DETAILED ACTION
Applicant's response, filed 5/11/2026, has been fully considered. The following rejections and/or objections are either reiterated or newly applied. They constitute the complete set presently being applied to the instant application.
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 .
Claim Status
Claims 31-33, 35-38, 40, 44-50 are pending.
Claims 34 and 41-43 are cancelled.
Claims 31-33, 35-38, 40, 44-50 are rejected.
Withdrawn Rejections/Objections
The rejection of claim 43 under 35 U.S.C. §112, Second Paragraph, in the Office action mailed 2/9/2026 is withdrawn in view of the amendments filed 5/11/2026.
The rejection of claims 31-33, 35-38, 40, and 44-48 under 35 U.S.C. §101, in the Office action mailed 2/9/2026 is withdrawn in view of the amendments filed 5/11/2026.
Claim Rejections - 35 USC § 112
Response to Amendment
In view of applicant’s amendments to the claims previous rejections under 35 U.S.C. 112 for written description of claim 43 is withdrawn.
The following is a quotation of the first paragraph of 35 U.S.C. 112(a):
(a) IN GENERAL.—The specification shall contain a written description of the invention, and of the manner and process of making and using it, in such full, clear, concise, and exact terms as to enable any person skilled in the art to which it pertains, or with which it is most nearly connected, to make and use the same, and shall set forth the best mode contemplated by the inventor or joint inventor of carrying out the invention.
The following is a quotation of the first paragraph of pre-AIA 35 U.S.C. 112:
The specification shall contain a written description of the invention, and of the manner and process of making and using it, in such full, clear, concise, and exact terms as to enable any person skilled in the art to which it pertains, or with which it is most nearly connected, to make and use the same, and shall set forth the best mode contemplated by the inventor of carrying out his invention.
Claims 49 and 50 are rejected under 35 U.S.C. 112(a) or 35 U.S.C. 112 (pre-AIA ), first paragraph, as failing to comply with the written description requirement. The claim(s) contains subject matter which was not described in the specification in such a way as to reasonably convey to one skilled in the relevant art that the inventor or a joint inventor, or for applications subject to pre-AIA 35 U.S.C. 112, the inventor(s), at the time the application was filed, had possession of the claimed invention. Both claims 49 and 50 are directed to a CRM and system respectively, however, the final step is one of outputting information that “causes the administering of the chemical product” and there is nothing within the description that describes a system or CRM component capable of administering the chemical product. As such the claims fail to comply the written description requirement.
Claim Rejections - 35 USC § 101
Response to Amendment
In response to applicant’s amendments to the claims, previous rejections under 35 U.S.C. 101 are withdrawn.
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 49 and 50 are rejected under 35 U.S.C. 101 because the claimed invention is directed to abstract ideas without significantly more. The claims recite a CRM and system for scoring the stability of chemical compounds produced in batch in order to determine a predicted preservation state at a particular time point for the chemical product. The judicial exception is not integrated into a practical application because while claims 49 and 50 attempt to integrate the exception into a practical application, said application is either generically recited computer elements that do not add a meaningful limitation to the abstract idea or it is insignificant extra solution activity and simply implementing the abstract idea on a computer. The claims do not include additional elements that are sufficient to amount to significantly more than the judicial exception because the computer elements only store and retrieve information in memory as well as perform basic calculations that are known to be well-understood, routine and conventional computer functions as recognized by the decisions listed in MPEP § 2105.05(d).
Framework with which to Analyze Subject Matter Eligibility:
Step 1: Are the claims directed to a category of statutory subject matter (a process, machine, manufacture, or composition of matter)? [See MPEP § 2106.03]
Claims are directed to statutory subject matter, specifically a CRM (claim 49) and a system (claim 50).
Step 2A Prong One: Do the claims recite a judicially recognized exception, i.e., an abstract idea, a law of nature, or a natural phenomenon? [See MPEP § 2106.04(a)]
The claims herein recite abstract ideas, mental processes and mathematical concepts.
With respect to the Step 2A Prong One evaluation, the instant claims are found herein to recite abstract ideas that fall into the grouping of mental processes and mathematical concepts.
Claims 49 and 50: Determining modified experimental stability data, selecting a chemical stability model, processing the model input, determining quality scores, selecting the stability model in part on the quality scores, determining a concentration, and determining the instance is acceptable for administration, are processes of identifying, calculating comparing/contrasting, and evaluating that can be done via pen and paper or within the human mind and are therefore abstract ideas, specifically mental processes. The modified experimental stability comprising the specified measurements is merely further limiting the data itself which is an abstract idea, specifically a mental process.
Step 2A Prong Two: If the claims recite a judicial exception under prong one, then is the judicial exception integrated into a practical application? [See MPEP § 2106.04(d) and MPEP § 2106.05(a)-(c) & (e)-(h)]
Because the claims do recite judicial exceptions, direction under Step 2A Prong Two provides that the claims must be examined further to determine whether they integrate the abstract ideas into a practical application.
The following claims recite the following additional elements in the form of non-abstract elements:
Claims 49 and 50: Receiving a temperature profile and model input, training the chemical stability model, and generating an output are insignificant extra solution activities, specifically necessary data gathering and necessary data outputting (See Mayo, 566 U.S. at 79, 101 USPQ2d at 1968; OIP Techs., Inc. v. Amazon.com, Inc., 788 F.3d 1359, 1363, 115 USPQ2d 1090, 1092-93 (Fed. Cir. 2015) (presenting offers and gathering statistics amounted to mere data gathering), Performing clinical tests on individuals to obtain input for an equation, In re Grams, 888 F.2d 835, 839- 40; 12 USPQ2d 1824, 1827-28 (Fed. Cir. 1989) and Determining the level of a biomarker in blood, Mayo, 566 U.S. at 79, 101 USPQ2d at 1968. See also PerkinElmer, Inc. v. Intema Ltd., 496 Fed. App'x 65, 73, 105 USPQ2d 1960, 1966 (Fed. Cir. 2012) (assessing or measuring data derived from an ultrasound scan, to be used in a diagnosis)) [See MPEP § 2106.05(g)]. A non-transitory computer storage media, instructions, computers, system, and storage devices are all generic and nonspecific elements of a computer that do not improve the functioning of any computer or technology described herein [See MPEP § 2106.04(d)(1) and MPEP § 2106.05(d)]. Generating an output that causes administering the instance of the chemical product is merely intended use as no structure is recited within either claim that is capable of providing such administration (See In re Otto, 312 F.2d 937, 938, 136 USPQ 458, 459 (CCPA 1963)) [See MPEP § 2111.02].
Step 2B: If the claims do not integrate the judicial exception, do the claims provide an inventive concept? [See MPEP 2106.05]
Because the additional claim elements do not integrate the abstract ideas into a practical application, the claims are further examined under Step 2B, which evaluates whether the additional elements, individually and in combination, amount to significantly more than the judicial exception itself by providing an inventive concept.
The claims do not recite additional elements that are sufficient to amount to significantly more than the judicial exception because the claims recite additional elements that are generic, conventional or nonspecific. These additional elements include:
The additional elements of receiving a model input (Conventional: Li et al., and Liu et al., each receive data from by downloading the COMDECOM dataset), receiving a temperature profile (Conventional: Specification – page 12, line 23 Conventional method for aggregating data and page 12, line 18 Conventional computer storage for said data), training the chemical stability model (Conventional: Conventional: Li et al., and Liu et al., each train their respective models (Methods Sections)), and generating an output are insignificant extra solution activities, specifically mere data gathering and necessary data outputting (See Mayo, 566 U.S. at 79, 101 USPQ2d at 1968; OIP Techs., Inc. v. Amazon.com, Inc., 788 F.3d 1359, 1363, 115 USPQ2d 1090, 1092-93 (Fed. Cir. 2015) (presenting offers and gathering statistics amounted to mere data gathering), Performing clinical tests on individuals to obtain input for an equation, In re Grams, 888 F.2d 835, 839-40; 12 USPQ2d 1824, 1827-28 (Fed. Cir. 1989) and Determining the level of a biomarker in blood, Mayo, 566 U.S. at 79, 101 USPQ2d at 1968. See also PerkinElmer, Inc. v. Intema Ltd., 496 Fed. App'x 65, 73, 105 USPQ2d 1960, 1966 (Fed. Cir. 2012) (assessing or measuring data derived from an ultrasound scan, to be used in a diagnosis)) [See MPEP § 2106.05(g)]. Therefore, taken both individually and as a whole, the additional elements do not amount to significantly more than the judicial exception by providing an inventive concept.
The additional elements of a non-transitory computer storage media, instructions, computers, system, and storage devices are all generic and nonspecific elements of a computer that are well-understood, routine and conventional within the art and therefore do not improve the functioning of any computer or technology described therein (Receiving or transmitting data over a network, e.g., using the Internet to gather data, Symantec, 838 F.3d at 1321, 120 USPQ2d at 1362 (utilizing an intermediary computer to forward information), Performing repetitive calculations, Flook, 437 U.S. at 594, 198 USPQ2d at 199 (recomputing or readjusting alarm limit values), and Storing and retrieving information in memory, Versata Dev. Group, Inc. v. SAP Am., Inc., 793 F.3d 1306, 1334, 115 USPQ2d 1681, 1701 (Fed. Cir. 2015)) [See MPEP § 2106.05(d)(II)]. Therefore, taken both individually and as a whole, the additional elements do not amount to significantly more than the judicial exception by providing an inventive concept.
Therefore, claims 49 and 50, when the limitations are considered individually and as a whole, are rejected under 35 U.S.C. 101 as being directed to non-statutory subject matter.
Response to Arguments
Applicant’s arguments, see page 2 of the Remarks, filed 5/11/2026, with respect to the rejection of claims 31-38 and 40-48 under 35 U.S.C. 101 have been fully considered and are persuasive. The rejection of claims 31-38 and 40-48 has been withdrawn.
Applicant’s arguments, see page 2 of Remarks, filed 5/11/2026, with respect to claims 49 and 50 have been fully considered and are not persuasive.
Applicant asserts similarity to Example 39 in so far as the limitations are not practically performable in the human mind. However, Example 39 is directed to facial detection and computer vision, specifically the manipulation/interpretation of data for which the human mind cannot process. The instant application recites data such as temperature, pressure, time, and light exposure, all of which along with structure and machine learning algorithms are performable within the human mind.
Additionally, applicant asserts an improvement to technology through the improvement to “accurate and reliable determination of a product’s preservation state”. However, applicant’s assertion is to an improvement to the judicial exception, the “determining of a preservation state”, not the additional elements of the system, CRM, or process and MPEP 2106.05(a) states It is important to note, the judicial exception alone cannot provide the improvement.
Finally, the administration step is not performable within the meets and bounds of the system and CRM described as there is nothing within the description that provides for a system/CRM that administers the chemical product. Therefore, the claim is not directed to a practical application.
Claim Rejections - 35 USC § 103
Response to Amendment
In view of applicant’s amendments to the claims, previous rejections under 35 U.S.C. 103 are withdrawn and new ground of rejections are set forth below.
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.
Claims 31-38, 40-43, and 48-50 are rejected under 35 U.S.C. 103 as being unpatentable over Liu et al. (Journal of Computationally Aided Molecular Design (2014) 941-950; previously cited) in view of Li et al. (Journal of Chemical Information and Modeling (2019) 1044-1049; previously cited), Clenet et al. (Journal of Pharmaceutical Sciences (2014) 3055-3064; newly cited), and Waterman et al. (International Journal of Pharmaceutics (2005) 101-25; newly cited).
Claim 31 is directed to a method of predicting chemical stability using model input of chemical stability attributes, training a model and selecting a model to produce quality scores which can then be used to determine chemical stability.
Claim 49 is directed to a computer implemented method of predicting chemical stability using model input of chemical stability attributes, training a model and selecting a model to produce quality scores which can then be used to determine chemical stability.
Claim 50 is directed to a computer implemented method of predicting chemical stability using model input of chemical stability attributes, training a model and selecting a model to produce quality scores which can then be used to determine chemical stability.
Liu et al. teaches on page 942, column 1, paragraph 2 “The COMDECOM data contain 12,810 structurally diverse compounds and their stability data” and on page 942, column 1, paragraph 1 “based upon the COMDECOM data measured from DMSO/H2O solutions stored at 50 C for 105 days”, reading on receive a model input defining an input temperature and an input duration of time. Li et al. teaches on page 1045, column 1, paragraph 5 “COMDECOM data consists of chemical structures and compound purity data measured at days 0, 14, 35, and 105”, Figure 7 on page 946 provides a “flow chart for the rule-embedded naive Bayesian classifier”, and on page 941, column 2, paragraph 1 “Stability is affected by a number of factors: chemical features, solvent types, temperature, humidity, freeze–thaw resistance, storage time, etc.”, the last of which would render obvious the use of such information in the prediction of stability, reading on receiving a temperature profile that defines, for each time interval in a sequence of time intervals preceding a current time point, a respective temperature at which an instance of a chemical product was stored during the time interval, processing the model input in accordance with the set of model parameters of the chemical stability model to generate a model output that comprises a predicted concentration of an ingredient of the chemical product resulting from storing the chemical product at temperatures within a tolerance of the input temperature for the input duration of time. Furthermore Liu et al. teaches on page 942, column 2, paragraph 1 “The compound purities and the stability indices were measured at 50 degrees C for 0, 14, 35, and 105 days. Here, we did a careful analysis of the problems in the dataset. The records containing erroneous purity profiles, such as a compound concentration ct at time t that is higher than ct–1 by +5 %, are removed”, reading on determining modified experimental stability data for the chemical product based on the temperature profile, wherein the modified experimental stability data excludes experimental measurements relating to at least one temperature higher than any temperature included in the temperature profile. Liu et al. teaches on page 946, column 1, paragraph 2 “To further validate the performance of the classifier, additional tests were executed by dividing the COMDECOM dataset (containing 9,746 compounds) into two groups: the training set (about 2/3 of the compounds), and the testing set (about 1/3 of the compounds)”, reading on training the chemical stability model on the modified experimental stability data, using a numerical optimization technique, to determine trained values of the set of model parameters of the chemical stability model. Liu et al. teaches on page 946, column 2, paragraph 2 “In this work, the bayesian score ranges from 0 to 1 and 0.5 is the cutoff value to distinguish the unstable or stable compounds”, reading on determining one or more quality scores for the chemical stability model based on the trained values of the set of model parameters of the chemical stability model and determining a predicted concentration of the ingredient of the instance of the chemical product using the chemical stability model selected for the chemical product. Liu et al. further teaches on page 942, column 1, paragraph 1 “The prediction will be run in the backend and the unstability score will be provided and the predicted potential reactive site will be highlighted”, reading on generating a quality score for the chemical stability model.
Liu et al. does not teach multiple models or the selection of a single model from a group of models.
Li et al. teaches in the abstract “In this paper, we report DeepChemStable, a model employing an attention-based graph convolution network based on the COMDECOM data”, on page 1044, column 1, paragraph 2 “COMDECOM data contain structurally diverse compounds whose stabilities were measured experimentally in a mixture of DMSO and H2O”, and on page 1046, column 1, paragraph 1 “As shown in Figure 2, instead of simply combining all learned fragment features with a global pooling step after several graph convolution layers, an attention layer is applied to capture the different importance of fragments in determination of stability. The attention mechanism captures the influence on an individual fragment of other fragments” along with the cost functions which denote y-hat as the prediction generated for the fragment features, which in view of Li et al. from above reads on generating a quality score for the chemical stability model based on an accuracy of predicted attribute values generated by the chemical stability model in accordance with the trained values of the set of model parameters of the chemical stability model, wherein the accuracy of predicted attribute values generated by the chemical stability model is measured.
Liu et al. and Liu et al. do not teach the administration of the chemical product in response to determining said product’s stability.
Clenet et al. teaches in the abstract “We have used a protein-based vaccine, a live virus vaccine, and an experimental adjuvant to evaluate the utility of an advanced kinetic modeling approach for stability prediction. The modeling approach uses a systematic and simple procedure for the selection of the most appropriate kinetic equation to describe the degradation rate of compounds subjected to accelerated conditions. One-step and two-step reactions with unlimited combinations of kinetic models were screened for the three products under evaluation. The most appropriate mathematical model for a given product was chosen based on the values of residual sum of squares and the weight parameter w. A relatively simple n-th order kinetic model best fitted the degradation of an adjuvanted protein vaccine with a prediction error lower than 10%”, on page 3055, column 1, paragraphs 1-3 “The stability of vaccines is a critical factor influencing their worldwide distribution and has a major impact on vaccine quality, potency, and storage conditions…The thermal stability of vaccines can be evaluated through numerous methods that look at physicochemical or biological changes in a vaccine upon exposure to elevated temperatures. There are essentially two approaches to studying the thermal stability of vaccines: temperature ramping experiments and accelerated stability studies under isothermal conditions… Temperature ramping experiments involve monitoring changes in the biophysical properties of a vaccine while temperature is increased at a given heating rate. In this case, thermal stability is evaluated via monitoring decomposition extent and/or some thermodynamic parameters such as the enthalpy and free energy. Since thermal ramping experiments can be completed within a few hours, these studies are broadly used in vaccine formulation development and screening of stabilizing conditions. Apart from thermodynamics, it is noteworthy to mention that kinetics also governs the proteins stability. High kinetic stability results in a low denaturation rate of the protein and thus long-term stability”, and page 3056, column 2, paragraph 1 “The stability of PhtD was evaluated as a function of time by RP-HPLC after desorption from the aluminum adjuvant as previously described.30 By this technique, only the peak area of intact PhtD is integrated and used for the calculation of the concentration of intact protein. Therefore, this provided a direct measurement of protein degradation induced by elevated temperatures. Stability data are presented as the concentration of intact protein as a function of time”, reading on receiving a temperature profile that defines, for each time interval in a sequence of time intervals preceding a current time point, a respective temperature at which an instance of a chemical product was stored during the time interval; determining modified experimental stability data for the chemical product based on the temperature profile, wherein the modified experimental stability data excludes experimental measurements relating to at least one temperature higher than any temperature included in the temperature profile, and determining that the instance of the chemical product is acceptable for administration based on determining that the predicted concentration is within a predefined threshold range.
Waterman et al. teaches in the abstract “Methods are discussed, with the appropriate caveats, for accelerated aging of liquid and solid dosage forms, including small and large molecule active pharmaceutical ingredients. In particular, this review covers general thermal methods, as well as accelerated aging methods appropriate to oxidation, hydrolysis, reaction with reactive excipient impurities, photolysis and protein denaturation”, and on page 101, column 1, paragraph 1 “In the development of pharmaceutical dosage forms, one of the persistent challenges is assuring acceptable stability. While classically stability refers to the ability to withstand loss of a chemical due to decomposition, in the pharmaceutical world, the term “stability” more often refers to the storage time allowed before any degradation product in the dosage form achieves a sufficient level to represent a risk to the patient. Based on this time, the expiration date (shelf-life) of a product is determined. The allowable level of any given impurity will depend on the dose and likelihood of toxicity; however, for most drugs, the allowable levels of a single impurity permissible without explicit toxicological clinical testing are generally well less than 1% based on the drug”, which renders obvious administering the instance of the chemical product to a subject based on determining that the instance of the chemical product is acceptable for administration, as the determination of stability of a pharmaceutical would then render obvious the administration of said pharmaceutical. Furthermore, in view of the teachings of Liu et al. and Li et al., this also reads on determining that the instance of the chemical product is acceptable for administration based on determining that the predicted concentration is within a predefined threshold range, as the previous cited references were looking at the stability of the chemical instance, and Waterman et al. is specifically tying stability to decomposition and concentration in pharmaceuticals.
It would have been obvious at the time of filing to modify the teachings of Li et al. and Liu et al. for the overall method of claims 31, 49, and 50, with the teachings of Waterman et al. for the predicting of pharmaceutical stability as both are chemical compounds with the same features of stability affecting both, i.e. pharmaceuticals are merely a sub-category of chemicals. Additionally, Li et al. and Liu et al. would be obvious to combine as they are specifically from the same lab, on the same subject, using the same data, and similar methods. It would also be obvious to a person skilled within the art to give users a choice from multiple models to predict chemical stability based upon each model’s performance on the data given, especially as these two papers come from the same lab on the same topic each attempting to address issues with modeling chemical stability. Furthermore, it would have been obvious to combine these with the teachings of Clenet et al. for predicting stability as a function of temperature, where temperature is measured as concentration of the active compound, as they found “results indicate that the long-term stability of vaccines can be predicted with great accuracy”. One would have had a reasonable expectation of success using the teachings of each as again two of these papers come from the same lab on the same topic each attempting to address issues with modeling chemical stability, while the third and fourth are merely focusing on a subcategory of the data. Therefore, it would be obvious to one with ordinary skill in the art to incorporate the teachings of each and to be successful.
Claim 32 is directed to the method of claim 31 but further specifies that the chemical product be a drug.
Liu et al. teaches on page 944, column 1, paragraph 5 “A scaffold-based classification approach (SCA) was employed to compare the structural diversities of the COMDECOM, Drug Bank, and WDI databases”.
Li et al. teaches in the abstract “In the drug discovery process, unstable compounds in storage can lead to false positive or false negative bioassay conclusions”.
Waterman et al. teaches in the abstract “Methods are discussed, with the appropriate caveats, for accelerated aging of liquid and solid dosage forms, including small and large molecule active pharmaceutical ingredients”.
All prior art pieces teach the chemical product being a drug.
Claim 33 is directed to the method of claim 32 and therefore claim 31, but further specifies that the chemical product be a vaccine.
Liu et al. teaches on page 944, column 1, paragraph 5 “A scaffold-based classification approach (SCA) was employed to compare the structural diversities of the COMDECOM, Drug Bank, and WDI databases”.
Li et al. teaches in the abstract “In the drug discovery process, unstable compounds in storage can lead to false positive or false negative bioassay conclusions”.
It would be obvious to a person skilled in the art that the use of a stability model for chemical stability of drugs would also be useful for the chemical stability of vaccines, as they are in fact a form of drug.
Claim 34 is directed to the method of claim 33 and therefore claim 31, but further specifies that the attribute of the chemical product comprise a concentration of an active component of the chemical product.
Li et al. teaches on page 1045, column 1, paragraph 5 “COMDECOM data consists of chemical structures and compound purity data measured at days 0, 14, 35, and 105”, reading on wherein the attribute of the chemical product comprises a concentration of an active component in the chemical product.
Claim 35 is directed to the method of claim 31 but further specifies that the experimental stability data for the chemical product comprise a collection of measurements from physical experiments.
Li et al. teaches on page 1044, column 1, paragraph 2 “COMDECOM data contain structurally diverse compounds whose stabilities were measured experimentally in a mixture of DMSO and H2O”, reading on wherein the set of experimental stability data for the chemical product comprises a collection of experimental measurements generated by way of one or more physical experiments.
Claim 36 is directed to the method of claim 35 and therefore claim 31 but further specifies that the experimental measure be an experimental temperature, duration of time and attribute value for the chemical product.
Liu et al. teaches on page 941, column 2, paragraph 1 “Stability is affected by a number of factors: chemical features, solvent types, temperature, humidity, freeze–thaw resistance, storage time, etc.”
Li et al. teaches on page 1044, column 1, paragraph 2 “COMDECOM data contain structurally diverse compounds whose stabilities were measured experimentally in a mixture of DMSO and H2O. The compound purity was monitored at 0-, 14-, 35-, and 105-day time points, and models were built from these data”.
It would be obvious to a person skilled in the art that measurements of stability over time would include a temperature and therefore be an attribute (stability), with both corresponding time and temperature data.
Claim 37 is directed to the method of claim 36 and thus claim 31, but further specifies optimizing the model parameters based upon minimizing an error between the experimental and the predicted values.
Li et al. teaches on page 1044, column 1, paragraph 2 “COMDECOM data contain structurally diverse compounds whose stabilities were measured experimentally in a mixture of DMSO and H2O”, and on page 1046, column 1, paragraph 1 “As shown in Figure 2, instead of simply combining all learned fragment features with a global pooling step after several graph convolution layers, an attention layer is applied to capture the different importance of fragments in determination of stability. The attention mechanism captures the influence on an individual fragment of other fragments” along with the cost functions which denote y-hat as the prediction generated for the fragment features, reading on wherein for each chemical stability model in the library of chemical stability models, training the chemical stability model comprises, for each experimental measurement in the collection of experimental measurements: optimizing values of the set of model parameters of the chemical stability model to minimize an error between: (i) the experimental attribute value specified by the experimental measurement, and (ii) a predicted attribute value generated by the chemical stability model by processing the experimental temperature and the experimental duration of time specified by the experimental measurement.
Claim 38 is directed to the method of claim 31 but further specifies that trained model parameters be compared to a target range of values for the model parameter.
Li et al. teaches on page 1046, column 2, paragraph 2 “Table 3 lists the hyperparameters to be optimized and their search range”, and on page 1047, column 1, paragraph 1 “the DeepChemStable model generated four graph convolution layers and output atom embedding size of 200. The best learning rate is 10−4. The best positive rate is 1.5, and the best L2 regulation parameter is 10−4”, reading on generating a quality score for the chemical stability model by comparing a trained value of a model parameter of the chemical stability model to a target range of values for the model parameter of the chemical stability model.
Claim 40 is directed to the method of claim 31 but further specifies that the quality score for the model be based on a confidence interval associated with a trained value of a model parameter.
Li et al. teaches on page 1046, column 2, paragraph 2 “Table 3 lists the hyperparameters to be optimized and their search range”, and on page 1047, column 1, paragraph 1 “the DeepChemStable model generated four graph convolution layers and output atom embedding size of 200. The best learning rate is 10−4. The best positive rate is 1.5, and the best L2 regulation parameter is 10−4”.
It would be obvious that any model’s performance would be associated the values of the trained hyperparameters, if any. Therefore, a confidence interval, which dictates the number of times you know a value will fall within the range specified, of a hyperparameter, would necessarily be associated with the performance of said model, and be an obvious way of determining model selection.
Claim 44 is directed to training the chemical stability model on the modified data, wherein the modified data is data that has gone through filtering based on the temperature.
Liu et al. teaches on page 946, column 1, paragraph 2 “To further validate the performance of the classifier, additional tests were executed by dividing the COMDECOM dataset (containing 9,746 compounds) into two groups: the training set (about 2/3 of the compounds), and the testing set (about 1/3 of the compounds)”. Furthermore, Liu et al. teaches on page 942, column 2, paragraph 1 “The compound purities and the stability indices were measured at 50 degrees C for 0, 14, 35, and 105 days. Here, we did a careful analysis of the problems in the dataset. The records containing erroneous purity profiles, such as a compound concentration ct at time t that is higher than ct–1 by +5 %, are removed”, which reads on training, for each chemical stability model in the library of chemical stability models, the chemical stability model on the modified experimental stability data, to determine trained values of the set of model parameters of the chemical stability model.
Claim 48 is directed to the method of claim 31 but further specifies that to identify the best model, said model will have the highest value of a quality score from among the models.
It would have been obvious to a person skilled in the art to identify a model from a group of models as the best model via a score, given a choice of multiple models. Additionally, both Li et al. and Liu et al. provide a score of each model in terms of their AUC to determine precision/recall rates.
Claim 45 is rejected under 35 U.S.C. 103 as being unpatentable over Li et al. (Journal of Chemical Information and Modeling (2019) 1044-1049; previously cited), Liu et al. (Journal of Computationally Aided Molecular Design (2014) 941-950; previously cited), Clenet et al. (Journal of Pharmaceutical Sciences (2014) 3055-3064; newly cited), and Waterman et al. (International Journal of Pharmaceutics (2005) 101-25; newly cited) as applied to claims 31-38, 40-43, and 48-50 above, and further in view of Ahlneck et al. (International Journal of Pharmaceutics (1990) 87-95; previously cited).
Claim 45 is directed to the method of claim 31 but further specifies that the model input defines a residual moisture and the output defines a predicted value of said attribute.
Li et al., Liu et al., Clenet et al., and Waterman et al. teach the method of claims 31-35, 38, 43, and 48-50 as previous described.
Ahlneck et al. teaches in on page 1, column 1, paragraph 1 “It is well recognized that residual water associated with drugs in the solid state can have significant effects on a variety of physical and chemical properties, such as chemical degradation, dissolution rate, flow and compactibility”, which in view of Li et al. and Liu et al., would render the use of residual moisture as model input, and thereby model output, obvious.
It would have been obvious at the time of invention to modify the teachings of Li et al., Liu et al., Clenet et al., and Waterman et al. for the method of claim 31, with the teachings of Ahlneck et al. as the latter mere informs that residual water is associated with variety of chemical properties including chemical degradation. One would have had a reasonable expectation of success given that Ahlneck et al. is merely informing a new source of data and not modifying the prior method. Therefore, it would be obvious to one with ordinary skill in the art to incorporate the teachings of each and to be successful.
Claim 46 is rejected under 35 U.S.C. 103 as being unpatentable over Li et al. (Journal of Chemical Information and Modeling (2019) 1044-1049; previously cited), Liu et al. (Journal of Computationally Aided Molecular Design (2014) 941-950; previously cited), Clenet et al. (Journal of Pharmaceutical Sciences (2014) 3055-3064; newly cited), and Waterman et al. (International Journal of Pharmaceutics (2005) 101-25; newly cited) as applied to claims 31-38, 40-43, and 48-50 above, and further in view of Lee et al. (Food Science Biotechnology (2013) 279-282; previously cited).
Claim 46 is directed to the method of claim 31 but further specifies that the model input defines a radiation state and the output defines a predicted value of said attribute.
Li et al., Liu et al., Clenet et al., and Waterman et al. teach the method of claims 31-35, 38, 43, and 48-50 as previous described.
Lee et al. teaches on page 1, column 1, paragraph 1 “Curcumin, a yellow pigment from turmeric (Curcuma longa), has shown various health beneficial effects. In the present study, changes in the chemical stability and bioactivities of curcumin by UV radiation were investigated. Curcumin degradation in water or phosphate buffered saline was accelerated under UV radiation (254 nm); the residual levels of curcumin were 36.9 and 16.8%, respectively, after 24 h radiation.”, which in view of Li et al. and Liu et al., would render the use of radiation state as model input, and thereby model output, obvious.
It would have been obvious at the time of invention to modify the teachings of Li et al., Liu et al., Clenet et al., and Waterman et al. for the method of claim 31, with the teachings of Lee et al. as the latter mere informs that UV radiation is associated with variety of chemical properties including chemical degradation of curcumin. One would have had a reasonable expectation of success given that Lee et al. is merely informing a new source of data and not modifying the prior method. Therefore, it would be obvious to one with ordinary skill in the art to incorporate the teachings of each and to be successful.
Claim 47 is rejected under 35 U.S.C. 103 as being unpatentable over Li et al. (Journal of Chemical Information and Modeling (2019) 1044-1049; previously cited), Liu et al. (Journal of Computationally Aided Molecular Design (2014) 941-950; previously cited), Clenet et al. (Journal of Pharmaceutical Sciences (2014) 3055-3064; newly cited), and Waterman et al. (International Journal of Pharmaceutics (2005) 101-25; newly cited) as applied to claims 31-38, 40-43, and 48-50 above, and further in view of Anarjan et al. (Journal of the American Oil Chemists' Society (2013) 1223-1227; previously cited).
Claim 47 is directed to the method of claim 31 but further specifies that the model input defines a lighting state and the output defines a predicted value of said attribute.
Li et al., Liu et al., Clenet et al., and Waterman et al. teach the method of claims 31-35, 38, 43, and 48-50 as previous described.
Anarjan et al. teaches in the abstract “Since astaxanthin is sensitive to oxidative damage, its degradations kinetics in the prepared nanodispersion systems were investigated as a function of storage temperature, atmosphere and light. The results showed that astaxanthin degradation followed a first-order kinetic and, in most cases, astaxanthin was more stable in optimum-formulated three-component-stabilized nanodispersions as compared to nanodispersion systems stabilized by individual stabilizers. In addition, high storage temperature and intense illumination significantly (P\0.05) increased the degradation of astaxanthin”, which in view of Li et al. and Liu et al., would render the use of lighting state as model input, and thereby model output, obvious.
It would have been obvious at the time of invention to modify the teachings of Li et al., Liu et al., Clenet et al., and Waterman et al. for the method of claim 31, with the teachings of Anarjan et al. as the latter mere informs that intense illumination is associated with increased chemical degradation of specific chemicals. One would have had a reasonable expectation of success given that Anarjan et al. is merely informing a new source of data and not modifying the prior method. Therefore, it would be obvious to one with ordinary skill in the art to incorporate the teachings of each and to be successful.
Response to Arguments
Applicant’s arguments, see page 5 of the Remarks, filed 5/11/2026, with respect to the rejections of claims 31, 49, and 50 under Liu et al. (Journal of Computationally Aided Molecular Design (2014) 941-950; previously cited) in view of Li et al. (Journal of Chemical Information and Modeling (2019) 1044-1049; previously cited), and Waterman et al. (International Journal of Pharmaceutics (2005) 101-25; previously cited) have been fully considered and are persuasive. Therefore, the rejection has been withdrawn. However, upon further consideration, a new ground of rejection is made in view of Clenet et al. (Journal of Pharmaceutical Sciences (2014) 3055-3064; newly cited).
Conclusion
Applicant's amendment necessitated the new ground(s) of rejection presented in this Office action. Accordingly, THIS ACTION IS MADE FINAL. See MPEP § 706.07(a). Applicant is reminded of the extension of time policy as set forth in 37 CFR 1.136(a).
A shortened statutory period for reply to this final action is set to expire THREE MONTHS from the mailing date of this action. In the event a first reply is filed within TWO MONTHS of the mailing date of this final action and the advisory action is not mailed until after the end of the THREE-MONTH shortened statutory period, then the shortened statutory period will expire on the date the advisory action is mailed, and any nonprovisional extension fee (37 CFR 1.17(a)) pursuant to 37 CFR 1.136(a) will be calculated from the mailing date of the advisory action. In no event, however, will the statutory period for reply expire later than SIX MONTHS from the mailing date of this final action.
Any inquiry concerning this communication or earlier communications from the examiner should be directed to KEENAN NEIL ANDERSON-FEARS whose telephone number is (571)272-0108. The examiner can normally be reached M-Th, alternate F, 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, Karlheinz Skowronek can be reached at 571-272-9047. 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.
/K.N.A./Examiner, Art Unit 1687
/LARRY D RIGGS II/Supervisory Patent Examiner, Art Unit 1686
Prosecution Insights