training a prediction model requires tunning of parameters according to mathematical formulas and algorithms. Therefore this step is directed to an abstract idea of mathematical concepts.
The affinity prediction model is used to predict the input to obtain an output of the affinity prediction model, and the output of the affinity prediction model is a predicted affinity between the training target and the training drug (claims 1, 9, and 13).
--This element encompasses observing a data set and performing an evaluation to identify affinity data. “Predict” in this element encompasses making a determination about predicted affinity.
Such mental observations or evaluations fall within the “mental processes” grouping of abstract ideas.
Wherein the predicted affinity is positively correlated to the binding capacity between a target and a drug and the inhibition of the target by the drug (claims 1, 5, 9, and 13).
-- This wherein clause describes the correlation between predicted affinity and the binding capacity, target inhibition. Correlating affinity to binding capacity between a target and drug can be performed by the human mind. Therefore, this clause equates to an abstract idea of mental processes.
Wherein the test data set comprises a plurality pieces of test data, each piece of test data comprises the information of the training target, information of each of all training drugs corresponding to the training target, and a known affinity obtained by experiments of the each training drug with the training target (claims 1, 5, 9, and 13).
--This wherein clause describes the test data set and known affinities used to train the model. Therefore, this clause equates to an abstract idea of mental processes.
Wherein the training the affinity prediction model using the plurality of training samples comprises:
selecting a group of training samples from the plurality of training samples to obtain a training sample group;
inputting the selected training sample group into the affinity prediction model, and acquiring a predicted affinity corresponding to each training sample in the training sample group and predicted and output by the affinity prediction model;
constructing a loss function according to the predicted affinity corresponding to each training sample in the training sample group and the known affinity between the training target and the training drug in the corresponding training sample;
detecting whether the loss function converges; and
if the loss function does not converge, adjusting parameters of the affinity prediction model to make the loss function tend to converge (claims 1, 9, and 13).
--This wherein clause describes the processes of training the affinity prediction model. The first step (“selecting …”) is interpreted as data manipulation under a BRI, which can be achieved in human mind; the second step (“inputting …”) is interpreted as a mathematical operations (under a BRI, the prediction model can be a regression model); the third step (“constructing …”) recites mathematical concepts explicitly; the forth step (“detecting …”) recites a judgement process based on data observation, which can be achieved in human mind; the last step (“if …”) recites a typical human mental activity of logical thinking. Therefore, this clause equates to abstract ideas of mental processes and mathematical concepts.
Screening information of at least one drug with a highest predicted affinity with a preset target from a preset drug library using a pre-trained affinity prediction model based on a test data set corresponding to the preset target (claim 5).
--This step reads on a mental judgement process based on data observation. Therefore this step is directed to an abstract idea of mental process.
Acquiring a known affinity obtained by experiments of each of the at least one drug with the preset target based on the screened information of the at least one drug (claim 5);
--This step is interpreted as searching literatures and databases to identify experimentally discovered affinity between the at least one drug with the preset target. This process can be achieved by a human being through searching and reading. Therefore, this step equates to an abstract idea of mental processes.
Updating the test data set corresponding to the preset target based on the information of the several drugs at least one drug and the real known affinity of each drug with the preset target,
wherein the input of the affinity prediction model comprises the information of the training target, the information of the training drug and the test data set corresponding to the training target, the affinity prediction model is used to predict the input to obtain an output of the affinity prediction model, and the output of the affinity prediction model is a predicted affinity between the training target and the training drug (claim 5);
--“Updating the test data” is interpreted as a data manipulation. The next “wherein” clause further describes the data and model involved. Therefore this step is directed to an abstract idea of mental processes.
Wherein the test data set comprises the information of the training target, information of each training drug in all training drugs corresponding to the training target, and a known affinity of the each training drug with the training target (claim 5).
--This wherein clause describes the test data set and known affinities, which is data used by the claimed abstract idea. Therefore, this clause equates to an abstract idea of mental processes.
Wherein all the training drugs corresponding to the training target are the training drugs which are screened from a preset drug library based on a historical affinity prediction model (claim 5).
--This wherein clause describes the training drugs which is the data used in the claimed abstract idea. Therefore, this clause equates to an abstract idea of mental processes.
Step 2A Prong Two: Consideration of Practical Application
The claims result in a process of training the affinity prediction model or updating the test data set, both are directed to abstract ideas (of mathematical concepts or mental processes). The claims do not recite any additional elements that integrate the abstract idea/judicial exception into a practical application.
This judicial exception is not integrated into a practical application because the claims do not meet any of the following criteria:
An additional element reflects an improvement in the functioning of a computer, or an improvement to other technology or technical field;
an additional element that applies or uses a judicial exception to effect a particular treatment or prophylaxis for a disease or medical condition;
an additional element implements a judicial exception with, or uses a judicial exception in conjunction with, a particular machine or manufacture that is integral to the claim;
an additional element effects a transformation or reduction of a particular article to a different state or thing; and
an additional element applies or uses the judicial exception in some other meaningful way beyond generally linking the use of the judicial exception to a particular technological environment, such that the claim as a whole is more than
a drafting effort designed to monopolize the exception.
Step 2B: Consideration of Additional Elements and Significantly More
The claimed method also recites "additional elements" that are not limitations drawn to an abstract idea:
Collecting a plurality of training samples, each training sample comprising information of a training target, information of a training drug and a test data set corresponding to the training target (claim 1);
An electronic device (claim 9);
At least one processor (claim 9);
A memory communicatively connected with the at least one processor (claim 9); and
A non-transitory computer readable storage medium (claim 13).
The recited additional elements are drawn to: acquiring data or generic computer components.
The claims do not include additional elements that are sufficient to amount to significantly more than the judicial exception because the additional elements are insignificant extra-solution activity or generic computer to execute abstract ideas.
The claims do not include additional elements that are sufficient to amount of significantly more than the judicial exception because it is routine and conventional to perform the acts of acquiring necessary data for analysis. Other elements of the method include hardware which are recitations of generic computer components that serve to perform generic computer functions that are well-understood, routine, and conventional activities previously known to the pertinent industry. Viewed as a whole, these additional claim element(s) do not provide meaningful limitation(s) to transform the abstract idea recited in the instantly presented claims into a patent eligible application of the abstract idea such that the claim(s) amounts to significantly more than the abstract idea itself. Therefore, the claim(s) are rejected under 35 U.S.C. 101 as being directed to non-statutory subject matter.
Response to Applicant’s Arguments
In the Remarks filed 3/17/2026, Applicant argues that (page 9, paras 2-3):
With the application as claimed, the accuracy and training effect of the affinity prediction model can be effectively improved. That is, a high-activity compound molecule which may be tightly bound to the protein target can be found and continuously optimized to finally form the drug available for treatment.
Therefore, the claimed features relate to a process implemented by a computer device, which is integrated into a practical application of drug screening, to improve the efficiency of the process implemented by a computer device. Therefore, the claimed invention is not a limitation in mind and is not directed to an abstract idea.
Applicant’s argument refers to Step 2A/Prong two in the 101 analysis, relating to whether claims are integrated into a practical application or not, due to technological improvement.
In response, Applicant’s argument is not persuasive. As discussed above in the 101 analysis, the claims result in a process of training the affinity prediction model or updating the test data set, both are directed to abstract ideas (of mathematical concepts or mental processes). The 101 analysis requires additional elements to apply, to capture and to reflect the judicial exceptions, but the claims do not recite any additional elements that integrate the abstract idea/judicial exception into a practical application.
Therefore, the 101 rejection is maintained.
Claim Rejections - 35 USC § 103
This rejection is maintained from the previous rejection. Modifications are necessitated by claims amendment.
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.
This application currently names joint inventors. In considering patentability of the claims the examiner presumes that the subject matter of the various claims was commonly owned as of the effective filing date of the claimed invention(s) absent any evidence to the contrary. Applicant is advised of the obligation under 37 CFR 1.56 to point out the inventor and effective filing dates of each claim that was not commonly owned as of the effective filing date of the later invention in order for the examiner to consider the applicability of 35 U.S.C. 102(b)(2)(C) for any potential 35 U.S.C. 102(a)(2) prior art against the later invention.
The factual inquiries for establishing a background for determining obviousness under 35 U.S.C. 103 are summarized as follows:
1. Determining the scope and contents of the prior art.
2. Ascertaining the differences between the prior art and the claims at issue.
3. Resolving the level of ordinary skill in the pertinent art.
4. Considering objective evidence present in the application indicating obviousness or nonobviousness.
Claims 1, 4, 9, 12-13 and 16 are rejected under 35 U.S.C. 103 as being unpatentable over Öztürk et al. ("DeepDTA: deep drug–target binding affinity prediction." Bioinformatics 34.17 (2018): i821-i829. Previously cited), in view of Altae-Tran et al.: ("Low data drug discovery with one-shot learning." ACS central science 3.4 (2017): 283-293. Newly cited).
Claim 1 is interpreted as method to train a model that predict affinity between drug and target. Regarding claim 1, Ozturk provides (page i821, section “Abstract/Motivation”) “we propose a deep-learning based model that uses only sequence information of both targets and drugs to predict DT interaction binding affinities”, which teaches A computer-implemented method for training an affinity prediction model
Ozturk provides (page i822, col 2, 1st para) “We use the Davis Kinase binding affinity dataset (Davis et al., 2011) and the KIBA large-scale kinase inhibitors bioactivity data (He et al., 2017; Tang et al., 2014) to evaluate the performance of our model”, which teaches using multiple drug-target affinity data points as training/evaluation samples.
Ozturk provides (page i821, section “Abstract/Motivation”) “sequence information of both targets and drugs” and (page i822, col 2, 1st para) “the sequences of the proteins and SMILES (Simplified Molecular Input Line Entry System) representations of the compounds are used rather than external features or 3D-structures of the binding complexes”, which teaches each training sample includes information of a training target and information of a training drug.
Ozturk provides (page i821, section “Abstract/Motivation”) “the modeling of protein sequences and compound 1D representations with convolutional neural networks (CNNs)”. Ozturk’s protein sequence is an identifying target representation. The claim’s phrase “expression means of a protein” is broader and unclear, but at least protein sequence/target identity is taught.
Ozturk provides (page i822, col 2, 1st para) “SMILES … representations of the compounds”. SMILES is a conventional unique/near-unique compound representation and is at least an identifier of the compound.
Ozturk provides (page i821, section “Abstract/Motivation”) “predict DT interaction binding affinities”, which teaches the output is predicted affinity between target and drug.
Ozturk (page i821, col 1, 1st para) uses Kd/Ki/IC50-related datasets, which teaches binding affinity maps directly to binding capacity. Inhibition is suggested by kinase inhibitor bioactivity datasets such as KIBA (page i822, col 2, 1st para), but the precise “positively correlated” wording may require sign normalization because lower Kd can mean stronger binding unless transformed to pKd or score.
Ozturk teaches Davis has inhibitors with Kd values (page i822, col 2, 3rd para); KIBA combines Ki, Kd, IC50; Altae-Tran (page 284, col 2, paras 1-2) provides that data points are compounds tested in experimental assays with labels. Ozturk teaches experimentally measured affinity/bioactivity datasets. Altae-Tran teaches the target/task support set.
Ozturk provides (page i824, col 1, last para) “A learning model tries to minimize the difference between the expected (real) value and the prediction during training. Since we work on a regression task, we used mean squared error (MSE) as the loss function”; and (page i824, col 2, 1st para) “mini-batch size of 256 was used to update the weights of the network. Adam was used as the optimization algorithm”. Ozturk teaches minibatch training, predicted values, MSE loss, and parameter updating. Convergence detection is a routine training-stop criterion.
Ozturk alone does not clearly input a target-specific support/test set.
Altae-Tran provides (page 284, col 2, 1st para) “each task typically corresponds to an experimental assay”; and (page 284, col 2, 2nd para) “We refer to the collection of available data points for a given task as a ‘support’ set”. Altae-Tran teaches conditioning prediction on a per-task/per-assay support set. This is close to the claimed target-specific test data set.
Ozturk provides (page i821, section “Abstract/Motivation”) “uses only sequence information of both targets and drugs”; and Altae-Tran provides (page 284, col 2, 2nd para) that function h is “parametrized upon choice of support S” and predicts a query compound. Combination of Ozturk and Altae-Tran supplies target/drug inputs from Ozturk and target/task support-set input from Altae-Tran.
Regarding claim 4, Ozturk provides (page i824, col 1, last para)| “Since we work on a regression task, we used mean squared error (MSE) as the loss function”, which teaches a loss function that calculates the sum of MSEs; Ozturk’s MSE differs only by normalization factor, which is a routine equivalent for optimization.
Claims 9, 12 are the electronic device version of the claims 1, 4 method, and claims 13, 16 are the computer readable storage medium version of the claims 1, 4 method respectively.
Regarding claims 9 and 12, Ozturk's DeepDTA implementation is a software (page i821, Title and section Abstract), with code availability at GitHub. Once the method is obvious, implementing it on a generic processor and memory would have been a routine computer implementation. Other than that, the art applied to claims 1 and 4 also teaches claims 9 and 12; and claims 13 and 16 respectively.
It would have been prima facie obvious to a person having ordinary skills in art, to modify Ozturk’s DeepDTA affinity regression framework to condition the prediction not only on the candidate target and candidate drug, but also on a target/task-specific set of experimentally tested compounds and corresponding labels/affinities as taught by Altae-Tran. The motivation would have been to improve prediction in low-data or target-specific settings, which Altae-Tran expressly identifies as a central problem in drug discovery. The modification would have amounted to applying Altae-Tran’s known support-set/one-shot learning architecture to Ozturk’s known drug-target affinity regression task.
One would reasonably expect success because the expected result would have been a model that predicts affinity for a target-drug query while using target-specific experimental context, with predictable benefits in low-data target-specific prediction.
This would be an classic example of “Combining prior art elements according to known methods to yield predictable results” (MPEP §2141.III.(A))
Claims 5-6 are rejected under 35 U.S.C. 103 as being unpatentable over Öztürk et al. ("DeepDTA: deep drug–target binding affinity prediction." Bioinformatics 34.17 (2018): i821-i829. Previously cited), in view of Altae-Tran et al.: ("Low data drug discovery with one-shot learning." ACS central science 3.4 (2017): 283-293. Newly cited), and Warmuth et al. ("Active learning with support vector machines in the drug discovery process." Journal of chemical information and computer sciences 43.2 (2003): 667-673. Newly cited).
Regarding claim 5, Warmuth discloses (page 667, Title and section Abstract) screening small batches from a large collection. Hence Warmuth teaches selecting compounds from a large collection for binding activity testing.
Warmuth screened compounds underrun “biochemical testing” for binding activity (page 667, Title and section Abstract), which teaches experimental follow-up testing of selected compounds.
Warmuth discloses active learning selects successive batches based on current active/inactive compounds (page 667, Title and section Abstract), which teaches iterative experimental update.
Warmuth discloses (page 667, Title and section Abstract) “active learning paradigm … selecting the successive batches” from a large compound collection. Warmuth hence teaches historical/current model-based selection of successive batches, which suggests prior model iteration selecting of later compounds.
Warmuth does not teaches a model conditioned on target/task support data.
Altae-Tran provides (page 284, col 2, 2nd para): “to learn a function h, parametrized upon choice of support S that predicts the probability of any query x”; Ozturk predicts drug-target affinity (page i821, section “Abstract/Motivation”). Combination of Altae-Tran and Ozturk teaches a model conditioned on target/task support data.
Regarding claim 6, Altae-Tran discloses (page 284, col 2, paras 2-4) support set is collection of available data points; Warmuth discloses (page 667, col 2, 2nd para) starts from iterative screening problem and obtains labels through tests. A null/empty support set is a design choice for an initial round. Existing drug/affinity data correspond to a support set.
It would have been prima facie obvious to modify Warmuth’s active-learning screening workflow, with Altae-Tran’s drug discovery method optionally conditioned on target-specific support/test data. Because Warmuth’s iterative active-learning method depends on biochemical testing of selected batches and using accumulated labeled results to guide subsequent selections.
One would reasonably expect success as Warmuth supplies the iterative experimental-screening/update loop, Altae-Tran supplies target-specific support-set conditioning for low-data assays; and both Warmuth and Altae-Tran are about drug discovery through machine learning.
It would have been prima facie obvious to modify the combined drug discovery pipeline of Warmuth and Altae-Tran, which featuring active-learning screening and optionally conditioned on target-specific support/test data, with Ozturk’s teaching of a deep-learning-based drug-target binding affinity prediction model. Because Warmuth’s iterative active-learning method depends on biochemical testing of selected batches and the subsequent selections need be guided by affinity data.
One would reasonably expect success as Warmuth supplies the iterative experimental-screening/update loop, Altae-Tran supplies target-specific support-set conditioning for low-data assays, and Ozturk supplies affinity data to guide subsequent selections, and all Warmuth, Altae-Tran and Ozturk are about drug discovery through machine learning.
Claims 7 and 8 are rejected under 35 U.S.C. 103 as being unpatentable over Warmuth, Altae-Tran, and Öztürk as applied to claims 5 and 6 above, and further in view of Thafar et al.:(“Comparison Study of Computational Prediction Tools for Drug-Target Binding Affinities,” Frontiers in Chemistry, 7:782, published Nov. 20, 2019. Previously cited).
Regarding claim 7, none of Warmuth, Altae-Tran, or Öztürk teaches affinity between drugs and targets. Thafar discloses (page 1, section Abstract) affinity regression can be converted to ranking; and (page 9, col 2, penultimate para) a ranking metric. Warmuth discloses (page 667, col 2, 2nd para), which suggests screening from a large collection.
Regarding claim 8, claim 8 appears substantially duplicative of claim 7, except for dependency through claim 6. The art applied to claim 7 also teaches claim 8.
It would have been prima facie obvious to modify the combined drug discovery pipeline of Warmuth, Altae-Tran, and Ozturk which featuring active-learning screening and optionally conditioned on target-specific support/test data, and generating drug-target binding affinity data. with Thafar’s ranking formulation that rank a preset drug library and select drugs having the highest predicted affinity for a preset target. Because Warmuth’s iterative active-learning method depends on biochemical testing of selected batches and subsequent selections need be prioritized by Thafar’s ranking data.
One would reasonably expect success as Warmuth supplies the iterative experimental-screening/update loop, Altae-Tran supplies target-specific support-set conditioning for low-data assays, Ozturk supplies affinity data, and Thafar supply a ranking formulation to guide subsequent selections, and all Warmuth, Altae-Tran, Ozturk and Thafar are about drug discovery through machine learning.
Response to Applicant’s Argument
In the Remarks filed 3/17/2026, Applicant points out(page 9 last para through page 10, 1st para) that amended claims 1, 5, 9 and 13 now recite: "the test data set comprises a plurality pieces of test data, each piece of test data comprises the information of the training target, information of each of all training drugs corresponding to the training target, and a known affinity obtained by experiments of the each training drug with the training target". .
In response, Applicant’s new amendment is not persuasive. The cited elements are taught by newly applied art. For example:
For the instant claims, Altae-Tran provides (page 284, col 2, 1st para) “each task typically corresponds to an experimental assay”; and (page 284, col 2, 2nd para) “We refer to the collection of available data points for a given task as a ‘support’ set”. Altae-Tran teaches conditioning prediction on a per-task/per-assay support set. This is close to the claimed target-specific test data set.
Ozturk provides (page i821, section “Abstract/Motivation”) “uses only sequence information of both targets and drugs”; and Altae-Tran provides (page 284, col 2, 2nd para) that function h is “parametrized upon choice of support S” and predicts a query compound. Combination of Ozturk and Altae-Tran supplies target/drug inputs from Ozturk and target/task support-set input from Altae-Tran.
Ozturk provides (page i822, col 2, 1st para) “We use the Davis Kinase binding affinity dataset (Davis et al., 2011) and the KIBA large-scale kinase inhibitors bioactivity data (He et al., 2017; Tang et al., 2014) to evaluate the performance of our model”, which teaches using multiple drug-target affinity data points as training/evaluation samples.
Therefore combined Ozturk and Altae-Tran teach the test data related to the targets and related to the drugs.
Applicants argue (page 1, par. 4) that Thafar fails to disclose at lease the test data set as claimed which comprises a plurality of pieces of test data, each piece comprising information of the training target, information of all training drugs corresponding to the training target, and a known affinity of the training drug with the training target.
In response, Applicant’s argument is not persuasive. As discussed above, Ozturk provides (page i821, section “Abstract/Motivation”) “uses only sequence information of both targets and drugs”; and Altae-Tran provides (page 284, col 2, 2nd para) that function h is “parametrized upon choice of support S” and predicts a query compound. Combination of Ozturk and Altae-Tran supplies target/drug inputs from Ozturk and target/task support-set input from Altae-Tran.
Altae-Tran provides (page 284, col 2, 1st para) “each task typically corresponds to an experimental assay”; and (page 284, col 2, 2nd para) “We refer to the collection of available data points for a given task as a ‘support’ set”. Altae-Tran teaches conditioning prediction on a per-task/per-assay support set. This is close to the claimed target-specific test data set.
Ozturk provides (page i822, col 2, 1st para) “We use the Davis Kinase binding affinity dataset (Davis et al., 2011) and the KIBA large-scale kinase inhibitors bioactivity data (He et al., 2017; Tang et al., 2014) to evaluate the performance of our model”, which teaches using multiple drug-target affinity data points as training/evaluation samples.
Therefore combined Ozturk and Altae-Tran teach the test data related to the targets and related to the drugs.
Therefore, the 103 rejection is maintained.
Conclusion
No claims are allowed.
Any inquiry concerning this communication or earlier communications from the examiner should be directed to GUOZHEN LIU whose telephone number is (571)272-0224. The examiner can normally be reached Monday-Friday 8-5.
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.
/GL/
Patent Examiner
Art Unit 1686
/Anna Skibinsky/
Primary Examiner, AU 1635
Prosecution Insights