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and the fee set forth in 37 CFR 1.17(e) has been timely paid, the finality of the previous Office action has been withdrawn pursuant to 37 CFR 1.114.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 2/23/2026 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 US application 62/991,471 filed 3/18/2020 is acknowledged.
Claim Status
Claim 2 is cancelled.
Claims 1 and 3-32 are pending, while claims 25-26 are withdrawn.
Claims 1, 3-24, 27-32 are examined on the merits.
Claim Rejections - 35 USC § 101
The instant rejection is maintained from the previous Office Action filed 10/22/2025 and modified in view of Applicant’s amendments filed 2/23/2026.
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, 3-24 and 27-32 are rejected under 35 USC § 101 because the claimed inventions are directed to non-statutory subject matter.
Step 1: The Four Categories of Statutory Subject Matter (MPEP 2106.03)
Claims 1, 3-17 and 31 are directed to a 101 process, here a "computer-implemented method," for completion of entries in drug-target interaction matrix by predicting drug and target interactions from coupled datasets.
Claim 18-24, 27-30 and 32 are directed to another 101 process, here another “computer-implemented method,” for performing completion of entries in a drug-target interaction matrix by predicting drug and target interactions from coupled datasets.
Step 2A, Prong One: Whether the Claims Set Forth or Describe a Judicial Exception(MPEP 2106.04 § II.A.1)
Claim 1 recites:
Identifying, by a computer processor, incomplete entries in the drug-target interaction matrix;
The size and dimension of the interaction matrix is not specified. Under a broadest reasonable interpretation (BRI), the matrix encompasses a 2D matrix. This step recites a judgement activity based on data observation, which can be achieved in human mind with the aid of a pen and a printed matrix. Therefore, this step equates to an abstract idea of mental processes.
Determining, by the computer processor, a matrix optimization function for use in accessing the drug-drug matrix and for use in accessing the target-target matrix;
How the matrix optimization function is determined is not specified. Under a BRI, this step recites a decision-making process (to pick-up a matrix optimization function) which can be achieved in a human mind. Therefore, this step equates to an abstract idea of mental processes.
Using the matrix optimization function, accessing, by the computer processor, a subset of entries in the drug-target interaction matrix, using the matrix optimization function, accessing one or more entries in the drug-drug matrix and one or more entries in the target-target matrix based on the subset of entries;
This step recites accessing matrix data by a program function, and since under a BRI, the matrix encompasses a 2D matrix. Hence this step can be achieved in human mind. Therefore, this step equates to an abstract idea of mental processes.
Performing, by the computer processor, an optimization of the matrix optimization function until a predicted interaction entry, corresponding to an entry from the drug-drug matrix and an entry from the target-target matrix, is identified for completing one of the incomplete entries of the drug-target interaction matrix and updating the drug-target interaction matrix forming an updated drug-target interaction matrix including the predicted interaction entry;
According to instant claim 9, the matrix optimization function is a mathematical function. Performing the matrix optimization function hence reads on mathematical operations according to mathematical formula given in claim 9. Also, “updating the drug-target interaction matrix forming an updated drug-target interaction matrix” reads on a data manipulation, which, under a BRI, can be achieved in human mind. Therefore, this step equates to abstract ideas of mathematical concepts and of mental processes.
Receiving, by the computer processor, subsequent drug and/or target data, comparing the subsequent drug and/or target interaction data to the updated drug-target interaction matrix;
Under a BRI, this step recites a data validation process which compares the predicted interactions (updated drug-target interaction matrix) with newly acquired data (such as experimental data). This data comparison can be achieved in human mind. Therefore, this step equates to an abstract idea of mental processes.
Performing, by the computer processor, an alternating optimization between the drug-drug matrix and the target-target matrix until the predicted interaction entry is identified, wherein the alternating optimization comprises alternatively (i) fixing a drug-drug matrix optimization while minimizing a target-target matrix optimization and (ii) fixing the target-target optimization while minimizing the drug-drug matrix optimization.
According to instant claim 9, the matrix optimization function is a mathematical function. Performing the matrix optimization function hence reads on mathematical operations according to mathematical formula given in claim 9. Therefore, this step equates to abstract ideas of mathematical concepts.
Claim 18 recites:
Identifying, by a computer processor, incomplete entries in the drug-target interaction matrix;
This step recites a judgement activity based on data observation, which can be achieved in human mind. Therefore this step equates to an abstract idea of mental processes.
Determining, by the computer processor, a tensor optimization function for use in accessing the drug-drug tensor and for use in accessing the target-target tensor;
This step recites a decision-making process for selecting a tensor optimization function that accesses data. Under a BRI, this step can be achieved in human mind. Therefore, this step equates to an abstract idea of mental processes.
Using the tensor optimization function, accessing, by the computer processor, a subset of entries in the drug-target interaction matrix and, using the tensor optimization function, accessing a plurality of slices of the drug-drug tensor and a plurality of slices of the target-target tensor;
The size of the tensor or matrix is not specified. Under one embodiment they can be small. Hence this step recites accessing a subset of data, which can be achieved in human mind. Therefore, this step equates to an abstract idea of mental processes.
Performing, by the computer processor, an optimization of the matrix optimization function for each of the plurality of slices of the drug-drug tensor and for each of the plurality of slices of the target-target tensor, thereby producing an optimum candidate drug entry and optimum candidate target entry for each slice;
According to instant claim 9, the matrix optimization function is a mathematical function. Performing the matrix optimization function hence reads on mathematical operations according to mathematical formula given in claim 9. Also, “producing an optimum candidate drug entry and optimum candidate target entry for each slice” reads on a data manipulation, which, under a BRI, can be achieved in human mind. Therefore, this step equates to abstract ideas of mathematical concepts and of mental processes.
Performing, by the computer processor, an optimization on the optimum candidate drug entries and on the optimum candidate target entries for each of the slices to identify a predicted interaction entry containing an optimum entry from the drug-drug tensor and an optimum entry from the target-target tensor, and populating one of the incomplete entries of the drug-target interaction matrix with the predicted interaction entry to form an updated drug-target interaction matrix;
According to instant claim 9, the matrix optimization function is a mathematical function. Performing the matrix optimization function hence reads on mathematical operations according to mathematical formula given in claim 9. Also, “populating one of the incomplete entries of the drug-target interaction matrix with the predicted interaction entry to form an updated drug-target interaction matrix” reads on a data manipulation, which, under a BRI, can be achieved in human mind. Therefore, this step equates to abstract ideas of mathematical concepts and of mental processes.
Receiving, by the computer processor, subsequent drug and/or target data, comparing the subsequent drug and/or target interaction data to the updated drug-target interaction matrix and
Under a BRI, this step recites a data validation process which compares the predicted interactions (updated drug-target interaction matrix) with newly acquired data (such as experimental data). This data comparison can be achieved in human mind. Therefore, this step equates to an abstract idea of mental processes.
Hence, the claims explicitly recite elements that, individually and in combination, constitute abstract ideas. The claims as a whole, read on data analysis and data modeling in a generic computer. The claims must therefore be examined further to determine whether they integrate that abstract idea into a practical application (MPEP 2106.04(d)).
Step 2A, Prong Two: Whether the Claims Contain Additional Elements that Integrate the Judicial Exception(s) into a Practical Application (MPEP 2106.04 § II.A.2)
Both claims 1 and 18 resulted in comparing the predicted drug-target interactions with additional data entries, which reads on a mental activity that can be achieved in the human mind. The claims as a whole, reads on mental and math steps which results in information. The claims do not recite additional elements that integrate the abstract idea 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 following additional elements are recited in claims:
accessing, by the computer processor, a drug-drug matrix and a target-target matrix (claim 1);
accessing, by the computer processor, a drug-drug tensor and a target-target tensor (claim 18);
receiving, by the computer processor, subsequent drug and/or target data (claims 1 and 18); and
outputting one or more resulting interaction entries from the updated drug-target interaction matrix (claims 1 and 18).
The additional elements that are not JEs recited in claims are drawn to data inputting (the “accessing, , by the computer processor, a drug-drug matrix and a target-target matrix” and the “receiving, , by the computer processor, subsequent drug and/or target data” steps both recited in claims 1 and 18) and data outputting (the “outputting one or more resulting interaction entries from the updated drug-target interaction matrix” recited in claims 1 and 18). The data inputting/outputting are insignificant extra-solution activities (MPEP 2106.05(g)); regarding the “computer-implemented” elements recited in many claims, the Court found that the recitation of the computer in the claim amounted to mere instructions to apply the abstract idea on a generic computer. (MPEP 2106.05(f)).
The claim(s) does/do not include additional elements that are sufficient to amount to significantly more than the judicial exception. Instead, the claims as a whole, reads on generating additional data from existing data, or an algorithm which is entirely an abstract idea or series of steps that can be performed by the human mind or mathematical calculations.
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.
As such, claims 1, 3-24 and 27-32 are not patent eligible under 35 U.S.C. §101.
Response to Applicant’s Arguments
In the Remarks filed 2/23/2026, Applicant argues (page 11, penultimate para through page 12, 1st para) that the human mind is incapable of practically performing the claimed steps. In response, Applicant’s argument is not persuasive. Claim elements are examined under the broadest reasonable interpretation (BRI). Under a BRI, the matrix and the tensor can have a small size and the operation to these matrix/tensor becomes practically possible in the human mind.
In the Remarks (page 12, 3rd para through page 14, 1st para), Applicant further argues that “alternating optimization across coupled matrices”, “scale and complexity of matrix operations”, “computational runtime data”, and “tensor optimization across multiple slices” are not practically performable in the human mind. In response, Applicant’s argument is not persuasive. The matrix/tensor optimization function, according to the dependent claim 9, reads on a mathematical function in the format of equations. Therefore, these optimization operations, are drawn to mathematical concepts. Afterall, “accessing, by the computer processor, a subset of entries…” and “accessing one or more entries in the drug-drug matrix” reads on something achievable in human mind, under a BRI. Even though the processes are recited as done by a computer processor, a mental process executed in computer is still a mental process. The argued size 678X477, is just one embodiment (“in an example” par [0140]). Claim 1 does not recite an iterative operation. Accessing a small piece of data one time is practically achievable in the human mind and reads on thinking.
In the Remarks (page 14, paras 2-3), Applicant argues that “the Federal Circuit's reasoning in SRI Int'l, Inc. V. Cisco Systems, Inc., 930 F.3d 1295 (Fed. Cir. 2019), is instructive here.” In response, Applicant’s argument is not persuasive. The network packet data is different from drug-target association data. The network packet data even at a small scale, is not readable to a human, which is different from the human readable, thinkable drug-target association data.
In the Remarks (page 14, para 4 through page 15, para 1), Applicant argues that “Thales Visionix Inc. V. United States, 850 F.3d 1343 (Fed. Cir. 2017), is instructive.” In response, Applicant’s argument is not persuasive. The motion tracking data is different from drug-target association data. The reason the claims in Thales were statutory is because they involved two sensors placed in unconventional locations. It is not because of the data or calculations. Thales claims had unconventional additional elements, which would make the claims statutory under Step 2B, Which is different form the human readable, thinkable drug-target association data. The claimed matrix or tensor data structure on the other hand, are readable and thinkable. Applicant cannot mix what is invented with what is claimed. Under a BRI, claim 1 (and similarly claim 18) accesses a small piece of data one time then throw the data into a mathematical equation. Such a process is practically achievable in the human mind. As to the mathematical operations, an long, iterative math operation is still drawn to a mathematical concept.
In the Remarks (page 16, para 2 through page 17, para 1), Applicant argues that “the claims address a specific technical problem in drug-target interaction prediction”. In response, Applicant’s argument is not persuasive. The claims as a whole, reads on a method of predicting new drug-target interactions based on existing data. The claimed invention might have done a good data modeling, but there are no additional elements to apply, to capture and to reflect the judicial exceptions, which are required at Step 2A/Prong two in order to be 101 eligible. The recited additional elements, such as receiving input drug-drug and target-target matrix data and “outputting one or more resulting interaction entries”, are insignificant extra-solution activities as they are necessary for data analysis. The “computer” and “processor”, amounted to mere instructions to apply the abstract idea on a generic computer. (MPEP 2106.05(f)).
In the Remarks (page 17, para 2 through page 18, para 1), Applicant argues that “the claims do not merely link a judicial exception to a particular technological environment”. In response, Applicant’s argument is not persuasive. It is generally true that the claims do not merely link a judicial exception to a particular technological environment. However, “the claims do not merely link a judicial exception to a particular technological environment” is not the only factor weighted in the 35 USC 101 analysis. The argued technological improvement through “ordered combination” of elements are not persuasive. The claimed invention might have good data modeling, but there are no additional elements to apply, to capture and to reflect the judicial exceptions, which are required at Step 2A/Prong two in order to be 101 eligible.
In the Remarks (page 18, para 2 through page 19, para 2), Applicant argues that “the combination of the above-recited limitations collectively constitutes a transformative technical solution that directly addresses deficiencies in existing drug-target interaction prediction systems. Taken together, these claim limitations recite a particular technological solution to the technical problem of completing sparse drug-target interaction matrices”. Applicant’s argument refers to Step 2A/Prong two or Step 2B in the 35 USC 101 analysis, relating to whether claims are integrated into a practical application due to “a transformative technical solution” or significantly more. In response, Applicant’s argument is not persuasive. Art exists that use similar matrix/tensor complete method as a way to predict drug-target interactions form existing information. Most importantly, there is no additional elements to apply, to capture and to reflect the judicial exceptions, which are required at Step 2A/Prong two in order to be 101 eligible. There is no additional elements qualify “significantly more” required at Step 2B to be 101 eligible.
In the Remarks (page 19, para 3 through page 20, para 2), Applicant argues that “new dependent claims tie in the prediction to a concrete outcome”. In response, Applicant’s argument is not persuasive. Newly added claims 31 and 32 both recite:
(New) The computer-implemented of claim 1 (or 18), further comprising:
generating, by the computer processor, from the one or more resulting interaction entries, interaction candidates for controlling activity of a biological target; and
selecting, by the computer processor, from the interaction candidates, a repositioned drug for a new therapeutic indication.
The limitations are drawn to no more than a prediction and an intended application. These are equivalent to a thinking, which is drawn to abstract ideas of mental processes. There are not “physical” process steps. It would be fair to say that “new dependent claims tie in the prediction and stays in computer”. The expected “controlling activity of a biological target”; and “a repositioned drug for a new therapeutic indication” are yet to be realized as additional elements. Again, there is no additional elements to apply, to capture and to reflect the judicial exceptions, which are required at Step 2A/Prong two in order to be 101 eligible.
In the Remarks (page 20, para 3 through page 20, para 2), Applicant argues that “the claims integrate any such exception into a practical application”. Applicant’s argument refers to Step 2B in the 35 USC 101 analysis, relating to whether claims recite significantly more. In response, Applicant’s argument is not persuasive. Because the argument “the claims nonetheless recite additional elements that amount to an inventive concept and thus recite significantly more than a judicial exception” (page 21, 1st para) is not supported by evidence. The additional elements we identified are drawn to input/out data and computer processors. They don’t amount to an inventive concept and thus recite significantly more than a judicial exception.
In the Remarks (page 21 para 3 through page 22, para 1), Applicant argues that “the claims recite processes that greatly improve the functioning of a computer”. Applicant’s argument refers to Step 2A/Prong two in the 35 USC 101 analysis. In response, Applicant’s argument is not persuasive.
The instant claims do embrace an abstract idea that is merely applied on a general purpose computer. Specifically, the claims still recite abstract data analysis involved in the matrix/tensor completion. The computer implemented analysis and modeling in these claims is defined only in terms of a functional description and a description the expected results. The computer is simply relied on for performing the mathematical operations and inputting/outputting data that would normally be performed by hand on a chalk board or with pen and paper.
As such, these claims only present a functional description of the results expected from a process using the actual algorithms or algorithmic procedures required to produce the functional results that are asserted in the claims. The recent CAFC decision of Internet Patents Corp. v. Active Networks (Fed. Cir. 2015) establishes that functional claiming at the point(s) of novelty for computer implemented invention fails under 35 USC 101 as being directed to an abstract idea. Further, in Eon Corp v. AT&T (Fed. Cir. 2015), the Federal Circuit has affirmed that Eon’s asserted patent claims are invalid for failing to specify the structure associated with a purely functional claim element. Upon review, neither the instant claims nor the instant disclosure provides the programming or algorithms required as the "structure" for a special purpose computer.
Therefore, taken as a whole, the instant claims are directed to an abstract idea that is merely applied on a general purpose computer. The claims do not delineate how the computer elements involved in the claims are improved. Rather, the recited steps themselves are generic and embrace only routine and conventional computer operations to be used in data analysis. Thus do not present anything
significantly more than the abstract idea to be “applied” using a generic computer with a processor. Thus the computer in the instant claims still only serves as the same computational work horse, which is a very well known, routine and conventional purpose that computers are used for in the related arts.
Contrary to appellant argument, the case of McRO is not analogous to the instant claims because the instant claims only involve a generic computer in the generation of an in silico realization of drugs and targets. Unlike McRO, there is no improvement to the generic computer of the instant claims beyond the conventional use as a computational workhorse running algorithmic based programming.
In the Remarks (page 22, 2nd para through last para), Applicant argues that “The claims recite other meaningful limitations beyond generally linking the use of the judicial exception to a particular technological environment”. In response, Applicant’s argument is not persuasive. As discussed above, it is generally true that the claims do not merely link a judicial exception to a particular technological environment. However, “the claims do not merely link a judicial exception to a particular technological environment” is not the only factor weighted in the 35 USC 101 analysis. The argued newly added claim 31 does not satisfy the Step 2B requirement in the 35 USC 101 analysis.
Newly added claims 31 and 32 both recite:
(New) The computer-implemented of claim 1 (or 18), further comprising:
generating, by the computer processor, from the one or more resulting interaction entries, interaction candidates for controlling activity of a biological target; and
selecting, by the computer processor, from the interaction candidates, a repositioned drug for a new therapeutic indication.
The limitations are drawn to no more than a prediction and an intended application. These are equivalent to thinking, which is drawn to abstract ideas of mental processes. There is nothing “physical” done here. It would be fair to say that “new dependent claims tie in the prediction and stays in computer”. The expected “controlling activity of a biological target”; and “a repositioned drug for a new therapeutic indication” are yet to be realized.
There is nothing “significantly more” than the judicial exceptions. Step 2B requirement is not matched.
Hence, the 101 rejection is maintained.
Claim Rejections - 35 USC § 103
Upon further consideration, this rejection is reinstated.
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.
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.
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.
Claims 1, 3-7, 10-17 and 31 are rejected under 35 U.S.C. 103 as being unpatentable over Wang et al.: ("Drug‐Target Interaction Prediction via Dual Laplacian Graph Regularized Matrix Completion." BioMed Research International 2018.1 (2018): 1425608. Newly cited), and further in view of Luo et al. ("A network integration approach for drug-target interaction prediction and computational drug repositioning from heterogeneous information." Nature communications 8.1 (2017): 573. Previously cited).
Claim 1 is interpreted as a computerized method for predicting drug-target interactions. Regarding claim 1, Wang provides (page 1, section Abstract lines 3-7) “we propose an effective computational model of dual Laplacian graph regularized matrix completion … DLGRMC transforms the task of drug-target interaction prediction into a matrix completion problem, in which the potential interactions between drugs and targets can be obtained based on the prediction scores after the matrix completion procedure” which teaches a computer-implemented method of performing completion in a drug-target interaction matrix by predicting drug and target interactions from coupled datasets.
Wang provides (page 3, col 2, 2nd para lines 1-2) “Matrix completion aims to fill in the missing entries of a partially observed matrix 𝑀” which teaches identifying incomplete entries in the drug-target interaction matrix.
Wang provides (page 4, col 1, 2nd para lines 6-14) “the drug-drug similarities and target-target similarities which have been demonstrated useful in previous works are not fully exploited to serve the matrix completion model. Thus, we believe that the two kinds of similarities can advantage the matrix completion model; of course, better DTIs prediction results can be expected. In this work, we present a new objective function through incorporation of the drug-drug similarities and target-target similarities into the standard matrix completion framework for DTIs prediction”, which teaches accessing a drug-drug matrix and a target-target matrix separate from the drug-target interaction matrix.
Wang provides (page 4, col 1, 2nd para center) “The optimization problem of
DLGRMC can be formulated as follows:
PNG
media_image1.png
144
598
media_image1.png
Greyscale
”, which teaches determining a matrix optimization function for use with the drug-drug matrix and the target-target matrix.
Wang provides (page 4, col 1, 2nd para last 14 lines) “𝐴 is an adjacency matrix with binary values which is defined to clearly describe the validated DTIs; i.e., if a specific drug 𝐷𝑖 is confirmed to be interacted with a target 𝑇𝑗, the entity 𝐴(𝑖, 𝑗) is assigned 1 or otherwise 0. Thus, the adjacency matrix 𝐴 is with size 𝑑 × 𝑡. Since 𝐴 is with 0 − 1 values, we use itself as the indicator matrix to indicate the indices of the observed DTIs. The forth term regularized by parameter 𝜆 constrains that drugs with similar chemical structure are more likely to be connected with similar targets and targets with similar genomic sequence similarity are forced to have interactions with similar drugs. 𝐷𝑆(𝑖, 𝑗) represents the chemical structure similarity between drugs 𝐷𝑖 and 𝐷𝑗, and 𝑇𝑆(𝑖, 𝑗) represents the genomic sequence similarity between targets 𝑇𝑖 and 𝑇𝑗”, which teaches using the matrix optimization function to access a subset of entries in the drug-target interaction matrix and entries in the drug-drug and target-target matrices based on the subset.
Wang provides (page 9, col 1, 1st para lines 2-5) “the potential interactions between drugs and targets can be obtained based on the prediction scores after the matrix completion procedure”, which teaches performing optimization until a predicted interaction entry is identified for completing an incomplete entry and updating the matrix.
Wang provides (page 4, col 2, 2nd para (including formula 5 and 6)) “we propose an alternative iterative algorithm to solve this problem based on Augmented Lagrange Multiplier (ALM) algorithm … Then the variables can be solved alternatively”, which teaches alternating optimization between drug-drug and target-target matrix optimizations.
Wang does not teach receiving subsequent drug and/or target data, and comparing to the updated matrix.
Luo provides (page 2, Fig. 1 legend) “DTINet then finds the best projection from drug space onto protein space, such that the projected feature vectors of drugs are
geometrically close to the feature vectors of their known interacting proteins. The projection matrix Z is learned to minimize the difference between the known interaction matrix P and XZYT (see Supplementary Note 1 for more details). After that, DTINet infers new interactions for a drug by sorting its target candidates”, which teaches receiving subsequent drug and/or target data, comparing to the updated matrix, and outputting resulting interaction entries.
Regarding claim 3, Wang does not teach interaction matrix entries indicate binding changing target behavior/function. Luo provides (page 1, section Abstract last 6 lines) “we experimentally validate the novel interactions between three drugs and the cyclooxygenase proteins predicted by DTINet, and demonstrate the new potential applications of these identified cyclooxygenase inhibitors in preventing inflammatory diseases. These results indicate that DTINet can provide a practically useful tool for integrating heterogeneous information to predict new drug–target interactions and repurpose existing drugs”, which teaches interaction matrix entries indicate binding changing target behavior/function.
Regarding claim 4, Wang provides (page 2, col 2, 2nd para lines 2-7) “four small-scale benchmark datasets which correspond to four different target protein types and a large-scale dataset are used in our experiments, including nuclear receptors (NRs), G protein coupled receptors (GPCRs), ion channels (ICs), enzymes (Es), and DrugBank (DB)”, which teaches the target entries of G-protein and enzymes.
Regarding claim 5, Wang does not teach target entries of side-effects. Luo provides (page 2, col 2, 2nd para) “previous works have incorporated pharmacological or phenotypic information, such as side-effects, transcriptional response data, drug–disease associations, public gene expression data and functional data for DTI prediction”, which teaches target entries of side-effects.
Regarding claim 6, Wang provides (page 4, col 1, 2nd para center) “The optimization problem of DLGRMC can be formulated as follows:
PNG
media_image1.png
144
598
media_image1.png
Greyscale
”, and later equation (4), which teaches the matrix optimization function is a fixed function.
Regarding claim 7, Wang provides (pages 3-4, connection para) “(1) is usually transformed to the following convex problem by relaxing the rank function into the
nuclear norm: … where ‖ ⋅ ‖∗ is the nuclear norm” and equation (2), which teaches the fixed function is a nuclear norm.
Regarding claim 10, Wang provides (page 4, col 1, 2nd para) “drugs with similar chemical structure are more likely to have connections with similar targets and targets with similar genomic sequence similarity are more likely to have interactions with similar drugs”, which teaches minimizing first distance for drug side and second distance for target side.
Regarding claim 11, Wang provides (page 1, section Abstract lines 7-8) “In DLGRMC, the drug pairwise chemical structure similarities and the target pairwise genomic sequence similarities are fully exploited to serve the matrix completion”, which teaches the drug-drug matrix is a drug similarity matrix.
Regarding claim 12, Wang does not teach receiving subsequent drug and/or target data. Luo provides (page 2, Fig. 1 legend) “With the learned compact features X and Y for drugs and proteins (i.e., each row in X and Y represents the feature vector of a drug and a protein, respectively), DTINet then finds the best projection from drug space onto protein space, such that the projected feature vectors of drugs are geometrically close to the feature vectors of their known interacting proteins. The projection matrix Z is learned to minimize the difference between the known interaction matrix P and XZYT (see Supplementary Note 1 for more details). After that, DTINet infers new interactions for a drug by sorting its target candidates”, which teaches receiving subsequent drug and/or target data, comparing to the updated matrix, and outputting resulting interaction entries”, which teaches the drug-drug matrix is a drug interaction matrix.
Regarding claim 13, Wang provides (page 1, section Abstract lines 7-8) “In DLGRMC, the drug pairwise chemical structure similarities and the target pairwise genomic sequence similarities are fully exploited to serve the matrix completion by using a dual Laplacian graph regularization term”, which teaches the target-target matrix is a target similarity matrix.
Regarding claim 14, Wang does not teach the target-target matrix is a target interaction matrix. Luo provides (page 2, Fig. 1 legend) “with the learned compact features X and Y for drugs and proteins (i.e., each row in X and Y represents the feature vector of a drug and a protein, respectively), DTINet then finds the best projection from drug space onto protein space, such that the projected feature vectors of drugs are geometrically close to the feature vectors of their known interacting proteins. The projection matrix Z is learned to minimize the difference between the known interaction matrix P and XZYT”, which teaches the target-target matrix is a target interaction matrix.
Regarding claim 15, Wang provides (page 3, col 2, 2nd para lines 1-2) “Matrix completion aims to fill in the missing entries of a partially observed matrix”, which teaches at least one of the drug-drug matrix and target-target matrix is incomplete or sparse.
Regarding claim 16, Wang does not teach the DTI matrix optimization function is scalable to a plurality of drug-drug and target-target matrices. Luo provides (page 10, col 2, 3rd para lines 1-6) “A future direction of our work is to include more heterogeneous network data in our framework. While we used only four domains (i.e., drugs, proteins, diseases and side-effects) of information in this work, we highlight that DTINet is a scalable framework in that more additional networks can be easily incorporated into the current prediction pipeline”, which teaches the DTI matrix optimization function is scalable to a plurality of drug-drug and target-target matrices.
Regarding claim 17, Wang provides (page 4, col 1, 2nd para) “𝐴 is an adjacency matrix with binary values which is defined to clearly describe the validated DTIs; i.e., if a specific drug 𝐷𝑖 is confirmed to be interacted with a target 𝑇𝑗, the entity 𝐴(𝑖, 𝑗) is assigned 1 or otherwise 0. Thus, the adjacency matrix 𝐴 is with size 𝑑 × 𝑡. Since 𝐴 is with 0 − 1 values, we use itself as the indicator matrix to indicate the indices of the observed DTIs”, which teaches identifying sparse or incomplete entries comprises identifying entries having a [0,1] value.
Regarding claim 31, Wang does not teach generating new drug candidates to control the cyclooxygenase (COX) protein target and repositioning the drugs for preventing inflammatory diseases. Luo provides (page 3, col 2, 1st para) “we have experimentally validated the new interactions predicted by DTINet between three drugs and the cyclooxygenase (COX) proteins that have not been reported in the literature (to the best of our knowledge), and demonstrated the potential novel applications of these drugs in preventing inflammatory diseases”, which teaches generating new drug candidates to control the cyclooxygenase (COX) protein target and repositioning the drugs for preventing inflammatory diseases.
It would have been prima facie obvious to combine Wang and Luo because both address the same problem of predicting unknown DTIs from incomplete/noisy heterogeneous biomedical data and both rely on side information for drugs and targets.
A person with ordinary skills in art would have been motivated to incorporate Luo’s broader heterogeneous drug/protein/disease/side-effect information and downstream prediction/output logic into Wang’s matrix-completion backbone to improve prediction coverage and utility, with a reasonable expectation of success because both references expressly report improved DTI prediction from such auxiliary information.
Claims 18-20, 22-23, 28-29 and 32 are rejected under 35 U.S.C. 103 as being unpatentable over Wang et al.: ("Drug‐Target Interaction Prediction via Dual Laplacian Graph Regularized Matrix Completion." BioMed Research International 2018.1 (2018): 1425608. Newly cited), and further in view of Wang et al. (“Predicting associations among drugs, targets and diseases by tensor decomposition for drug repositioning.” BMC bioinformatics 20 (2019): 1-19. Previously cited. Hereafter referred as “Wang_2”).
Claim 18 is interpreted as another computerized method for predicting drug-target interactions. Regarding claim 18, Wang provides (page 1, section Abstract lines 3-7) “we propose an effective computational model of dual Laplacian graph regularized matrix completion … DLGRMC transforms the task of drug-target interaction prediction into a matrix completion problem, in which the potential interactions between drugs and targets can be obtained based on the prediction scores after the matrix completion procedure” which teaches a computer-implemented method of performing completion in a drug-target interaction matrix by predicting drug and target interactions from coupled datasets.
Wang provides (page 3, col 2, 2nd para lines 1-2) “Matrix completion aims to fill in the missing entries of a partially observed matrix 𝑀” which teaches identifying incomplete entries in the drug-target interaction matrix.
Wang provides (page 2, col 2, 2nd para lines 2-7) “four small-scale benchmark datasets which correspond to four different target protein types and a large-scale dataset are used in our experiments, including nuclear receptors (NRs), G protein coupled receptors (GPCRs), ion channels (ICs), enzymes (Es), and DrugBank (DB)”, and (page 1, section Abstract lines 3-7) “we propose an effective computational model of dual Laplacian graph regularized matrix completion … DLGRMC transforms the task of drug-target interaction prediction into a matrix completion problem, in which the potential interactions between drugs and targets can be obtained based on the prediction scores after the matrix completion procedure”, which teaches receiving subsequent drug/target data and outputting resulting interaction entries.
Wang does not teach accessing a drug-drug tensor and a target-target tensor.
Wang_2 provides (page 2, col 2, last para) “we construct three-dimensional tensors representing DTD associations … We investigate the role of different additional information related to drugs and targets and the effects of other factors. Then we examine the ability of predicting new associations of the proposed method”, which teaches accessing a drug-drug tensor and a target-target tensor.
Wang_2 provides (page 2, col 2, last para) “decompose the tensors to derive latent factors and discover new predictions”, which teaches determining a tensor optimization function.
Wang_2 provides (page 3, col 1, 2nd para lines 11-14) “Each association tensor is decomposed together with different kinds of additional information separately, resulting in three factor matrices for drugs, targets and diseases”, which teaches accessing subset of DTI entries and plurality of slices of drug-drug and drug-target tensors.
Wang_2 provides (page 3, col 1, 2nd para lines 11-14) “Each association tensor is decomposed together with different kinds of additional information separately, resulting in three factor matrices for drugs, targets and diseases, respectively”, which teaches optimizing for each slice producing candidate entries.
Wang_2 provides (page 14, col 2, 2nd para lines 1-5) “we have proposed a novel framework for drug repositioning based on decomposing the triplet association tensors of drugs, targets and diseases. The proposed method is able to recover missing associations and predict unobserved triplet associations”, which teaches optimization on optimum candidate entries to identify predicted interaction entry and populate incomplete entry.
Regarding claim 19, Wang provides (page 9, col 2, last para) “a drug can be represented by its chemical structure or by its chemical response in different cells”, which teaches the drug-drug tensor slices differ in drug structure and chemical response.
Regarding claim 20, Wang does not teach the target-target tensor slices. Wang_2 provides (page 3, col 2, last para lines 3-4) “similarities of drugs and targets, pairwise associations, as well as drug-drug interactions (DDIs) and PPIs (Methods)”, which teaches the target-target tensor slices differ in binary interaction.
Regarding claim 22, Wang does not teach the tensor optimization function is a fixed function. Wang_2 provides (pages 15-16, the connection para) “we use three kinds of additional information, including (1) the similarity between the drugs and targets (Eq. 2), due to the widely accepted assumption that similar drugs might interact with similar targets, (2) DDIs and PPIs (Eq. 3), which reflect the functional patterns of drugs and targets, and (3) pairwise associations among drugs, targets and diseases (Eq. 4), since some pairwise associations are lost in the tensor construction stage”, which teaches the tensor optimization function is a fixed function.
Regarding claim 23, Wang provides (pages 3-4, connection para) “due to the fact that problem (1) is nonconvex and no efficient solution can be obtained, (1) is usually transformed to the following convex problem by relaxing the rank function into the nuclear norm”, which teaches the matrix optimization function is a nuclear norm. It would be obvious to transform the matrix optimization function of Wang into a tensor optimization function of nuclear norm.
Regarding claim 28, Wang does not teach the first dataset comprises drug interaction. Wang_2 provides (page 15, col 1, 2nd para) “if drug A interacts with target B, target B is associated with disease C, meanwhile, drug A is associated with disease C, then we believe that the triplet association among drug A, target B and disease C exists, represented by one at the corresponding entry in the tensor”, which teaches the first dataset comprises drug interaction and second dataset comprises drug target data.
Regarding claim 29, Wang does not teach the first and second tensor are each 3D tensors. Wang_2 provides (page 2, col 2, last para) “we construct three-dimensional tensors representing DTD associations”, which teaches the first and second tensor are each 3D tensors.
Regarding claim 32, Wang does not teach repositioning drug LEF away from pancreatic neoplasms. Wang_2 provides (page 14, col 1, last para) “it is found that some of these new discovered associations indicate new therapeutic uses of drugs, including the anti-tumor effect of Prednisolone, while others demonstrate adverse effects, for example, LEF might increase the risk for pancreatic neoplasms”, which teaches identifying new target by existing drug candidate from predicted interaction entries. Which also teaches repositioning drug LEF away from pancreatic neoplasms.
It would have been prima facie obvious to extend Wang’s coupled-completion framework into a tensor-based framework in view of Wang_2, because Wang_2 taught that tensor decomposition over drug-target-disease association recovers missing associations better than pairwise matrix approaches and makes direct use of richer muti-way relationship structure.
A person with ordinary skills in art would have been motivated to replace or supplement Wang’s pairwise side information matrices with multi-slice drug-side and target-side tensors, while preserving the same drug/target similarity regularization logic, in order to capture multiple representations/association types per entity and thereby improve prediction performance.
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