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 .
This action is responsive to the application filed 03/20/2024. Claims 1-19 are presented for examination.
Priority
Applicant’s claim for the benefit of a provisionally filed application, filed 03/20/2023, is acknowledged.
Information Disclosure Statement
The information disclosure statement (IDS) submitted 03/20/2024, has been considered by the examiner.
Claim Rejections - 35 USC § 101
35 U.S.C. 101 reads as follows:
Whoever invents or discovers any new and useful process, machine, manufacture, or composition of matter, or any new and useful improvement thereof, may obtain a patent therefore, subject to the conditions and requirements of this title.
Claims 5, 12 and 19 are rejected under 35 U.S.C. 101 because the claimed invention is directed to non-statutory subject matter. The claims do not fall within at least one of the four categories of patent eligible subject matter.
Claims 5, 12 and 19 recite a “computer-readable medium”. The specification did not define what is considered as a computer-readable medium. At best, the specification only provides the definition for a “non-transitory computer readable medium” (see spec, page 16, lines 1-10). When the specification is silent, the BRI of a CRM in view of the state of the art covers a signal per se. Carrier wave signal is non-statutory subject matter. Therefore, claims 5, 12 and 19 are directed to a non-statutory subject matter.
Claims 1-19 are rejected under 35 U.S.C. 101 because the claimed invention is directed to an abstract idea without significantly more.
Claim 1
Step 1: The claim recites “A computer-implemented method for training a machine learning model for predicting molecular property differences, the method comprising:”; therefore, it is directed to the statutory category of a process.
Step 2A Prong 1: The claim recites, inter alia:
creating a set of training data with the set of data: These limitations recite a mentally performable process with the aid of pen and paper of using observation and judgement to create a set of training data with the set of data.
creating a set of molecule pairs using each molecule of the set of training data: These limitations recite a mentally performable process with the aid of pen and paper of using observation and judgement to create a set of molecule pairs using each molecule of the set of training data.
generating a shared molecular representation of each pair of molecules in the set of training data: These limitations recite a mentally performable process with the aid of pen and paper of using judgement and evaluation to generate a shared molecular representation of each pair of molecules in the set of training data.
and for two molecules forming a molecule pair, predicting a property difference of molecular derivatization using the machine learning model as trained based on property differences of each pair of molecules: These limitations recite mathematical calculations similar to an act of calculating using mathematical methods to determine a variable or number per MPEP 2106.04(a)(2)(I)(C).
Thus, the claim recites a judicial exception.
Step 2A Prong 2: This judicial exception is not integrated into a practical application. The additional elements of the claim are as follows:
receiving a set of data including molecules, wherein each molecule of the set of data includes a molecular representation and a known absolute property value of each molecule in the set of data: These additional elements amount to insignificant extra-solution activity in the form of mere data gathering per MPEP § 2106.05(g).
training a machine learning model of an artificial intelligence (Al) system using the set of training data, wherein the set of training data includes the shared molecular representation and property difference of each pair of molecules in the set of training data: These additional elements recite only the idea of training a machine learning model of an AI system using the set of training data, wherein the set of training data includes the shared molecular representation and property difference of each pair of molecules in the set of training data and attempts to cover any implementation of the machine learning model training without any restriction as to the specific style or process of training, how the training data is used in the training process, and no details of the artificial intelligence system’s mechanisms, e.g. is the model a deep neural network, RNN, CNN, etc. Thus, these additional elements do not meaningfully limit the claim and do not integrate the judicial exception into a practical application because this type of recitation is equivalent to the words "apply it". See MPEP 2106.05(f).
Thus, the way in which the additional elements use or interact with the judicial exception do not integrate the judicial exception into a practical application.
Step 2B: The additional elements from Step 2A Prong 2 include adding words equivalent to "apply it" with the judicial exception and insignificant extra-solution activity of data gathering recited by “receiving a set of data including molecules, wherein each molecule of the set of data includes a molecular representation and a known absolute property value of each molecule in the set of data” which are well-understood routine and conventional activities similar to presenting offers and gathering statistics per MPEP 2106.05(d)(II). Thus, the additional elements, viewed individually or in combination, do not provide an inventive concept or otherwise amount to significantly more than the abstract idea itself. See MPEP § 2106.05.
Claim 2
Step 1: A process, as above.
Step 2A Prong 1: The claim recites the abstract ideas of claim 1 as well as, inter alia:
splitting the set of training data into a training set and a test set: These limitations recite a mentally performable process with the aid of pen and paper of using observation and judgement to split the set of training data into a training set and a test set.
Thus, the claim recites a judicial exception.
Step 2A Prong 2 & Step 2B: There are no additional elements recited so the claim does not provide a practical application and is not considered to be significantly more. As such, the claim is patent ineligible.
Claim 3
Step 1: A process, as above.
Step 2A Prong 1: The claim recites the abstract ideas of claim 2 as well as, inter alia:
generating a pair of molecules for at least one of the training set and the test set by cross merging a first molecule of the training set or the test set and a second molecule of the training set or the test set, wherein all possible molecule pairs of the training set and the test set are generated, wherein cross merging of the training set is limited to molecules of the training set, and wherein cross merging of the test set is limited to molecules of the test set: These limitations recite mathematical relationships similar to organizing information and manipulating information through mathematical correlations per MPEP 2106.04(a)(2)(I)(A)(iv).
Thus, the claim recites a judicial exception.
Step 2A Prong 2 & Step 2B: There are no additional elements recited so the claim does not provide a practical application and is not considered to be significantly more. As such, the claim is patent ineligible.
Claim 4
Step 1: A process, as above.
Step 2A Prong 1: The claim recites the abstract ideas of claim 1 as well as, inter alia:
concatenating a first molecular representation of a first molecule and a second molecular representation of a second molecule of each pair of molecules in the set of training data: These limitations recite mathematical relationships similar to organizing information and manipulating information through mathematical correlations per MPEP 2106.04(a)(2)(I)(A)(iv).
Thus, the claim recites a judicial exception.
Step 2A Prong 2 & Step 2B: There are no additional elements recited so the claim does not provide a practical application and is not considered to be significantly more. As such, the claim is patent ineligible.
Claim 5
Step 1: A process, as above.
Step 2A Prong 1: The claim recites the abstract ideas as the judicial exception of claim 1.
Step 2A Prong 2: This judicial exception is not integrated into a practical application. The additional elements of the claim are as follows:
A computer program product comprising program instructions stored on a machine-readable storage medium, wherein when the program instructions are executed by a computer processor, the program instructions cause the computer processor to execute the method: These additional elements are recited at a high level of generality and merely indicate a field of use or technological environment in which to apply a judicial exception, e.g. a computer program product comprising program instructions stored on a machine-readable storage medium, to a particular technological environment or field of use, e.g. wherein when the program instructions are executed by a computer processor, the program instructions cause the computer processor to execute the method. See MPEP 2106.05(h). Elements that use or interact with the judicial exception do not integrate the judicial exception into a practical application.
Thus, the way in which the additional elements use or interact with the judicial exception do not integrate the judicial exception into a practical application.
Step 2B: The additional elements from Step 2A Prong 2 include generally linking the use of the judicial exception to indicate a field of use or technological environment. Thus, the additional elements, viewed individually or in combination, do not provide an inventive concept or otherwise amount to significantly more than the abstract idea itself. See MPEP § 2106.05.
Claim 6
Step 1: The claim recites “A computer-implemented method for training a machine learning model for retrieving a compound with a desired characteristic from a set of data, the method comprising:”; therefore, it is directed to the statutory category of a process.
Step 2A Prong 1: The claim recites, inter alia:
creating a set of training data with the set of data: These limitations recite a mentally performable process with the aid of pen and paper of using observation and judgement to create a set of training data with the set of data.
creating a set of molecule pairs using each molecule of the set of training data: These limitations recite a mentally performable process with the aid of pen and paper of using observation and judgement to create a set of molecule pairs using each molecule of the set of training data.
generating a shared molecular representation of each pair of molecules in the set of molecule pairs: These limitations recite a mentally performable process with the aid of pen and paper of using judgement and evaluation to generate a shared molecular representation of each pair of molecules in the set of training data.
identifying a first compound of the set of training data based on a property of the identified compound: These limitations recite a mentally performable process with the aid of pen and paper of using observation and judgement to identify a first compound of the set of training data based on a property of the identified compound.
pairing the identified compound with each compound of a learning dataset, wherein the learning data set is based on the set of data: These limitations recite a mentally performable process with the aid of pen and paper of using judgement to pair the identified compound with each compound of a learning dataset.
for a pair of molecules from the learning dataset, predicting a property difference of the pair of molecules from the learning dataset using the machine learning model as trained based on property differences of each pair of molecules of the set of training data, wherein the pair of molecules include the identified compound: These limitations recite mathematical calculations similar to an act of calculating using mathematical methods to determine a variable or number per MPEP 2106.04(a)(2)(I)(C).
and adding a compound paired with the identified compound from the learning dataset to the training data set based on a property increase of the compound and the identified compound: These limitations recite a mentally performable process with the aid of pen and paper of using observation and judgement to add a compound paired with the identified compound from the learning dataset to the training data set based on a property increase of the compound and the identified compound.
Thus, the claim recites a judicial exception.
Step 2A Prong 2: This judicial exception is not integrated into a practical application. The additional elements of the claim are as follows:
receiving a set of data including molecules, wherein each molecule of the set of data includes a molecular representation and a known absolute property: These additional elements amount to insignificant extra-solution activity in the form of mere data gathering per MPEP § 2106.05(g).
training a machine learning model of an Al system using the set of training data, wherein the set of training data includes the shared molecular representation and respective property differences of each pair of molecules of the set of training data: These additional elements recite only the idea of training a machine learning model of an AI system using the set of training data, wherein the set of training data includes the shared molecular representation and respective property differences of each pair of molecules of the set of training data and attempts to cover any implementation of the machine learning model training without any restriction as to the specific style or process of training, how the training data is used for training, and no details of the artificial intelligence system’s mechanisms, e.g. is the model a deep neural network, RNN, CNN, etc. Thus, these additional elements do not meaningfully limit the claim and do not integrate the judicial exception into a practical application because this type of recitation is equivalent to the words "apply it". See MPEP 2106.05(f).
Thus, the way in which the additional elements use or interact with the judicial exception do not integrate the judicial exception into a practical application.
Step 2B: The additional elements from Step 2A Prong 2 include adding words equivalent to "apply it" with the judicial exception and insignificant extra-solution activity of data gathering recited by “receiving a set of data including molecules, wherein each molecule of the set of data includes a molecular representation and a known absolute property” which are well-understood routine and conventional activities similar to presenting offers and gathering statistics per MPEP 2106.05(d)(II). Thus, the additional elements, viewed individually or in combination, do not provide an inventive concept or otherwise amount to significantly more than the abstract idea itself. See MPEP § 2106.05.
Claim 7
Step 1: A process, as above.
Step 2A Prong 1: The claim recites the abstract ideas of claim 7 as well as, inter alia:
creating a second set of molecule pairs using each molecule of the set of training data, wherein the set of training data includes the added compound: These limitations recite a mentally performable process with the aid of pen and paper of using observation and judgement to create a second set of molecule pairs using each molecule of the set of training data.
Thus, the claim recites a judicial exception.
Step 2A Prong 2: This judicial exception is not integrated into a practical application. The additional elements of the claim are as follows:
and retraining the machine learning model using the set of training data, wherein the set of training data includes shared molecular representations and respective property differences of each pair of molecules of the second set of molecule pairs of the set of training data: These additional elements recite only the idea of training a machine learning model of an AI system using the set of training data and attempts to cover any implementation of the machine learning model training without any restriction as to the specific style or process of training, and no details of the artificial intelligence system’s mechanisms, e.g. is the model a deep neural network, RNN, CNN, etc. Thus, these additional elements do not meaningfully limit the claim and do not integrate the judicial exception into a practical application because this type of recitation is equivalent to the words "apply it". See MPEP 2106.05(f).
Thus, the way in which the additional elements use or interact with the judicial exception do not integrate the judicial exception into a practical application.
Step 2B: The additional elements from Step 2A Prong 2 include adding words equivalent to "apply it" with the judicial exception. Thus, the additional elements, viewed individually or in combination, do not provide an inventive concept or otherwise amount to significantly more than the abstract idea itself. See MPEP § 2106.05.
Claim 8
Step 1: A process, as above.
Step 2A Prong 1: The claim recites the abstract ideas as the judicial exception of claim 6.
Step 2A Prong 2: This judicial exception is not integrated into a practical application. The additional elements of the claim are as follows:
wherein the compound paired with the identified compound has a property improvement greater than other compounds paired with the identified compound in the learning dataset: These additional elements are recited at a high level of generality and merely indicate a field of use or technological environment in which to apply a judicial exception, e.g. the compound paired with the identified compound, to a particular technological environment or field of use, e.g. has a property improvement greater than other compounds paired with the identified compound in the learning dataset. See MPEP 2106.05(h). Elements that use or interact with the judicial exception do not integrate the judicial exception into a practical application.
Thus, the way in which the additional elements use or interact with the judicial exception do not integrate the judicial exception into a practical application.
Step 2B: The additional elements from Step 2A Prong 2 include generally linking the use of the judicial exception to indicate a field of use or technological environment. Thus, the additional elements, viewed individually or in combination, do not provide an inventive concept or otherwise amount to significantly more than the abstract idea itself. See MPEP § 2106.05.
Claim 9
Step 1: A process, as above.
Step 2A Prong 1: The claim recites the abstract ideas of claim 6 as well as, inter alia:
splitting the set of training data into one or more sets selected from the group consisting of: a training set, a test set, and the learning dataset: These limitations recite a mentally performable process with the aid of pen and paper of using observation and judgement to split the set of training data into one or more sets from the group consisting of a training, test, and learning dataset.
Thus, the claim recites a judicial exception.
Step 2A Prong 2 & Step 2B: There are no additional elements recited so the claim does not provide a practical application and is not considered to be significantly more. As such, the claim is patent ineligible.
Claim 10
Step 1: A process, as above.
Step 2A Prong 1: The claim recites the abstract ideas of claim 9 as well as, inter alia:
generating a pair of molecules for at least one of the training set, the test set, and the learning dataset by cross merging a first molecule and a second molecule of the training set, the test set, or the learning dataset, wherein all possible molecule pairs of the training set, the test set, or the learning dataset are generated and the learning set is cross merged with one molecule of the training set, wherein cross merging of the training set is limited to molecules of the training set, wherein cross merging of the test set is limited to molecules of the test set, and wherein cross merging of the learning set is limited to the one molecule of the training set and molecules of the learning set, wherein the one molecule of the training set includes a desired property value: These limitations recite mathematical relationships similar to organizing information and manipulating information through mathematical correlations per MPEP 2106.04(a)(2)(I)(A)(iv).
Thus, the claim recites a judicial exception.
Step 2A Prong 2 & Step 2B: There are no additional elements recited so the claim does not provide a practical application and is not considered to be significantly more. As such, the claim is patent ineligible.
Claim 11
Step 1: A process, as above.
Step 2A Prong 1: The claim recites the abstract ideas of claim 10 as well as, inter alia:
for a pair of molecules from an external dataset, predicting a property difference of the pair of molecules from the external dataset using the machine learning model as trained based on property differences of each pair of molecules of the set of training data: These limitations recite mathematical calculations similar to an act of calculating using mathematical methods to determine a variable or number per MPEP 2106.04(a)(2)(I)(C).
Thus, the claim recites a judicial exception.
Step 2A Prong 2 & Step 2B: There are no additional elements recited so the claim does not provide a practical application and is not considered to be significantly more. As such, the claim is patent ineligible.
Claim 12
Step 1: A process, as above.
Step 2A Prong 1: The claim recites the abstract ideas as the judicial exception of claim 6.
Step 2A Prong 2: This judicial exception is not integrated into a practical application. The additional elements of the claim are as follows:
A computer program product comprising program instructions stored on a machine- readable storage medium, wherein when the program instructions are executed by a computer processor, the program instructions cause the computer processor to execute the method: These additional elements are recited at a high level of generality and merely indicate a field of use or technological environment in which to apply a judicial exception, e.g. a computer program product comprising program instructions stored on a machine-readable storage medium, to a particular technological environment or field of use, e.g. wherein when the program instructions are executed by a computer processor, the program instructions cause the computer processor to execute the method. See MPEP 2106.05(h). Elements that use or interact with the judicial exception do not integrate the judicial exception into a practical application.
Thus, the way in which the additional elements use or interact with the judicial exception do not integrate the judicial exception into a practical application.
Step 2B: The additional elements from Step 2A Prong 2 include generally linking the use of the judicial exception to indicate a field of use or technological environment. Thus, the additional elements, viewed individually or in combination, do not provide an inventive concept or otherwise amount to significantly more than the abstract idea itself. See MPEP § 2106.05.
Claim 13
Step 1: The claim recites “A computer-implemented method for training a machine learning model for predicting which of a pair of molecules has an improved property value, the method comprising:”; therefore, it is directed to the statutory category of a process.
Step 2A Prong 1: The claim recites, inter alia:
creating a set of training data with the set of data: These limitations recite a mentally performable process with the aid of pen and paper of using observation and judgement to create a set of training data with the set of data.
creating a set of molecule pairs using each molecule of the set of training data: These limitations recite a mentally performable process with the aid of pen and paper of using observation and judgement to create a set of molecule pairs using each molecule of the set of training data.
filtering the set of training data based on a set of rules, wherein the filtered set of training data includes molecule pairs of the set of molecule pairs having at least one molecule with a property value improved compared to the other molecule: These limitations recite a mentally performable process with the aid of pen and paper of using judgement and evaluation to filter the set of training data based on a set of rules.
and for datapoints of a pair of molecules, predicting a property value improvement of molecular derivatization using the machine learning model as trained based on property differences of the datapoints, wherein the property value improvement indicates at least one molecule of the pair of molecules includes a property value greater than the other molecule: These limitations recite mathematical calculations similar to an act of calculating using mathematical methods to determine a variable or number per MPEP 2106.04(a)(2)(I)(C).
Thus, the claim recites a judicial exception.
Step 2A Prong 2: This judicial exception is not integrated into a practical application. The additional elements of the claim are as follows:
receiving a set of data including molecules, wherein each molecule of the set of data includes a molecular representation and a value selected from a group consisting of: a known exact absolute property value and a known bound absolute property value, wherein the known exact absolute property value and the known bound absolute property value are related to a property of each molecule of the set of data: These additional elements amount to insignificant extra-solution activity in the form of mere data gathering per MPEP § 2106.05(g).
training a machine learning model of an Al system using datapoints of the filtered set of training data, wherein the datapoints include molecular pairs of the filtered set of training data with shared representations, and wherein the datapoints include at least one selected from the group consisting of: bounded datapoints and exact regression datapoints: These additional elements recite only the idea of training a machine learning model of an AI system using datapoints of the filtered set of training data, wherein the datapoints include molecular pairs of the filtered set of training data with shared representations, and wherein the datapoints include at least one selected from the group consisting of: bounded datapoints and exact regression datapoints and attempts to cover any implementation of the machine learning model training without any restriction as to the specific style or process of training, how the datapoints are used in training, and no details of the artificial intelligence system’s mechanisms, e.g. is the model a deep neural network, RNN, CNN, etc. Thus, these additional elements do not meaningfully limit the claim and do not integrate the judicial exception into a practical application because this type of recitation is equivalent to the words "apply it". See MPEP 2106.05(f).
Thus, the way in which the additional elements use or interact with the judicial exception do not integrate the judicial exception into a practical application.
Step 2B: The additional elements from Step 2A Prong 2 include adding words equivalent to "apply it" with the judicial exception and insignificant extra-solution activity of data gathering recited by “receiving a set of data including molecules, wherein each molecule of the set of data includes a molecular representation and a value selected from a group consisting of: a known exact absolute property value and a known bound absolute property value, wherein the known exact absolute property value and the known bound absolute property value are related to a property of each molecule of the set of data” which are well-understood routine and conventional activities similar to presenting offers and gathering statistics per MPEP 2106.05(d)(II). Thus, the additional elements, viewed individually or in combination, do not provide an inventive concept or otherwise amount to significantly more than the abstract idea itself. See MPEP § 2106.05.
Claim 14
Step 1: A process, as above.
Step 2A Prong 1: The claim recites the abstract ideas of claim 13 as well as, inter alia:
removing, from the set of training data, molecular pairs of the set of training data with a property difference below a property difference threshold value: These limitations recite a mentally performable process with the aid of pen and paper of using observation and judgement to remove molecular pairs of the set of training data with a property difference below a property difference threshold value.
Thus, the claim recites a judicial exception.
Step 2A Prong 2 & Step 2B: There are no additional elements recited so the claim does not provide a practical application and is not considered to be significantly more. As such, the claim is patent ineligible.
Claim 15
Step 1: A process, as above.
Step 2A Prong 1: The claim recites the abstract ideas of claim 13 as well as, inter alia:
removing, from the set of training data, molecular pairs of the set of training data having a first molecule and a second molecule with equal property values: These limitations recite a mentally performable process with the aid of pen and paper of using observation and judgement to remove molecular pairs of the set of training data having a first molecule and a second molecule with equal property value.
Thus, the claim recites a judicial exception.
Step 2A Prong 2 & Step 2B: There are no additional elements recited so the claim does not provide a practical application and is not considered to be significantly more. As such, the claim is patent ineligible.
Claim 16
Step 1: A process, as above.
Step 2A Prong 1: The claim recites the abstract ideas of claim 13 as well as, inter alia:
removing, from the set of training data, molecular pairs of the set of training data when an improved property is unknown: These limitations recite a mentally performable process with the aid of pen and paper of using observation and judgement to remove molecular pairs of the set of training data when an improved property is unknown.
Thus, the claim recites a judicial exception.
Step 2A Prong 2 & Step 2B: There are no additional elements recited so the claim does not provide a practical application and is not considered to be significantly more. As such, the claim is patent ineligible.
Claim 17
Step 1: A process, as above.
Step 2A Prong 1: The claim recites the abstract ideas of claim 13 as well as, inter alia:
splitting the set of training data into a training set and a test set: These limitations recite a mentally performable process with the aid of pen and paper of using observation and judgement to split the set of training data into a training set and a test set.
Thus, the claim recites a judicial exception.
Step 2A Prong 2 & Step 2B: There are no additional elements recited so the claim does not provide a practical application and is not considered to be significantly more. As such, the claim is patent ineligible.
Claim 18
Step 1: A process, as above.
Step 2A Prong 1: The claim recites the abstract ideas of claim 17 as well as, inter alia:
generating a pair of molecules for at least one of the training set and the test set by cross merging a first molecule and a second molecule of the training set and the test set, wherein all possible molecule pairs of the training set and the test set are generated, wherein cross merging of the training set is limited to molecules of the training set, and wherein cross merging of the test set is limited to molecules of the test set: These limitations recite mathematical relationships similar to organizing information and manipulating information through mathematical correlations per MPEP 2106.04(a)(2)(I)(A)(iv).
Thus, the claim recites a judicial exception.
Step 2A Prong 2 & Step 2B: There are no additional elements recited so the claim does not provide a practical application and is not considered to be significantly more. As such, the claim is patent ineligible.
Claim 19
Step 1: A process, as above.
Step 2A Prong 1: The claim recites the abstract ideas as the judicial exception of claim 13.
Step 2A Prong 2: This judicial exception is not integrated into a practical application. The additional elements of the claim are as follows:
A computer program product comprising program instructions stored on a machine-readable storage medium, wherein when the program instructions are executed by a computer processor, the program instructions cause the computer processor to execute the method: These additional elements are recited at a high level of generality and merely indicate a field of use or technological environment in which to apply a judicial exception, e.g. a computer program product comprising program instructions stored on a machine-readable storage medium, to a particular technological environment or field of use, e.g. wherein when the program instructions are executed by a computer processor, the program instructions cause the computer processor to execute the method. See MPEP 2106.05(h). Elements that use or interact with the judicial exception do not integrate the judicial exception into a practical application.
Thus, the way in which the additional elements use or interact with the judicial exception do not integrate the judicial exception into a practical application.
Step 2B: The additional elements from Step 2A Prong 2 include generally linking the use of the judicial exception to indicate a field of use or technological environment. Thus, the additional elements, viewed individually or in combination, do not provide an inventive concept or otherwise amount to significantly more than the abstract idea itself. See MPEP § 2106.05.
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.
Claims 1-5 are rejected under 35 U.S.C. 102(a)(1) as being anticipated by Tynes et al. ("Pairwise Difference Regression: A Machine Learning Meta-algorithm for Improved Prediction and Uncertainty Quantification in Chemical Search", J. Chem. Inf. Model. 61 (8): 3846-3857, ACM) (Year: 2021), hereafter Tynes.
Regarding independent claim 1, Tynes teaches a computer-implemented method for training a machine learning model for predicting molecular property differences ([Abstract & Methods] discusses training a machine learning model to compute pairwise difference regression for property values of pairs of molecules), the method comprising:
receiving a set of data including molecules, wherein each molecule of the set of data includes a molecular representation and a known absolute property value of each molecule in the set of data ([Fig. 1 & Methods] discusses receiving a set of molecules, each molecule including a molecular representation and a set of reference properties);
creating a set of training data with the set of data ([Fig. 1 & Methods] discusses partitioning the set of data into training data sets);
creating a set of molecule pairs using each molecule of the set of training data ([Fig. 1 & Methods] discusses establishing pairs from each molecule of the training data set);
generating a shared molecular representation of each pair of molecules in the set of training data ([Eq. 2 & Methods] discusses building a shared representation by concatenating the two features and the differences between the features);
training a machine learning model of an artificial intelligence (Al) system using the set of training data, wherein the set of training data includes the shared molecular representation and property difference of each pair of molecules in the set of training data ([Eq. 3-6] discusses using the set of training data to train a learning algorithm or architecture, the training data includes the difference of each pair and the shared molecular representation);
and for two molecules forming a molecule pair, predicting a property difference of molecular derivatization using the machine learning model as trained based on property differences of each pair of molecules ([Eq. 5-10] discusses predicting property differences using the machine learning model as trained on the pairs of molecules).
Regarding dependent claim 2, Tynes teaches the invention as claimed in claim 1, including splitting the set of training data into a training set and a test set ([Computational Experiments & Results] discusses splitting the data items into both a training set and a test set).
Regarding dependent claim 3, Tynes teaches the invention as claimed in claim 2, including generating a pair of molecules for at least one of the training set and the test set by cross merging a first molecule of the training set or the test set and a second molecule of the training set or the test set, wherein all possible molecule pairs of the training set and the test set are generated, wherein cross merging of the training set is limited to molecules of the training set, and wherein cross merging of the test set is limited to molecules of the test set ([Methods & Fig. 1] discusses generating n^2 points from n points for the training and test sets by cross merging a molecule with each of the available second molecules, and this is done for each possible molecule of the molecules in both training and test sets).
Regarding dependent claim 4, Tynes teaches the invention as claimed in claim 1, including concatenating a first molecular representation of a first molecule and a second molecular representation of a second molecule of each pair of molecules in the set of training data ([Eq. 2 & Methods] discusses building a shared representation by concatenating each representation within the pair of molecules, each pair’s two features and the differences between the features).
Regarding dependent claim 5, Tynes teaches the invention as claimed in claim 1, including a computer program product comprising program instructions stored on a machine- readable storage medium, wherein when the program instructions are executed by a computer processor, the program instructions cause the computer processor to execute the method as claimed in claim 1 ([Implementation & Notes] discusses the method is stored as source code to be implemented on any computer; thus, the program instructions, when executed by a processor, cause the processor to execute the method within the source code).
Claim Rejections - 35 USC § 103
In the event the determination of the status of the application as subject to AIA 35 U.S.C. 102 and 103 (or as subject to pre-AIA 35 U.S.C. 102 and 103) is incorrect, any correction of the statutory basis (i.e., changing from AIA to pre-AIA ) for the rejection will not be considered a new ground of rejection if the prior art relied upon, and the rationale supporting the rejection, would be the same under either status.
The following is a quotation of 35 U.S.C. 103 which forms the basis for all obviousness rejections set forth in this Office action:
A patent for a claimed invention may not be obtained, notwithstanding that the claimed invention is not identically disclosed as set forth in section 102, if the differences between the claimed invention and the prior art are such that the claimed invention as a whole would have been obvious before the effective filing date of the claimed invention to a person having ordinary skill in the art to which the claimed invention pertains. Patentability shall not be negated by the manner in which the invention was made.
Claims 6-12 are rejected under 35 U.S.C. 103 as being unpatentable over Tynes et al. ("Pairwise Difference Regression: A Machine Learning Meta-algorithm for Improved Prediction and Uncertainty Quantification in Chemical Search", J. Chem. Inf. Model. 61 (8): 3846-3857, ACM) (Year: 2021), hereafter Tynes, in view of Reker et al. ("Active-learning strategies in computer-assisted drug discovery", Drug Discovery Today Volume 20, Number 4, ScienceDirect) (Year: 2015), hereafter Reker, and further in view of Zhang et al. ("Siamese Neural Networks for Regression: Similarity-Based Pairing and Uncertainty Quantification", Research Square) (Year: 2022), hereafter Zhang.
Regarding independent claim 6, Tynes teaches a method comprising:
receiving a set of data including molecules, wherein each molecule of the set of data includes a molecular representation and a known absolute property ([Fig. 1 & Methods] discusses receiving a set of molecules, each molecule including a molecular representation and a set of reference properties, which are absolute);
creating a set of training data based on the set of data ([Fig. 1 & Methods] discusses partitioning the set of data into training data sets);
creating a set of molecule pairs using each molecule of the set of training data ([Fig. 1 & Methods] discusses establishing pairs from each molecule of the training data set);
generating a shared molecular representation of each pair of molecules in the set of molecule pairs ([Eq. 2 & Methods] discusses building a shared representation by concatenating the two features and the differences between the features);
training a machine learning model of an Al system using the set of training data, wherein the set of training data includes the shared molecular representation and respective property differences of each pair of molecules of the set of training data ([Eq. 3-6] discusses using the set of training data to train a learning algorithm or architecture, the training data includes the difference between each pair and the shared molecular representation);
for a pair of molecules from the learning dataset, predicting a property difference of the pair of molecules from the learning dataset using the machine learning model as trained based on property differences of each pair of molecules of the set of training data ([Eq. 5-10] discusses predicting property differences using the machine learning model as trained on the pairs of molecules).
Tynes does not explicitly teach identifying a first compound of the set of training data based on a property of the identified compound; pairing the identified compound with each compound of a learning dataset, wherein the learning data set is based on the set of data; for a pair of molecules from the learning dataset, predicting a property difference of the pair of molecules from the learning dataset using the machine learning model as trained based on property differences of each pair of molecules of the set of training data, wherein the pair of molecules include the identified compound; and adding a compound paired with the identified compound from the learning dataset to the training data set based on a property increase of the compound and the identified compound.
However, Reker teaches active-learning strategies in a computer setting, including identifying the compounds within a chemical space that have the best chance of success ([Abstract & Introduction] discusses identifying the best or most promising compound according to a standard, which according to broadest reasonable interpretation, constitutes its known property); and adding a readout back to a training dataset ([Abstract & Introduction] discusses iteratively adding the identified readout, which is the compound, to the training data set based on the performance, which constitutes a property increase of the compound).
Because Tynes teaches receiving a set of data including molecules, creating a set of training data based on the set of data, creating a set of molecule pairs using each molecule of the set of training data, generating a shared molecular representation of each pair of molecules in the set of molecule pairs, training a machine learning model of an AI system using the set of training data, and predicting a property difference of the pair of molecules from the learning dataset using the machine learning model as trained based on property differences of each pair of molecules of the set of training data; and Reker teaches identifying a first compound of the set of training data based on a property of the identified compound, and adding a compound paired with the identified compound from the learning dataset to the training data set based on a property increase of the compound and the identified compound, accordingly, it would have been obvious to one of ordinary skill in the art, before the effective filing date of the claimed invention, to incorporate identifying a first compound of the set of training data based on a property of the identified compound, and adding a compound paired with the identified compound from the learning dataset to the training data set based on a property increase of the compound and the identified compound as taught by Reker into Tynes’ method, with a reasonable expectation of success, to teach receiving a set of data including molecules, wherein each molecule of the set of data includes a molecular representation and a known absolute property; creating a set of training data based on the set of data; creating a set of molecule pairs using each molecule of the set of training data; generating a shared molecular representation of each pair of molecules in the set of molecule pairs; training a machine learning model of an Al system using the set of training data, wherein the set of training data includes the shared molecular representation and respective property differences of each pair of molecules of the set of training data; identifying a first compound of the set of training data based on a property of the identified compound; for a pair of molecules from the learning dataset, predicting a property difference of the pair of molecules from the learning dataset using the machine learning model as trained based on property differences of each pair of molecules of the set of training data; and adding a compound paired with the identified compound from the learning dataset to the training data set based on a property increase of the compound and the identified compound. This combination would have been motivated by the desire to implement cost-effective active learning approaches and outline a potential for molecule or drug-discovery (Reker [Abstract]).
The combination of Tynes and Reker does not explicitly teach pairing the identified compound with each compound of a learning dataset, wherein the learning data set is based on the set of data; for a pair of molecules from the learning dataset, predicting a property difference of the pair of molecules from the learning dataset using the machine learning model as trained based on property differences of each pair of molecules of the set of training data, wherein the pair of molecules include the identified compound.
However, Zhang teaches a method for generating compound pairs, wherein an identified compound is paired with the other compounds of a dataset ([Methods] discusses pairing a compound with each compound of a second dataset, and the second dataset Is based on the set of data); and predicting a property difference of the pair including the identified compound ([Pg. 5-6] discusses training the model on, and outputting, the pairwise property difference of molecules from the learning dataset, which includes the identified compound).
Because the combination of Tynes and Reker teaches receiving a set of data including molecules, wherein each molecule of the set of data includes a molecular representation and a known absolute property, creating a set of training data based on the set of data, creating a set of molecule pairs using each molecule of the set of training data, generating a shared molecular representation of each pair of molecules in the set of molecule pairs, training a machine learning model of an Al system using the set of training data, wherein the set of training data includes the shared molecular representation and respective property differences of each pair of molecules of the set of training data, identifying a first compound of the set of training data based on a property of the identified compound, for a pair of molecules from the learning dataset, predicting a property difference of the pair of molecules from the learning dataset using the machine learning model as trained based on property differences of each pair of molecules of the set of training data, and adding a compound paired with the identified compound from the learning dataset to the training data set based on a property increase of the compound and the identified compound; and Zhang teaches pairing the identified compound with each compound of a learning dataset, and predicting a property difference of the pair of molecules from the learning dataset using the machine learning model as trained based on property differences of each pair of molecules of the set of training data, wherein the pair of molecules include the identified compound, accordingly, it would have been obvious to one of ordinary skill in the art, before the effective filing date of the claimed invention, to incorporate pairing the identified compound with each compound of a learning dataset, and predicting a property difference of the pair of molecules from the learning dataset using the machine learning model as trained based on property differences of each pair of molecules of the set of training data, wherein the pair of molecules include the identified compound as taught by Zhang into the combination of Tynes and Reker’s method, with a reasonable expectation of success, to teach receiving a set of data including molecules, wherein each molecule of the set of data includes a molecular representation and a known absolute property; creating a set of training data based on the set of data; creating a set of molecule pairs using each molecule of the set of training data; generating a shared molecular representation of each pair of molecules in the set of molecule pairs; training a machine learning model of an Al system using the set of training data, wherein the set of training data includes the shared molecular representation and respective property differences of each pair of molecules of the set of training data; identifying a first compound of the set of training data based on a property of the identified compound; pairing the identified compound with each compound of a learning dataset, wherein the learning data set is based on the set of data; for a pair of molecules from the learning dataset, predicting a property difference of the pair of molecules from the learning dataset using the machine learning model as trained based on property differences of each pair of molecules of the set of training data, wherein the pair of molecules include the identified compound; and adding a compound paired with the identified compound from the learning dataset to the training data set based on a property increase of the compound and the identified compound. This combination would have been motivated by the desire for the method to result in a better prediction performance consistently (Zhang [Abstract]).
Regarding dependent claim 7, the combination of Tynes, Reker, and Zhang teaches the invention as claimed in claim 6, including:
creating a second set of molecule pairs using each molecule of the set of training data, wherein the set of training data includes the added compound (Tynes [Computational Experiments & Results & Fig. 6] discusses experiments re-pairing the training set after each newly selected molecule is added);
and retraining the machine learning model using the set of training data, wherein the set of training data includes shared molecular representations and respective property differences of each pair of molecules of the second set of molecule pairs of the set of training data (Tynes [Computational Experiments & Results & Fig. 7] discusses retraining the pairwise regressor at each iteration of the pair set; thus, the model is retrained on the training data, which has been built on shared molecular representations and respective property differences of each pair of molecules of the second set of molecule pairs of the set of training data).
Regarding dependent claim 8, the combination of Tynes, Reker, and Zhang teaches the invention as claimed in claim 6, including wherein the compound paired with the identified compound has a property improvement greater than other compounds paired with the identified compound in the learning dataset (Reker [Exploration versus exploitation] discusses comparing all candidates and the selected compound has the best improvement, or predicted activity, when compared to other compounds).
Regarding dependent claim 9, the combination of Tynes, Reker, and Zhang teaches the invention as claimed in claim 6, including splitting the set of training data into one or more sets selected from the group consisting of: a training set, a test set, and the learning dataset (Tynes [Computational Experiments & Results] discusses splitting the data items into both a training set and a test set).
Regarding dependent claim 10, the combination of Tynes, Reker, and Zhang teaches the invention as claimed in claim 9, including generating a pair of molecules for at least one of the training set, the test set, and the learning dataset by cross merging a first molecule and a second molecule of the training set, the test set, or the learning dataset, wherein all possible molecule pairs of the training set, the test set, or the learning dataset are generated and the learning set is cross merged with one molecule of the training set, wherein cross merging of the training set is limited to molecules of the training set, wherein cross merging of the test set is limited to molecules of the test set, and wherein cross merging of the learning set is limited to the one molecule of the training set and molecules of the learning set, wherein the one molecule of the training set includes a desired property value (Tynes [Methods & Fig. 1] discusses generating n^2 points from n points for the training and test sets by cross merging a molecule with each of the available second molecules, and this is done for each possible molecule of the molecules in both training and test sets).
Regarding dependent claim 11, the combination of Tynes, Reker, and Zhang teaches the invention as claimed in claim 10, including for a pair of molecules from an external dataset, predicting a property difference of the pair of molecules from the external dataset using the machine learning model as trained based on property differences of each pair of molecules of the set of training data (Tynes [Table 2] discusses taking external datasets and using the method to predict a property difference of the pair of molecules from the external dataset using the same trained machine learning model).
Regarding dependent claim 12, the combination of Tynes, Reker, and Zhang teaches the invention as claimed in claim 6, including a computer program product comprising program instructions stored on a machine- readable storage medium, wherein when the program instructions are executed by a computer processor, the program instructions cause the computer processor to execute the method as claimed in claim 6 (Tynes [Implementation & Notes] discusses the method is stored as source code to be implemented on any computer; thus, the program instructions, when executed by a processor, cause the processor to execute the method within the source code).
Claims 13 & 17-19 are rejected under 35 U.S.C. 103 as being unpatentable over Tynes et al. ("Pairwise Difference Regression: A Machine Learning Meta-algorithm for Improved Prediction and Uncertainty Quantification in Chemical Search", J. Chem. Inf. Model. 61 (8): 3846-3857, ACM) (Year: 2021), hereafter Tynes, in view of Argawal et al. ("Ranking Chemical Structures for Drug Discovery: A New Machine Learning Approach", J. Chem. Inf. Model., Vol. 50, No. 5, ACM) (Year: 2010), hereafter Argawal.
Regarding independent claim 13, Tynes teaches a method comprising:
receiving a set of data including molecules, wherein each molecule of the set of data includes a molecular representation and a value selected from a group consisting of: a known exact absolute property value and a known bound absolute property value, wherein the known exact absolute property value and the known bound absolute property value are related to a property of each molecule of the set of data ([Fig. 1 & Methods] discusses receiving a set of molecules, each molecule including a molecular representation and a set of reference properties, which constitutes known exact absolute property values);
creating a set of training data with the set of data ([Fig. 1 & Methods] discusses partitioning the set of data into training data sets);
creating a set of molecule pairs using each molecule of the set of training data ([Fig. 1 & Methods] discusses establishing pairs from each molecule of the training data set);
filtering the set of training data based on a set of rules ([Uncertainty Quantification Performance] discusses filtering the training data set based on rules of uncertainty, excluding points with high uncertainty values);
training a machine learning model of an Al system using datapoints of the filtered set of training data, wherein the datapoints include molecular pairs of the filtered set of training data with shared representations, and wherein the datapoints include at least one selected from the group consisting of: bounded datapoints and exact regression datapoints ([Eq. 3-6 & Computational Experiments and Results] discusses using the set of training data to train a learning algorithm or architecture using datapoints of the filtered set of training data, the training data includes the shared molecular representation and are exact regression datapoints);
and for datapoints of a pair of molecules, predicting a property value improvement of molecular derivatization using the machine learning model as trained based on property differences of the datapoints ([Eq. 5-10] discusses calculating predictions of molecular derivatization using the machine learning model as trained based on property differences).
Tynes does not explicitly teach filtering the set of training data based on a set of rules, wherein the filtered set of training data includes molecule pairs of the set of molecule pairs having at least one molecule with a property value improved compared to the other molecule; and for datapoints of a pair of molecules, predicting a property value improvement of molecular derivatization using the machine learning model as trained based on property differences of the datapoints, wherein the property value improvement indicates at least one molecule of the pair of molecules includes a property value greater than the other molecule.
However, Argawal teaches ranking chemical structures using machine learning wherein training data is filtered based on property value ([Introduction] discusses the end goal is to rank or filter the training data, comprising compounds in which the property value is ranked higher than others); and predicting property value improvement of a molecule having a property value greater than the other molecule ([Abstract & Introduction] discusses the system is used for prediction of compounds and then further ranking value of datapoints of molecules to determine which values are higher).
Because Tynes teaches receiving a set of data including molecules, wherein each molecule of the set of data includes a molecular representation and a value selected from a group consisting of: a known exact absolute property value and a known bound absolute property value, wherein the known exact absolute property value and the known bound absolute property value are related to a property of each molecule of the set of data, creating a set of training data with the set of data, creating a set of molecule pairs using each molecule of the set of training data, filtering the set of training data based on a set of rules, training a machine learning model of an Al system using datapoints of the filtered set of training data, wherein the datapoints include molecular pairs of the filtered set of training data with shared representations, and wherein the datapoints include at least one selected from the group consisting of: bounded datapoints and exact regression datapoints, and for datapoints of a pair of molecules, predicting a property value improvement of molecular derivatization using the machine learning model as trained based on property differences of the datapoints; and Argawal teaches the filtered set of training data includes molecule pairs of the set of molecule pairs having at least one molecule with a property value improved compared to the other molecule, and the property value improvement indicates at least one molecule of the pair of molecules includes a property value greater than the other molecule, accordingly, it would have been obvious to one of ordinary skill in the art, before the effective filing date of the claimed invention, to substitute filtering the set of training data based on pairs having at least one molecule with a property value improved compared to the other molecule, and incorporate predicting property value improvement of a molecule having a property value greater than the other molecule as taught by Argawal into Tynes’ system, with a reasonable expectation of success, to teach receiving a set of data including molecules, wherein each molecule of the set of data includes a molecular representation and a value selected from a group consisting of: a known exact absolute property value and a known bound absolute property value, wherein the known exact absolute property value and the known bound absolute property value are related to a property of each molecule of the set of data; creating a set of training data with the set of data; creating a set of molecule pairs using each molecule of the set of training data; filtering the set of training data based on a set of rules, wherein the filtered set of training data includes molecule pairs of the set of molecule pairs having at least one molecule with a property value improved compared to the other molecule; training a machine learning model of an Al system using datapoints of the filtered set of training data, wherein the datapoints include molecular pairs of the filtered set of training data with shared representations, and wherein the datapoints include at least one selected from the group consisting of: bounded datapoints and exact regression datapoints; and for datapoints of a pair of molecules, predicting a property value improvement of molecular derivatization using the machine learning model as trained based on property differences of the datapoints, wherein the property value improvement indicates at least one molecule of the pair of molecules includes a property value greater than the other molecule. This combination would have been motivated by the desire to optimize performance of the method on large datasets or libraries (Argawal [Abstract]).
Regarding dependent claim 17, the combination of Tynes and Argawal teaches the invention as claimed in claim 13, including splitting the set of training data into a training set and a test set (Tynes [Computational Experiments and Results] discusses splitting the data items into both a training set and a test set).
Regarding dependent claim 18, the combination of Tynes and Argawal teaches the invention as claimed in claim 17, including generating a pair of molecules for at least one of the training set and the test set by cross merging a first molecule and a second molecule of the training set and the test set, wherein all possible molecule pairs of the training set and the test set are generated ,wherein cross merging of the training set is limited to molecules of the training set, and wherein cross merging of the test set is limited to molecules of the test set (Tynes [Methods & Fig. 1] discusses generating n^2 points from n points for the training and test sets by cross merging a molecule with each of the available second molecules, and this is done for each possible molecule of the molecules in both training and test sets).
Regarding dependent claim 19, the combination of Tynes and Argawal teaches the invention as claimed in claim 13, including a computer program product comprising program instructions stored on a machine- readable storage medium, wherein when the program instructions are executed by a computer processor, the program instructions cause the computer processor to execute the method as claimed in claim 13 (Tynes [Implementation & Notes] discusses the method is stored as source code to be implemented on any computer; thus, the program instructions, when executed by a processor, cause the processor to execute the method within the source code).
Claim 14 is rejected under 35 U.S.C. 103 as being unpatentable over Tynes, in view of Argawal, as applied in claim 13, and further in view of Stumpfe et al. ("Evolving Concept of Activity Cliffs", ACS Omega, 4, 14360-14368) (Year: 2019), hereafter Stumpfe.
Regarding dependent claim 14, the combination of Tynes and Argawal teaches the invention as claimed in claim 13, including filtering the set of training data based on a set of rules (Tynes [Uncertainty Quantification Performance] discusses filtering the training data set based on rules of uncertainty, excluding points with high uncertainty values; Argawal [Introduction] discusses the end goal is to rank or filter the training data, comprising compounds in which the property value is ranked higher than others).
The combination of Tynes and Argawal does not explicitly teach removing, from the set of training data, molecular pairs of the set of training data with a property difference below a property difference threshold value.
However, Stumpfe teaches a method for comparing compounds and removing, or filtering out, pairs that have a difference below a threshold ([Fig. 3 & Conclusions] discusses setting a threshold of property difference, and if below that threshold they are flagged; thus, the pairs below the threshold would be removed from the set of training data).
Because the combination of Tynes and Argawal teaches filtering the set of training data based on a set of rules; and Stumpfe teaches comparing compounds and removing, or filtering out, pairs that have a difference below a threshold, accordingly, it would have been obvious to one of ordinary skill in the art, before the effective filing date of the claimed invention, to incorporate comparing compounds and removing, or filtering out, pairs that have a difference below a threshold as taught by Stumpfe into the combination of Tynes and Argawal’s method, with a reasonable expectation of success, to teach removing, from the set of training data, molecular pairs of the set of training data with a property difference below a property difference threshold value. This combination would have been motivated by the desire to restrict the set of training data to only pairs of molecules with a relative, larger property difference (Stumpfe [Sec. 4]).
Claim 15 is rejected under 35 U.S.C. 103 as being unpatentable over Tynes, in view of Argawal, as applied in claim 13, and further in view of Krüger ("Integration and analysis of protein evolutionary relationships and small molecule bioactivity data", University of Cambridge) (Year: 2013), hereafter Krüger.
Regarding dependent claim 15, the combination of Tynes and Argawal teaches the invention as claimed in claim 13, including filtering the set of training data based on a set of rules (Tynes [Uncertainty Quantification Performance] discusses filtering the training data set based on rules of uncertainty, excluding points with high uncertainty values; Argawal [Introduction] discusses the end goal is to rank or filter the training data, comprising compounds in which the property value is ranked higher than others).
The combination of Tynes and Argawal does not explicitly teach removing, from the set of training data, molecular pairs of the set of training data having a first molecule and a second molecule with equal property values.
However, Krüger teaches a method of comparison where pairs of a set are removed if they have an equal property value ([Sec. 4.2] discusses measuring potency differences between pairs and removing a pair, or excluding it, from the set if they were equal).
Because the combination of Tynes and Argawal teaches filtering the set of training data based on a set of rules; and Krüger teaches removing, from the set of training data, molecular pairs of the set of training data having a first molecule and a second molecule with equal property values, accordingly, it would have been obvious to one of ordinary skill in the art, before the effective filing date of the claimed invention, to incorporate removing, from the set of training data, molecular pairs of the set of training data having a first molecule and a second molecule with equal property values as taught by Krüger into the combination of Tynes and Argawal’s method, with a reasonable expectation of success, to teach removing, from the set of training data, molecular pairs of the set of training data having a first molecule and a second molecule with equal property values. This combination would have been motivated by the desire to restrict the set of training data to only pairs with a noticeable property difference and reduce the amount of redundant value (Krüger [Sec. 4.2]).
Claim 16 is rejected under 35 U.S.C. 103 as being unpatentable over Tynes, in view of Argawal, as applied in claim 13, and further in view of Arany et al. ("SparseChem: Fast and accurate machine learning model for small molecules", ESAT STADIUS, arXiv) (Year: 2022), hereafter Arany.
Regarding dependent claim 16, the combination of Tynes and Argawal teaches the invention as claimed in claim 13, including filtering the set of training data based on a set of rules (Tynes [Uncertainty Quantification Performance] discusses filtering the training data set based on rules of uncertainty, excluding points with high uncertainty values; Argawal [Introduction] discusses the end goal is to rank or filter the training data, comprising compounds in which the property value is ranked higher than others).
The combination of Tynes and Argawal does not explicitly teach removing, from the set of training data, molecular pairs of the set of training data when an improved property is unknown.
However, Arany teaches a machine learning model for molecules, comprising an upper and lower bound for the set of data, wherein the data is excluded if the property value is unknown ([Sec. 2] discusses if a predicted property loss is above a bound, it is set to 0, which constitutes being unknown due to being larger than the upper bound and is effectively being filtered out or removed from affecting the rest of the set of training data).
Because the combination of Tynes and Argawal teaches filtering the set of training data based on a set of rules; and Arany teaches removing, from the set of training data, molecular pairs of the set of training data when an improved property is unknown, accordingly, it would have been obvious to one of ordinary skill in the art, before the effective filing date of the claimed invention, to incorporate removing, from the set of training data, molecular pairs of the set of training data when an improved property is unknown as taught by Arany into the combination of Tynes and Argawal’s method, with a reasonable expectation of success, to teach removing, from the set of training data, molecular pairs of the set of training data when an improved property is unknown. This combination would have been motivated by the desire to censor values that are unknown (Arany [Sec. 2]).
Conclusion
The prior art made of record and not relied upon is considered pertinent to applicant's disclosure.
Kramer et al. ("Matched Molecular Pair Analysis: Significance and the Impact of Experimental Uncertainty", J. Med. Chem. 2014, 57, 3786-3802, ACM) (Year: 2014) ([Introduction] During the past years, matched molecular pair analysis (MMPA) has become a standard tool for the extraction of medicinal chemistry knowledge from large databases. The basic idea of MMPA is to search large chemical databases for sets of molecular pairs that are linked by identical chemical transformations. Predictions about the effect of the transformations are made based on an analysis of the past distribution of differences in biochemical or biophysical properties. The most promising modifications can then be used to prioritize synthesis and subsequent testing.
Any inquiry concerning this communication or earlier communications from the examiner should be directed to RILEY S ACOSTA whose telephone number is (571)272-8714. The examiner can normally be reached Monday-Thursday 6am-4pm.
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, Jennifer N Welch can be reached at (571)272-7212. 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.
/RILEY S ACOSTA/Examiner, Art Unit 2143
/JENNIFER N WELCH/Supervisory Patent Examiner, Art Unit 2143
Prosecution Insights