Machine learning for risk assessment for nitrosamine formation

Abstract

The following relates generally to using artificial intelligence (Al) and / or machine learning (ML) to analyze chemical structures and / or assess risk. In some embodiments, one or more processors: receive information of a first chemical component; receive information of a second chemical component; input: (i) the information of the first chemical component and (ii) the information of the second chemical component into an Al algorithm to predict a plurality of potential molecules that could be created by an interaction between the first chemical component and the second chemical component; and filter the plurality of potential molecules based on a chemical structure type.

Description

MACHINE LEARNING FOR RISK ASSESSMENT FOR NITROSAMINE FORMATIONCROSS REFERENCE TO RELATED APPLICATIONS

[0001] This application claims priority to all of: U.S. Provisional Application No. 63 / 558,290, entitled “Machine Learning For Risk Assessment For Nitrosamine Formation” (filed February 27, 2024), which is incorporated by reference herein in their entirety.FIELD

[0002] The present disclosure generally relates to using artificial intelligence (Al) and / or machine learning (ML) to analyze chemical structures and / or assess risk.BACKGROUND

[0003] Often, when a pharmaceutical manufacturer finds a new active ingredient, the pharmaceutical manufacturer may desire to also find an excipient(s) to mix with the active ingredient to produce a suitable, stable pharmaceutical dosage form. However, mixing active ingredient(s) with excipient(s) may be dangerous. For example, such mixing sometimes produces potential carcinogens, such as nitrosamines.

[0004] The systems and methods disclosed herein provide solutions to these problems and may provide solutions to the ineffectiveness, insecurities, difficulties, inefficiencies, encumbrances, and / or other drawbacks of conventional techniques.SUMMARY

[0005] In one aspect, a computer-implemented method for using artificial intelligence (Al) to analyze chemical structures and / or assess risk may be provided. In one example, the method may include: (1) receiving, via one or more processors, information of a first chemical component; (2) receiving, via the one or more processors, information of a second chemical component; (3) inputting, via the one or more processors: (i) the information of the first chemical component and (ii) the information of the second chemical component into an Al algorithm to predict a plurality of potential molecules that could be created by an interaction between the first chemical component and the second chemical component; and (4) filtering, via the one or more processors, the plurality of potential molecules based on a chemical structure type. The method may include additional, fewer, or alternate actions, including those discussed elsewhere herein.

[0006] In another aspect, another computer-implemented method may be provided. In one example, the method may include: (1) displaying, via one or more processors, a first entry screen configured to allow a user to enter information of a first chemical component; (2) receiving, via one or more processors, information of the first chemical component entered via the first entry screen; (3) displaying, via one or more processors, a second entry screen configured to allow the user to enter information of a second chemical component; (4) receiving, via the one or more processors, information of the second chemical component via the second entry screen; (5) receiving, via the one or more processors, a prediction of a plurality of potential molecules that could be created by an interaction between the first chemical component and the second chemical component, wherein the prediction is based on: (i) the information of the first chemical component, and (ii) the information of the second chemical component; (6) receiving a risk assessment based on the plurality of potential molecules; and (7) displaying, via the one or more processors, an indication of the risk assessment. The method may include additional, fewer, or alternate actions, including those discussed elsewhere herein.BRIEF DESCRIPTION OF THE DRAWINGS

[0007] Advantages will become more apparent to those skilled in the art from the following description of the preferred embodiments which have been shown and described by way of illustration. As will be realized, the present embodiments may be capable of other and different embodiments, and their details are capable of modification in various respects. Accordingly, the drawings and description are to be regarded as illustrative in nature and not as restrictive.

[0008] The figures described below depict various aspects of the applications, methods, and systems disclosed herein. It should be understood that each figure depicts an embodiment of a particular aspect of the disclosed applications, systems and methods, and that each of the figures is intended to accord with a possible embodiment thereof. Furthermore, wherever possible, the following description refers to the reference numerals included in the following figures, in which features depicted in multiple figures are designated with consistent reference numerals.

[0009] Figure 1 depicts an exemplary computer system for using artificial intelligence (Al) to analyze chemical structures and / or assess risk.

[0010] Figure 2 depicts a flow diagram representing an example computer-implemented method or implementation for using Al to analyze chemical structures and / or assess risk.

[0011] Figure 3 depicts an example first entry screen.

[0012] Figure 4A depicts an example second entry screen.

[0013] Figure 4B depicts an example display including both a first entry screen and a second entry screen.

[0014] Figure 5 depicts an example third entry screen.

[0015] Figure 6 depicts an example process leading to an example risk assessment.

[0016] Figure 7 depicts an example cancer risk assessment.

[0017] Figure 8 depicts an example screen showing a nitrosamine molecule that could potentially be formed from an interaction between first and second chemical components.

[0018] Figure 9 depicts an example screen including a risk assessment.

[0019] Figure 10 illustrates a block diagram of an exemplary machine learning modeling method for training and evaluating a machine learning model.

[0020] Figure 11 illustrates an exemplary table of historical information that may be used to train an artificial intelligence or machine learning algorithm.DETAILED DESCRIPTION

[0021] Often, when a pharmaceutical manufacturer finds a new active ingredient, the pharmaceutical manufacturer may desire to also find an excipient(s) to mix with the active ingredient to produce a suitable, stable pharmaceutical dosage form. However, mixing active ingredient(s) with excipient(s) may be dangerous or lead to a reduced stability of the drugproduct. For example, such mixing sometimes produces potential carcinogens, such as nitrosamines.

[0022] To solve this problem and others, some embodiments input, into an artificial intelligence algorithm: (i) information of a first chemical component (e.g., an active ingredient), (ii) information of a second chemical component e.g., an excipient), and / or (iii) information of an impurity (e.g., information of an impurity in the active ingredient and / or excipient) to predict potential molecules that may be produced by an interaction between the first chemical component, and the second chemical component.

[0023] Risk analysis, such as a cancer risk analysis, may then be performed based on the predicted potential molecules.Example System for Using Artificial Intelligence (Al) to Analyze Chemical Structures and / or Assess Risk

[0024] To this end, Figure 1 illustrates an exemplary computer system 100 for using artificial intelligence (Al) to analyze chemical structures and / or assess risk in which the exemplary computer-implemented methods described herein may be implemented. The high-level architecture includes both hardware and software applications, as well as various data communications channels for communicating data between the various hardware and software components.

[0025] The computing device 102 may include one or more processors 120, such as one or more microprocessors, controllers, and / or any other suitable type of processor. The computing device 102 may further include a memory 122 (e.g., volatile memory, non-volatile memory) accessible by the one or more processors 120, (e.g., via a memory controller). The one or more processors 120 may interact with the memory 122 to obtain and execute, for example, computer- readable instructions stored in the memory 122. Additionally or alternatively, computer-readable instructions may be stored on one or more removable media e.g., a compact disc, a digital versatile disc, removable flash memory, etc.) that may be coupled to the computing device 102 to provide access to the computer-readable instructions stored thereon. In particular, the computer- readable instructions stored on the memory 122 may include instructions for executing various applications, such as interaction predictor Al algorithm 124 (e.g., including a interaction predictor Al model, etc.), likelihood prediction Al algorithm 125, artificial intelligence or machine learning (ML) training application 126, risk assessment engine 127, and / or filtering engine 128.

[0026] In operation, the interaction predictor Al algorithm 124 may predict a plurality of potential molecules that could be created by an interaction between a first chemical component and a second chemical component. In some embodiments, this may include creating a list of potential molecules that could be created by the interaction.

[0027] In some implementations, the first chemical component is an active ingredient, and / or the second chemical component is an excipient. Examples of the active ingredients may include:naproxen, nifedipine, sulfadimidine, sulfamerazine, sulfamethoxazole, sulfathiazole, tadalafil, etc.

[0028] Examples of the excipient include polymers, lipids, surfactants (e.g., nonionic surfactants, such as Lutensols®, Exxals™, Agnique®, Glucopon®, Plurafac®, Taximul®, etc.; and ionic surfactants, such as Aspiro™, Dehyton®, Disponil®, Hostapur, Texapon®, sodium lauryl sulfate (SLS), sodium laureth sulfate (SLES), etc.), and plasticizers. Additionally or alternatively, examples of the excipient include a diluent, such as lactose, spray dried lactose, micro crystalline cellulose (Avicel 101 and 102), Pvpk30 (Pearlitol SD200 and 25C), Sorbitol, Dibasic calcium phosphate dehydrate, Calcium sulphate dehydrate etc. Additionally or alternatively, examples of the excipient include a binder, such as gelatin, glucose, lactose, cellulose derivatives-Methyl cellulose, Ethyl cellulose, Hydroxy propylmethyl cellulose, Hydroxy propyl cellulose, starch, Poly vinyl pyrrolidone (Povidone), Sodium alginate, Carboxymethylcellulose, Acacia etc. Additionally or alternatively, examples of the excipient include an insoluble lubricant, such as stearic acid, magnesium stearate, calcium stearate, talc, paraffin, etc. Additionally or alternatively, examples of the excipient include a soluble lubricant, such as sodium lauryl sulphate, sodium benzoate, PEG 400, 600, 8000, etc. Additionally or alternatively, examples of the excipient include a glidant, such as colloidal silicon dioxide (Aerosil), cornstarch, talc, etc. Additionally or alternatively, examples of the excipient include an anti-adherent. Additionally or alternatively, examples of the excipient include a superdisintegrant, such as croscarmellose sodium (Ac-di-sol), crospovidone (polyplasdone), sodium starch glycollate, starch etc.

[0029] Examples of the polymers include homopolymers (e.g, polyvinylpyrrolidone, polyvinylacetate, polyvinylalcohol, polyethylene glycole, etc.), block polymers, random copolymers, triblock polymers, graft polymers, bottlebrush, and star polymers.

[0030] Examples of block polymers include: pol oxamer 188, and poloxamer 407. Examples of random copolymers include: copovidone, and poly(lactic-co-glycolic acid). An example of a graft polymer is soluplus.

[0031] Examples of other excipients include: polyoxyl 40 castor oil, water, sodium lauryl sulfate, colloidal silica, and organic solvents. In some examples, the excipient is a reagent.

[0032] The interaction predictor Al algorithm 124 may also predict plurality of potential molecules that could be created based on an impurity (e.g., an impurity of the excipient and / or an impurity of the active ingredient). Information of the impurities may be retrieved from the impurities database 180 and / or the internal database 118, for example.

[0033] Examples of the impurities include: nitrite, peroxides, solvent residues, and metal residues.

[0034] The predicted plurality of potential molecules that could be created by the interaction between the first and second chemical components may be filtered by the filtering engine 128. For example, the filtering engine 128 may receive an input indicating that the plurality of potential molecules should be filtered based on a chemical structure type. Examples of the chemical structure types include: N-Nitrosamines, glycosylamine (e.g., melanoidins), peroxides, dimers, etc.

[0035] A likelihood prediction Al algorithm 125 may receive the (filtered or unfiltered) plurality of potential molecules as input. The likelihood prediction Al algorithm 125 may output a likelihood that respective molecules of the plurality of potential molecules will be formed (e.g., a percentage or other indicator of likelihood) (e.g., a likelihood that a potential molecule will form as compared to other potential molecules of the plurality of potential molecules).

[0036] A risk assessment engine 127 may then perform a risk assessment based on the plurality of potential molecules, and optionally the likelihood(s) that any of the potential molecules will form. The risk assessment engine 127 may assess for any type of risk. Examples of the risk assessment include: a cancer risk assessment; a Maillard-based risk assessment; an oxidation-based risk assessment; a hydrolysis-based risk assessment; a complexation-based risk assessment; a dimerization-based risk assessment; and a salt formation-based risk assessment. As will be described elsewhere herein the risk assessment engine 127 may also output a recommended maximum amount of the first and / or second chemical component(s) per time period; and / or potency categories.

[0037] In some embodiments, the interaction predictor Al algorithm 124 may use an artificial intelligence or machine learning ML algorithm to (wholly or partially) predict the plurality of potential molecules that could be created by an interaction between a first chemical component and a second chemical component. In this regard, the artificial intelligence or ML trainingapplication 126 may train the artificial intelligence or machine learning ML algorithm. For example, as will be described elsewhere herein, the artificial intelligence or ML training application 126 may route historical information into the ML algorithm to train the ML algorithm. It should be appreciated that the interaction predictor Al algorithm 124 may include a interaction predictor Al model.

[0038] Similarly, in some embodiments, the likelihood prediction Al algorithm 125 may use an artificial intelligence or machine learning ML algorithm to (wholly or partially) predict the likelihood(s) of formation of the potential molecules that could be created by the interaction between the first chemical component and the second chemical component. In this regard, the artificial intelligence or ML training application 126 may train the artificial intelligence or machine learning ML algorithm. For example, as will be described elsewhere herein, the artificial intelligence or ML training application 126 may route historical information into the ML algorithm to train the ML algorithm. It should be appreciated that the likelihood prediction Al algorithm 125 may include a likelihood prediction Al model.

[0039] The computing device 102 may further include a display 140 to display information, as discussed herein.

[0040] In some examples, the computing device 102 receives a request to perform a risk assessment from the manufacturer device 150. For example, the manufacturer device 150 may be associated with a drug company that is considering producing a drug with the first chemical component, such as an active ingredient; in this regard, the drug company may be searching for an excipient (e.g., the first chemical component; it should also be appreciated that the first chemical component may include a plurality of excipients) to mix with the active ingredient (e.g., the second chemical component) (e.g., to produce a drug with the active ingredient in a solid dosage form).

[0041] The manufacturer device 150 may include one or more processors 155 such as one or more microprocessors, controllers, and / or any other suitable type of processor. The manufacturer device 150 may further include a memory 157 (e.g., volatile memory, non-volatile memory) accessible by the one or more processors 155, (e.g., via a memory controller). The one or more processors 155 may interact with the memory 157 to obtain and execute, for example, computer-readable instructions stored in the memory 157. Additionally or alternatively,computer-readable instructions may be stored on one or more removable media (e.g., a compact disc, a digital versatile disc, removable flash memory, etc.) that may be coupled to the manufacturer device 150 to provide access to the computer-readable instructions stored thereon. In particular, the computer-readable instructions stored on the memory 1 7 may include instructions for executing various applications. The manufacturing device 150 may further include a display 159 to display information, as discussed herein.

[0042] Additionally or alternatively to sending the request for the risk assessment, the manufacturer device 150 may also send information of the active ingredient (e.g., possibly including a glass-transition temperature of the active ingredient, a melting point temperature of the active ingredient, enthalpy of fusion of the active ingredient, a density of the active ingredient, molecular weight (MW) of the active ingredient, etc., which, in some embodiments, may be stored in the active ingredient database 158) and / or identifying information of the active ingredient, which the interaction predictor Al algorithm 124 may advantageously use to aide in predicting potential molecules. Additionally or alternatively, the interaction predictor Al algorithm 124 may request and / or receive information (e.g, information of the active ingredient) from the internal database 118. For example, the interaction predictor Al algorithm 124 may query the internal database 118 for information of the active ingredient based on identifying information of the active ingredient received from the manufacturer device 150.

[0043] In addition, further regarding the example system 100, the illustrated exemplary components may be configured to communicate, e.g., via a network 104 (which may be a wired or wireless network, such as the internet), with any other component. Furthermore, although the example system 100 illustrates only one of each of the components, any number of the example components are contemplated (e.g., any number of computing devices, internal databases, external databases, manufacturer devices, active ingredient databases, etc.).Example Method for Esing Artificial Intelligence (Al) to Analyze Chemical Structures and / or Assess Risk

[0044] Figure 2 illustrates a flow diagram representing an example computer-implemented method or implementation 200 for using Al to analyze chemical structures and / or assess risk. The example method or implementation 200 may be implemented by a computing system 100,such as the computing device 102, the manufacturer computing device 150, and / or any suitable device, including those discussed elsewhere herein.

[0045] The example method or implementation 200 may begin at block 202 when a first entry screen is displayed, such as on the displays 140 and / or 159 (e.g., via the or more processors 120 and / or one or more processors 155). The first entry screen may be configured to allow a user to enter information of a first chemical component.

[0046] Figure 3 depicts an example first entry screen 300 (e.g., that may be displayed at display device 140 and / or 159) into which a user may enter the information of the first chemical component (e.g., an active ingredient, etc.). For instance, a user may use the dropdown menu 305 to select the active ingredient. The example screen 300 may also be used to enter any other information of the active ingredient, such as: name, chemical formulation (e.g, in SMILES form) melting point, glass transition temperature, enthalpy of fusion, density, molecular properties, molar weight of the active ingredient, molar volume, and Hansen solubility parameters (e.g., contributions from dispersion forces, contribution from dipolar intermolecular force, and / or contribution for hydrogen bonds).

[0047] At block 204 the one or more processors 120 and / or the one or more processors 155 receive information of the first chemical component. In some examples, the first chemical component is an active ingredient, as discussed above.

[0048] The information of the first chemical component may include any suitable information. Examples of the information of the first chemical component include the examples discussed above that may be entered into screen 300. For instance, examples of the information of the first chemical component include the first chemical component’s: name, melting point, glass transition temperature, enthalpy of fusion, density, molecular properties (e.g., C-H binding enthalpy) or physical properties (e.g., dlO, d50, d90, and true density to determine the surface area, etc.), molar weight, molar volume, and Hansen solubility parameters (e.g., contributions from dispersion forces, contribution from dipolar intermolecular force, and contribution for hydrogen bonds).

[0049] In some embodiments, the information of the first chemical component is received via a user entering it into the first entry screen. However, the information of the first chemical component may, additionally or alternatively, be received be retrieving it from the internaldatabase 118 and / or any other source. For example, the one or more processors 120 and / or the one or more processors 155 may receive the name of an active ingredient (e.g., via a user entering the name via dropdown arrow 305); and, in response, the one or more processors 120 and / or the one or more processors 155 may retrieve additional information of the first chemical component from the internal database 118 and / or any other source (e.g., by querying a database based on the name of the active ingredient).

[0050] At block 206, a second entry screen is displayed, such as on the displays 140 and / or 159 (e.g., via the or more processors 120 and / or one or more processors 155). The second entry screen may be configured to allow a user to enter information of a second chemical component.

[0051] Figure 4A depicts an example second entry screen 400 (e.g., that may be displayed at display device 140 and / or 159) into which a user may enter the information of the second chemical component, such as an excipient, such as a polymer as in the illustrated example. In the illustrated example, the example excipient comprises a polymer; however, it should be appreciated that the excipient may additionally or alternatively include other examples, as discussed elsewhere herein. As illustrated, a user may use the dropdown menu 405 to select the second chemical component. The example second entry screen 400 may also be used to enter any other information of the second chemical component, such as: name, chemical formulation (e.g., in SIMLES form) glass transition temperature, density, molar weight, molar volume, and Hansen solubility parameters (e.g., contributions from dispersion forces, contribution from dipolar intermolecular force, and contribution for hydrogen bonds).

[0052] In addition, it should be appreciated that the first and second entry screens may be displayed simultaneously. For instance, Figure 4B depicts example display 450 including first entry screen 460, and second entry screen 470. In the illustrated example, the first entry screen 460 is configured to allow a user to enter, for example, physical properties (e.g., dlO, d50, d90, and true density to determine the surface area, etc.) of the first chemical component. Further in the illustrated example, in the second entry screen 470, the second chemical component is an excipient (e.g., an inactive ingredient); and the second entry screen 470 is configured to allow a user to enter a weight fraction of the second chemical component.

[0053] At block 208, the one or more processors 120 and / or the one or more processors 155 receive information of the second chemical component. In some examples, the second chemical component is an excipient, as discussed above.

[0054] The information of the second chemical component may include any suitable information. Examples of the information of the second chemical component include the examples discussed above that may be entered into screen 400. For instance, examples of the information of the second chemical component include the second chemical component’s: name, melting glass transition temperature, density, molar weight, molar volume, and Hansen solubility parameters (e.g., contributions from dispersion forces, contribution from dipolar intermolecular force, and contribution for hydrogen bonds).

[0055] In some embodiments, the information of the second chemical component is received via a user entering it into the second entry screen. However, the information of the second chemical component may, additionally or alternatively, be received by retrieving it from the internal database 118 and / or any other source. For example, the one or more processors 120 and / or the one or more processors 155 may receive the name of a second chemical component (e.g, via a user entering the name via dropdown arrow 405); and, in response, the one or more processors 120 and / or the one or more processors 155 may retrieve additional information of the second chemical component from the internal database 118 and / or any other source (e.g., by querying a database based on the name of the second chemical component).

[0056] It should be appreciated that although the example method 200 illustrates two chemical components, any number of chemical components may be used. Thus, in some embodiments, following block 208, the system may receive information of a third chemical component, and so on. In some such embodiments, the system may predict potential molecules formed from the interactions between all the chemical components (e.g., information of three chemical components is received, and the system predicts potential molecules formed from an interaction of the three chemical components).

[0057] At optional block 210, a third entry screen is displayed, such as on the displays 140 and / or 159 (e.g., via the or more processors 120 and / or one or more processors 155). The third entry screen may be configured to allow a user to: (i) select to include information of an impurity (e.g., an impurity in an active ingredient and / or excipient) in the prediction (e.g., the predictionof the plurality of potential molecules that could be created by an interaction between the first chemical component and the second chemical component enter), (ii) select to automatically retrieve the impurity information (e.g, from the impurities database 180 and / or the internal database 118), and / or (iii) enter information of the impurity.

[0058] Figure 5 depicts an example third entry screen 500 (e.g., that may be displayed at display device 140 and / or 159) into which a user may: (i) select to include the information of an impurity (e.g., via box 510), (ii) select to automatically retrieve the impurity information (e.g., via box 520) and / or (iii) enter the information of the impurity (e.g., via box 530). In some examples, a user may select the impurity from a list of impurities (e.g., box 540 may include a dropdown arrow to display a list of impurities).

[0059] At optional block 212, the one or more processors 120 and / or the one or more processors 155 receive information of the second chemical component. For example, the information may be received via a user entering the impurity information into box 540.

[0060] Additionally or alternatively, the impurity information may be retrieved (e.g., from the impurities database 180 and / or the internal database 118). For example, in response to an indication (e.g., via box 520) that impurity information should be automatically retrieved, the one or more processors 120 and / or 155 may retrieve impurity information from a database, such as database 118 and / or 180.

[0061] At optional block 213a, a fourth entry screen is displayed, such as on the displays 140 and / or 159 (e.g., via the or more processors 120 and / or one or more processors 155). The fourth entry screen may be configured to allow a user to: (i) enter mixing information, and / or (ii) retrieve mixing information from a database. Examples of the mixing information include mixing temperature, free energy of first and / or second chemical component, mixing pressure, ultraviolet (UV) light to initiate, mixing catalyst(s) (e.g., in SMILES form), amount of catalyst(s), reaction procedure (e.g., stirring, time, solvent(s), order of additions, refluxing solvents, temperature, etc.), post-reaction processing conditions (e.g., cooled for XYZ hours, kept in vacuum at XYZ degrees C for XYZ hours, purification steps, solvent extraction, solvent precipitation, etc.).

[0062] At optional block 213b, the one or more processors 120 and / or the one or more processors 155 receive the mixing information. For example, the information may be receivedvia a user entering the fourth entry screen. Additionally or alternatively, the impurity information may be retrieved (e.g., from the internal database 118).

[0063] At block 214, the one or more processors 120 and / or the one or more processors 155 may predict a plurality of potential molecules that could be created by an interaction between the first chemical component and the second chemical component based on: (i) the information of the first chemical component, (ii) the information of the second chemical component, and / or (iii) the impurity information.

[0064] In some embodiments, the prediction is made by inputting: (i) the information of the first chemical component, (ii) the information of the second chemical component, (iii) the impurity information into an interaction predictor Al algorithm 124 to predict the plurality of potential molecules that could be created by an interaction between the first chemical component and the second chemical component, and / or (iv) mixing information. In some embodiments, any or all of the: (i) the information of the first chemical component, (ii) the information of the second chemical component, and / or (iii) the impurity information is input into the interaction predictor Al algorithm 124 in Simplified Molecular Input Line Entry System (SMILES) form. The training of the interaction predictor Al algorithm 124 (e.g., via the Al or ML training application 126) will be described elsewhere herein.

[0065] In some embodiments, the information of the first chemical component and / or the information of the second chemical component includes information derived from the SMILES of the first and / or second chemical components. For example, any of the following may be derived from the SMILES of the first and / or second chemical components and used as part of the information of the first and / or second chemical component(s) input into the interaction prediction Al algorithm 124: molecular fingerprints, quantum chemistry descriptors, molecular radial distribution functions, specific molecule site-site association, diffusivities as a function of composition and temperature, phase separation, etc. For instance, the SMILES of the first and / or second chemical components may be used with any of the following methods to produce the above-mentioned examples: a Hartree-Fock method, a density functional theory method, a quantum calculation method, an experimental or group contribution methods (e.g., free energies of reaction, free energies of formation, etc.), a molecular simulation method (e.g., a molecular dynamics or coarse-grained molecular dynamics simulation method).

[0066] Examples of the mixing information include mixing temperature, free energy of first and / or second chemical component, mixing pressure, ultraviolet (UV) light to initiate, mixing catalyst(s) (e.g., in SMILES form), amount of catalyst(s), reaction procedure (e.g., stirring, time, solvent(s), order of additions, refluxing solvents, temperature, etc.), post-reaction processing conditions (e.g., cooled for XYZ hours, kept in vacuum at XYZ degrees C for XYZ hours, purification steps, solvent extraction, solvent precipitation, etc.).

[0067] Furthermore, some embodiments include a recursive aspect (e.g., as illustrated by the dotted arrow recursively feeding from / to box 214). That is, some embodiments feed the predicted plurality of potential molecules back into the interaction predictor Al algorithm 124 to predict an additional plurality of potential molecules. Thus, in some embodiments, the predicted plurality of potential molecules is further based on (v) a previously predicted plurality of potential molecules. In one such example, tertiary amines and a de-alkylating agent (e.g., the first and second chemical components) interact and a predicted potential molecule is a secondary amine (which directly susceptible to nitrosation); the predicted potential model (the secondary amine) is fed back into the interaction predictor Al algorithm 124 for further prediction of the plurality of potential molecules (e.g., which may now likely include an N-nitrosamine produced from the secondary amine).

[0068] Furthermore, the one or more processors 120 and / or the one or more processors 155 may rank the plurality of potential molecules (e.g., in order of likelihood of formation).

[0069] At optional block 216, the one or more processors 120 and / or the one or more processors 155 may detect that the plurality of potential molecules include at least one molecule with a particular chemical structure (e.g., a N-Nitrosamine as chemical structure, a glycosylamine chemical structure from a Maillard-reaction, etc., as discussed elsewhere herein). Advantageously, as will be seen, such a detection may be used to trigger a particular type of risk analysis. Triggering a particular type of risk analysis advantageously saves computational resources (e.g., memory and processing power) and reduces processing time because not all types of risk analysis must be run.

[0070] At block 218, the one or more processors 120 and / or the one or more processors 155 may (e.g., via the filtering engine 128) filter the plurality of potential molecules based on a chemical structure type (e.g., a N-nitrosamine chemical structure, a glycosylamine chemicalstructure, etc., as discussed elsewhere herein). In some embodiments, the chemical structure filtered for is based on a user input. For example, a user may input into computing device 102 and / or manufacturer device 150 to filter for N-nitrosamine chemical structures, glycosylamine chemical structures, etc. In another example, a user may input into computing device 102 and / or manufacturer device 150 to perform a particular type of risk assessment, and the type of chemical structure may be based on the particular type of risk assessment. For instance, a user may input to perform a carcinogenic risk assessment, and, in response, a N-nitrosamine chemical structure is filtered for.

[0071] The risk assessment may additionally or alternatively include a quantitative estimation of an unwanted product. For example, the risk assessment may include a quantitative estimation of a molecule of the plurality of potential molecules with a particular structure type (e.g., a N- nitrosamine structure type, a glycosylamine-structure type, etc.).

[0072] At block 220, the one or more processors 120 and / or the one or more processors 155 may predict likelihoods of the (filtered or unfiltered) potential molecules forming. In some embodiments, the likelihood(s) are predicted by inputting the: (i) (filtered or unfiltered) plurality of potential molecules, (ii) information of the first chemical component, (iii) information of the second chemical component, and / or (iv) impurity information into a likelihood predication Al algorithm 125 to predict the likelihoods of respective formations of the potential molecules of the (filtered or unfiltered) plurality of potential molecules that could be created by the interaction between the first chemical component and the second chemical component (e.g., likelihoods that potential molecules will form as compared to other potential molecules of the plurality of potential molecules). In some embodiments, any or all of the: (i) (filtered or unfiltered) plurality of potential molecules, (ii) information of the first chemical component, (iii) information of the second chemical component, and / or (iv) impurity information is input into the likelihood predication Al algorithm 125 in Simplified Molecular Input Line Entry System (SMILES) form. The training of the likelihood predication Al algorithm 125 (e.g., via the Al or ML training application 126) will be described elsewhere herein.

[0073] In some examples, the likelihood predication Al algorithm 125 receives (i)-(iv) above, and attempts to determine a particular molecule of the plurality of potential molecules which will be the most likely to be created by the interaction between the first and second chemicalcomponents. In doing so, the likelihood predication Al algorithm 125 yields a probability distribution over all the plurality of potential molecules (e.g., the probability distribution is the predicted likelihoods of molecules of the fdtered plurality of molecules forming).

[0074] It should be appreciated that either the fdtered or unfdtered plurality of potential molecules may be input into the likelihood predication Al algorithm 125. However, inputting the fdtered plurality of potential molecules has a technical advantage. Namely, because the fdtering prior to entry into the likelihood predication Al algorithm 125 reduces the amount of data that must be processed, computational resources (e.g., memory and processing power) are saved.

[0075] Furthermore, the one or more processors 120 and / or the one or more processors 155 may rank the (fdtered or unfdtered) plurality of potential molecules.

[0076] At block 222, the one or more processors 120 and / or the one or more processors 155 may perform a risk analysis (e.g., via the risk assessment engine 127, etc.). Any type of risk assessment may be performed. Examples of the risk assessment include: a cancer risk assessment; a Maillard-based risk assessment (e.g., including a determination of a likelihood of browning of the drug-excipient blend); an oxidation-based risk assessment (e.g., including a determination of a likelihood of degradation of an active ingredient); a hydrolysis-based risk assessment; a complexati on-based risk assessment; a dimerization-based risk assessment; and a salt formation-based risk assessment.

[0077] In some embodiments, the type of risk assessment performed is dependent on a type of molecule (and / or a type of chemical structure thereof) that was detected at block 216 or predicted at block 214. For example, if a N-nitrosamine is detected, a cancer risk assessment may be performed. In another example, if a molecule (and / or structure thereof) that would be formed as part of a Maillard reaction is detected, a Maillard-based risk assessment may be performed.

[0078] Figure 6 broadly depicts this concept by way of illustrative example. With reference thereto, at predict! on / detecti on 610 (e.g., block(s) 214 and / or 216 of Figure 2) any of the items 620 may be predicted or detected. In one example of this, a nitrosamine (or a nitrite, which may lead to formation of a nitrosamine) may be predicted or detected. In other examples, a molecule (or structure thereof) of any of the following types of reactions may be predicted or detected:Maillard, oxidation, hydrolysis, complexations, and dimerization. Tn yet another example, a salt molecule may be predicted or detected.

[0079] The type of risk assessment performed at 630 (e.g., block 222 of Figure 2) may then be dependent on the prediction or detection (e.g., a cancer risk assessment is performed if a nitrosamine is detected or predicted). Advantageously, such triggering of a particular type of risk analysis saves computational resources e.g., memory and processing power) and reduces processing time because not all types of risk analysis must be run.

[0080] The risk assessment may generate a recommended maximum amount of the first chemical component per time period (e.g., per hour, per day, per two days, per week, etc.). For instance, if the first chemical component is an active ingredient, the risk assessment may generate a daily recommended maximum amount of the active ingredient.

[0081] Additionally or alternatively, the risk assessment may generate a potency category indicating a level of risk (e.g., a higher category number may indicate a lower risk, etc.).

[0082] Figure 7 illustrates an example cancer risk assessment 700. With reference thereto, the determinations from column 710 lead to the cancer risk assessment 700 generating the potency categories and recommended maximum amount of the first chemical component per time period of column 720.

[0083] At block 224, indications of the risk analysis and / or the potential molecules may be displayed (e.g., at one or both of the displays 140, 159). Figures 8-10 show example screens that may be displayed (e.g., at one or both of the displays 140, 159).

[0084] More particularly, Figure 8 depicts an example screen showing a nitrosamine molecule that could potentially be formed. In the illustrated example, the indications of the potential molecule include an indication of the potential molecule 810, and a percentage likelihood that the potential molecule will form 820. The illustrated example screen further includes ranked list of molecules 830 (e.g., a ranked list of the potential molecules including likelihood percentages that respective molecules of the plurality of potential molecules will form). In some embodiments, the ranked list further includes the names of the ranked potential molecules.

[0085] Figure 9 depicts an example screen including a risk assessment. The illustrated example includes: an indication of potential molecule 910; potency category 920 (e.g., anindication of the risk assessment); and recommended maximum amount of the active ingredient 930 (e.g., another indication of the risk assessment).

[0086] It should be appreciated that any of the displaying described with respect to block 224 may additionally or alternatively be performed at any other point in the example method 200. For example, the predicted plurality of potential molecules may be displayed after block 214, etc.

[0087] It should be further appreciated that not all blocks and / or events of the exemplary signal diagrams and / or flowcharts are required to be performed. Moreover, the exemplary signal diagrams and / or flowcharts are not mutually exclusive (e.g., block(s) / events from each example signal diagram and / or flowchart may be performed in any other signal diagram and / or flowchart). The exemplary signal diagrams and / or flowcharts may include additional, less, or alternate functionality, including that discussed elsewhere herein.Example Artificial Intelligence (Al) and / or Machine Learning (ML) Techniques

[0088] Figure 10 is a block diagram of an example machine learning modeling method 1000 for training and evaluating a machine learning model (e.g., a machine learning algorithm) (e.g., the first, second, third, fourth, or fifth machine learning models mentioned above), in accordance with various embodiments. In some embodiments, the model “learns” an algorithm capable of performing the desired function, such as determining one or more output parameters. It should be understood that the principles of Figure 10 may apply to any artificial intelligence and / or machine learning algorithm discussed herein. For example, although the following discussion refers to the example machine learning modeling method 1000, it should be understood that it applies equally to be an example artificial intelligence modeling method 1000.

[0089] At a high level, the machine learning modeling method 1000 includes a block 1010 to prepare the data, a block 1020 to build and train the model, and a block 1030 to run the model.

[0090] Block 1010 may include sub-blocks 1012 and 1016. At block 1012, the artificial intelligence or ML training application 126 may receive the historical information to train the machine learning algorithm. In some embodiments, the historical information includes historical information of (i) historical first chemical compounds (e.g., historical active ingredients), (ii) historical second chemical compounds (e.g., historical excipients), and / or (iii) historical impurities, (iv) historical mixing information, and (v) historical molecules (e.g., historicalmolecules formed by interactions between the historical first chemical components, and the historical second chemical components).

[0091] In some embodiments, the ML algorithm may be trained using the above (i)-(i v) as inputs to the machine learning model (e.g., also referred to as independent variables, or explanatory variables), and the above (v) is used as the output of the machine learning model (e.g., also referred to as a dependent variable, or response variable). Put another way, the above (i)-(iv) (e.g., the historical information of historical first chemical compounds, historical second chemical compounds, historical impurities, and / or historical mixing information) may have an impact on (v) (e.g., the historical molecules); and the ML algorithm may be trained to find this impact.

[0092] Broadly speaking, regarding (i), examples of the historical information of the historical first chemical components include: name of the historical first chemical component, formulation of the historical first chemical component (e.g., in SMILES form, etc.), melting point of the historical first chemical component, glass transition temperature of the historical first chemical component, enthalpy of fusion of the historical first chemical component, density of the historical first chemical component, molecular properties of the historical first chemical component, molar weight of the historical first chemical component, molar volume of the historical first chemical component, and Hansen solubility parameters of the historical first chemical component.

[0093] Broadly speaking, regarding (ii), examples of the historical information of the historical second chemical component include historical second chemical component’s: name, formulation (e.g., in SMILES form, etc.), melting point, glass transition temperature, enthalpy of fusion, density, molecular properties, molar weight, molar volume, and Hansen solubility parameters.

[0094] Broadly speaking, regarding (iii), examples of the historical information of the historical impurities include historical information of historical impurities in historical active ingredients, and historical information of historical impurities of historical excipients. However, it should be appreciated that in some embodiments, the historical impurity information is not included. For instance, in some embodiments, the independent variables may include only thehistorical information of the historical first chemical components and the historical information of the historical second chemical components.

[0095] Broadly speaking, regarding (iv), examples of the historical mixing information include historical mixing temperatures, historical free energy of first and / or second historical chemical components, historical mixing pressures, historical ultraviolet (UV) lights to initiate, historical mixing catalyst(s) (e.g., in SMILES form), historical amounts of catalyst(s), historical reaction procedures (e.g., stirring, time, solvent(s), order of additions, refluxing solvents, temperature, etc.), historical post-reaction processing conditions (e.g., cooled for XYZ hours, kept in vacuum at XYZ degrees C for XYZ hours, purification steps, solvent extraction, solvent precipitation, etc.).

[0096] Broadly speaking, regarding (v), examples of the historical information of the historical molecules include the historical molecule’s: name, formulation (e.g, in SMILES form, etc.), melting point, glass transition temperature, enthalpy of fusion, density, molecular properties, molar weight, molar volume, and Hansen solubility parameters.

[0097] It should be appreciated that based on the above (i)-(iv) the ML algorithm may be trained to determine the percentage likelihood of molecules forming from the interactions between the historical first chemical components and the second chemical components.

[0098] In some embodiments, the historical information may be held in the form of a table, such as the exemplary table 1100 illustrated in the example of Figure 11. However, it should be appreciated that, in other examples, the computing device 102 may implement one or more alternate data structures that represent the historical information.

[0099] To this end, Figure 11 depicts an example table 1100 of historical data including parameters used for ML models that determine potential molecules forming from an interaction between a first chemical component and a second chemical component. The example table 1100 includes independent variables 1105, and further includes dependent variable 1110. When inputting the example table 1100 into a ML model for training, it may be specified if each of the variables is an independent or dependent variable. Furthermore, it should be appreciated that in some embodiments, the historical impurity information is not included. For instance, in some embodiments, the independent variables 1105 may include only the historical information of thehistorical first chemical components and the historical information of the historical second chemical components.

[0100] Any of the historical information discussed above (e.g., in the table 1100, etc.) may come from any suitable source. For example, the historical information may be received (e.g., by the one or more processors 120 and / or 155) from the internal database 118, and / or the impurities database 180.

[0101] In addition, it may be noted that the ML algorithm may be any suitable ML algorithm, such as a deep learning algorithm, a neural network, a convolutional neural network, Gaussian processes regression, random forest, etc.

[0102] Returning now to Figure 10, at block 1014, the data is split into training and validation datasets. Advantageously, this allows for the model to first be trained on the training data, and later evaluated based on reactions it was not trained on (e.g., the validation data) to avoid overfitting.

[0103] At block 1016 the artificial intelligence or ML training application 126 may extract features from the received data, and put them into vector form. For example, the features may correspond to the values associated with the historical information used as input factors.Furthermore, at block 1016, the received data may be assessed and cleaned, including handling missing data and handling outliers. For instance, missing records, zero values (e.g., values that were not recorded), incomplete data sets (e.g., for scenarios when data collection was not completed), outliers, and inconclusive data may be removed.

[0104] Block 1020 may include sub-blocks 1022 and 1026. At block 1022, the machine learning (ML) model is trained (e.g. based upon the data received from block 1010).

[0105] At block 1026, the artificial intelligence or ML training application 126 may evaluate the machine learning model, and determine whether or not the machine learning model is ready for deployment.

[0106] Further regarding block 1026, evaluating the model sometimes involves testing the model using testing data or validating the model using validation data. Testing / validation data typically includes both predictor feature values and target feature values (e.g., including known inputs and outputs), enabling comparison of target feature values predicted by the model to theactual target feature values, enabling one to evaluate the performance of the model. This testing / validation process is valuable because the model, when implemented, will generate target feature values for future input data that may not be easily checked or validated.

[0107] Thus, it is advantageous to check one or more accuracy metrics of the model on data for which the target answer is already known (e.g., testing data or validation data, such as data including historical information, such as the historical information discussed above), and use this assessment as a proxy for predictive accuracy on future data. Exemplary accuracy metrics include key performance indicators, comparisons between historical trends and predictions of results, cross-validation with subject matter experts, comparisons between predicted results and actual results, etc.

[0108] At block 1030, the artificial intelligence or ML training application 126 runs the ML model. For example, information received at blocks 204, 208, and / or 212 of Figure 2 may be input into the ML model to determine the predicted plurality of potential molecules.

[0109] In addition, advantageously, to even further improve the accuracy of the ML algorithm, the ML algorithm may be “updated” by continuing to train the ML algorithm as additional data is received. Thus, as illustrated in the example of Figure 10, following block 1030, the example method 1000 may return to block 1010. For example, following an initial training phase and running of the ML model, the one or more processors 120 and / or 155 may receive additional historical information to further train the ML algorithm.

[0110] It should be understood that not all blocks and / or events of the exemplary signal diagrams and / or flowcharts 200, 600, 1000 are required to be performed. Moreover, the exemplary signal diagrams and / or flowcharts are not mutually exclusive (e.g., block(s) / events from each example signal diagram and / or flowchart may be performed in any other signal diagram and / or flowchart). The exemplary signal diagrams and / or flowcharts may include additional, less, or alternate functionality, including that discussed elsewhere herein.Additional Example Embodiments - Prediction and Risk Analysis

[0111] Aspect 1. A computer-implemented method for using artificial intelligence (Al) to analyze chemical structures and / or assess risk, the computer-implemented method comprising: receiving, via one or more processors, information of a first chemical component;receiving, via the one or more processors, information of a second chemical component; inputting, via the one or more processors: (i) the information of the first chemical component and (ii) the information of the second chemical component into an Al algorithm to predict a plurality of potential molecules that could be created by an interaction between the first chemical component and the second chemical component; and filtering, via the one or more processors, the plurality of potential molecules based on a chemical structure type.

[0112] Aspect 2. The computer-implemented method of aspect 1, wherein: the first chemical component includes an active ingredient; and the second chemical component includes an excipient.

[0113] Aspect 2A. The computer-implemented method of aspect 1, wherein: the first chemical component includes an active ingredient, and the information of the first chemical component is information of the active ingredient; and the information of the active ingredient includes the active ingredient: name, chemical formulation, melting point, glass transition temperature, enthalpy of fusion, density, molecular properties, molar weight, molar volume, and / or Hansen solubility parameters.

[0114] Aspect 2B. The computer-implemented method of aspect 1, wherein: the second chemical component includes an excipient, and the information of the second chemical component is information of the excipient; and the information of the excipient includes the excipient: name, chemical formulation, melting point, glass transition temperature, enthalpy of fusion, density, molecular properties, molar weight, molar volume, and / or Hansen solubility parameters.

[0115] Aspect 2C. The computer-implemented method of aspect 1, wherein the information of the first chemical component and / or the information of the second chemical component includes information in Simplified Molecular Input Line Entry System (SMILES) form.

[0116] Aspect 2D. The computer-implemented method of aspect 1, further comprising receiving, via one or more processors, information of a third chemical component, and wherein the inputting further includes inputting, via the one or more processors, the information of the third chemical component into the Al algorithm to predict the plurality of potential moleculesthat could be created by the interaction between the first chemical component and the second chemical component.

[0117] Aspect 3. The computer-implemented method of aspect 1, wherein the chemical structure type includes: a N-nitrosamine structure type; or a glycosylamine structure type.

[0118] Aspect 4. The computer-implemented method of aspect 1, further including: detecting, via the one or more processors, that the plurality of potential molecules include at least one N-nitrosamine; and in response to the detection that the plurality of potential molecules include at least one N-nitrosamine, performing, via the one or more processors, a cancer risk assessment.

[0119] Aspect 5. The computer-implemented method of aspect 4, wherein: the preforming the cancer risk assessment includes determining, via the one or more processors, a recommended maximum amount of the first chemical component per time period.

[0120] Aspect 6. The computer-implemented method of aspect 1, further including performing a risk assessment based on the filtered plurality of potential molecules, wherein the risk assessment includes at least one of: a Maillard-based risk assessment; an oxidation-based risk assessment; a hydrolysis-based risk assessment; a complexati on-based risk assessment; a dimerization-based risk assessment; or a salt formation-based risk assessment.

[0121] Aspect 6A. The computer-implemented method of aspect 1, further including performing a risk assessment based on the filtered plurality of potential molecules, wherein the risk assessment is a Maillard-based risk assessment, and includes determining a likelihood of browning.

[0122] Aspect 6B. The computer-implemented method of aspect 1, wherein the risk assessment includes a quantitative estimation of a molecule of the plurality of potentialmolecules with a particular structure type, and wherein the particular structure type is a N- nitrosamine structure type or a glycosylamine structure type.

[0123] Aspect 7. The computer-implemented method of aspect 1, wherein the Al algorithm is an interaction predictor Al algorithm, and the method further includes: inputting, via the one or more processors, the filtered plurality of potential molecules into a likelihood prediction Al algorithm to predict likelihoods of molecules of the filtered plurality of molecules forming.

[0124] Aspect 7A. The computer-implemented method of aspect 7, wherein the filtered plurality of potential molecules is input in Simplified Molecular Input Line Entry System (SMILES) form into the likelihood prediction Al algorithm.

[0125] Aspect 7B. The computer-implemented method of aspect 7, wherein inputting, via the one or more processors, the filtered plurality of potential molecules into the likelihood prediction Al algorithm to predict the likelihoods of molecules of the filtered plurality of molecules forming further includes inputting a N-nitrosamine of the filtered plurality of potential molecules in Simplified Molecular Input Line Entry System (SMILES) form into the likelihood prediction Al algorithm to predict the likelihoods of molecules forming.

[0126] Aspect 8. The computer-implemented method of aspect 1, wherein: the first chemical component includes an active ingredient; the method further includes retrieving, via the one or more processors, an impurity information of an impurity associated with the active ingredient from an impurities database; and the inputting further includes inputting, via the one or more processors, (iii) the impurity information into the Al algorithm to predict the plurality of potential molecules that could be created by the interaction between the first chemical component and the second chemical component.

[0127] Aspect 9. The computer-implemented method of aspect 1, further including training, via the one or more processors, the Al algorithm by inputting, into the Al algorithm, historical information of: (i) historical first chemical compounds, (ii) historical second chemical compounds, (iii) historical impurities, (iv) historical mixing information, and (v) historical molecules.

[0128] Aspect 10. A computer device for using artificial intelligence (AT) to analyze chemical structures and / or assess risk, the computer device comprising one or more processors configured to: receive information of a first chemical component; receive information of a second chemical component; input: (i) the information of the first chemical component and (ii) the information of the second chemical component into an Al algorithm to predict a plurality of potential molecules that could be created by an interaction between the first chemical component and the second chemical component; and filter the plurality of potential molecules based on a chemical structure type.

[0129] Aspect 11. The computer device of aspect 10, wherein: the first chemical component includes an active ingredient; and the second chemical component includes an excipient.

[0130] Aspect 12. The computer device of aspect 10, wherein the chemical structure type includes: a N-nitrosamine structure type; or a glycosylamine structure type.

[0131] Aspect 13. The computer device of aspect 10, wherein the one or more processors are further configured to: detect that the plurality of potential molecules include at least one N-nitrosamine; and in response to the detection that the plurality of potential molecules include at least one N-nitrosamine, perform a cancer risk assessment.

[0132] Aspect 14. The computer device of aspect 13, wherein the one or more processors are further configured to perform the cancer risk assessment by determining a recommended maximum amount of the first chemical component per time period.

[0133] Aspect 15. The computer device of aspect 10, wherein the Al algorithm is an interaction predictor Al algorithm, and the one or more processors are further configured to: input the filtered plurality of potential molecules into a likelihood prediction Al algorithm to predict likelihoods of molecules of the filtered plurality of molecules forming.

[0134] Aspect 16. A computer system for using artificial intelligence (Al) to analyze chemical structures and / or assess risk, the computer system comprising: one or more processors; and one or more non-transitory memories, the one or more non-transitory memories having stored thereon computer-executable instructions that, when executed by the one or more processors, cause the one or more processors to: receive information of a first chemical component; receive information of a second chemical component; input: (i) the information of the first chemical component and (ii) the information of the second chemical component into an Al algorithm to predict a plurality of potential molecules that could be created by an interaction between the first chemical component and the second chemical component; and filter the plurality of potential molecules based on a chemical structure type.

[0135] Aspect 17. The computer system of aspect 16, wherein: the first chemical component includes an active ingredient; and the second chemical component includes an excipient.

[0136] Aspect 18. The computer system of aspect 16, wherein the chemical structure type includes: a N-nitrosamine structure type; or a glycosylamine structure type.

[0137] Aspect 19. The computer system of aspect 16, wherein the one or more non-transitory memories having stored thereon computer executable instructions that, when executed by the one or more processors, further cause the one or more processors to: detect that the plurality of potential molecules include at least one N-nitrosamine; and in response to the detection that the plurality of potential molecules include at least one N- nitrosamine, perform a cancer risk assessment.

[0138] Aspect 20. The computer system of aspect 16, wherein the Al algorithm is an interaction predictor Al algorithm, and wherein the one or more non-transitory memories having stored thereon computer executable instructions that, when executed by the one or more processors, further cause the one or more processors to:input the filtered plurality of potential molecules into a likelihood prediction Al algorithm to predict likelihoods of molecules of the filtered plurality of molecules forming.Additional Example Embodiments - Graphical User Interface (GUI)

[0139] Aspect 1. A computer-implemented method, comprising: displaying, via one or more processors, a first entry screen configured to allow a user to enter information of a first chemical component; receiving, via one or more processors, information of the first chemical component entered via the first entry screen; displaying, via one or more processors, a second entry screen configured to allow the user to enter information of a second chemical component; receiving, via the one or more processors, information of the second chemical component via the second entry screen; receiving, via the one or more processors, a prediction of a plurality of potential molecules that could be created by an interaction between the first chemical component and the second chemical component, wherein the prediction is based on: (i) the information of the first chemical component, and (ii) the information of the second chemical component; receiving a risk assessment based on the plurality of potential molecules; and displaying, via the one or more processors, an indication of the risk assessment.

[0140] Aspect 2. The computer-implemented method of aspect 1, wherein: the first chemical component includes an active ingredient; and the second chemical component includes an excipient.

[0141] Aspect 2A. The computer-implemented method of aspect 1, wherein: the first chemical component includes an active ingredient, and the information of the first chemical component is information of the active ingredient including the active ingredient: name, chemical formulation, melting point, glass transition temperature, enthalpy of fusion, density, molecular properties, molar weight, molar volume, and / or Hansen solubility parameters; and the first entry screen is further configured to allow the user to enter the: name, chemical formulation, melting point, glass transition temperature, enthalpy of fusion, density, molecular properties, molar weight, molar volume, and / or Hansen solubility parameters.

[0142] Aspect 2B. The computer-implemented method of aspect 1, wherein: the second chemical component includes an excipient, and the information of the second chemical component is information of the excipient including the excipient: name, chemical formulation, melting point, glass transition temperature, enthalpy of fusion, density, molecular properties, molar weight, molar volume, and / or Hansen solubility parameters; and the second entry screen is further configured to allow the user to enter the: name, chemical formulation, melting point, glass transition temperature, enthalpy of fusion, density, molecular properties, molar weight, molar volume, and / or Hansen solubility parameters.

[0143] Aspect 2C. The computer-implemented method of aspect 1, wherein: the first entry screen is further configured to allow the user to enter the information of the first chemical component in Simplified Molecular Input Line Entry System (SMILES) form; and / or the second entry screen is further configured to allow the user to enter the information of the second chemical component in SMILES form.

[0144] Aspect 3. The computer-implemented method of aspect 1, wherein the risk assessment includes at least one of: a Maillard-based risk assessment; an oxidation-based risk assessment; a hydrolysis-based risk assessment; a complexation-based risk assessment; a dimerization-based risk assessment; or a salt formation-based risk assessment.

[0145] Aspect 4. The computer-implemented method of aspect 1, wherein the displayed indication of the risk assessment includes: a recommended maximum amount of the first chemical component per time period; or a potency category.

[0146] Aspect 5. The computer-implemented method of aspect 1, wherein: the risk assessment includes a cancer risk assessment; the displayed indication of the risk assessment includes an indication of a potency category according to a potency category, wherein the potency category is determined based on acarcinogenic risk; and the displayed indication of the risk assessment further includes a maximum recommended amount of the first chemical component per time period.

[0147] Aspect 6. The computer-implemented method of aspect 1, wherein the first chemical component includes an active ingredient, and the method further includes: displaying, via the one or more processors, a prompt asking if impurity information should be included; receiving, via the one or more processors, an indication that the impurity information should be included; and in response to receiving the indication that the impurity information should be included, retrieving, via the one or more processors, impurity information of an impurity associated with the active ingredient from an impurities database; and wherein the prediction of the plurality of potential molecules that could be created by the interaction between the first chemical component and the second chemical component is further based on: (iii) the retrieved impurity information.

[0148] Aspect 7. The computer-implemented method of aspect 1, further including: displaying, via one or more processors, a third entry screen configured to allow the user to enter information of an impurity; and receiving, via the one or more processors, impurity information via the third entry screen; and wherein the prediction of the plurality of potential molecules that could be created by the interaction between the first chemical component and the second chemical component is further based on: (iii) the received impurity information.

[0149] Aspect 8. The computer-implemented method of aspect 1, further including displaying, via the one or more processors, a list of the plurality of potential molecules that could be created by the interaction between the first chemical component and the second chemical component.

[0150] Aspect 9. The computer-implemented method of aspect 1, further including displaying, via the one or more processors a list of the plurality of potential molecules that could be created by the interaction between the first chemical component and the second chemicalcomponent, and wherein: the displayed list includes likelihood percentages that respective molecules included in the list will be formed; and / or the displayed list is ranked according to likelihood percentages the respective molecules included in the list will be formed.

[0151] Aspect 10. A computer device comprising one or more processors configured to: display a first entry screen configured to allow a user to enter information of a first chemical component; receive information of the first chemical component entered via the first entry screen; display a second entry screen configured to allow the user to enter information of a second chemical component; receive information of the second chemical component via the second entry screen; receive a prediction of a plurality of potential molecules that could be created by an interaction between the first chemical component and the second chemical component, wherein the prediction is based on: (i) the information of the first chemical component, and (ii) the information of the second chemical component; receive a risk assessment based on the plurality of potential molecules; and display an indication of the risk assessment.

[0152] Aspect 1 1 . The computer device of aspect 10, wherein: the first chemical component includes an active ingredient; and the second chemical component includes an excipient.

[0153] Aspect 12. The computer device of aspect 10, wherein the displayed indication of the risk assessment includes: a recommended maximum amount of the first chemical component per time period; or a potency category.

[0154] Aspect 13. The computer device of aspect 10, wherein: the risk assessment includes a cancer risk assessment; the displayed indication of the risk assessment includes an indication of a potency category according to a potency category, wherein the potency category is determined based on a carcinogenic risk; andthe displayed indication of the risk assessment further includes a maximum recommended amount of the first chemical component per time period.

[0155] Aspect 14. The computer device of aspect 10, wherein the one or more processors are further configured to: display a third entry screen configured to allow the user to enter information of an impurity; and receive impurity information via the third entry screen; and wherein the prediction of the plurality of potential molecules that could be created by the interaction between the first chemical component and the second chemical component is further based on: (iii) the received impurity information.

[0156] Aspect 15. The computer device of aspect 10, wherein the one or more processors are further configured to display a list of the plurality of potential molecules that could be created by the interaction between the first chemical component and the second chemical component.

[0157] Aspect 16. A computer system comprising: one or more processors; and one or more non-transitory memories, the one or more non-transitory memories having stored thereon computer-executable instructions that, when executed by the one or more processors, cause the one or more processors to: display a first entry screen configured to allow a user to enter information of a first chemical component; receive information of the first chemical component entered via the first entry screen; display a second entry screen configured to allow the user to enter information of a second chemical component; receive information of the second chemical component via the second entry screen; receive a prediction of a plurality of potential molecules that could be created by an interaction between the first chemical component and the second chemical component, wherein the prediction is based on: (i) the information of the first chemical component, and (ii) the information of the second chemical component;receive a risk assessment based on the plurality of potential molecules; and display an indication of the risk assessment.

[0158] Aspect 17. The computer system of aspect 16, wherein: the first chemical component includes an active ingredient; and the second chemical component includes an excipient.

[0159] Aspect 18. The computer system of aspect 16, wherein the displayed indication of the risk assessment includes: a recommended maximum amount of the first chemical component per time period; or a potency category.

[0160] Aspect 19. The computer system of aspect 16, wherein: the risk assessment includes a cancer risk assessment; the displayed indication of the risk assessment includes an indication of a potency category according to a potency category, wherein the potency category is determined based on a carcinogenic risk; and the displayed indication of the risk assessment further includes a maximum recommended amount of the first chemical component per time period.

[0161] Aspect 20. The computer system of aspect 16, wherein the one or more non-transitory memories having stored thereon computer executable instructions that, when executed by the one or more processors, further cause the one or more processors to display a list of the plurality of potential molecules that could be created by the interaction between the first chemical component and the second chemical component.Other matters

[0162] Although the text herein sets forth a detailed description of numerous different embodiments, it should be understood that the legal scope of the invention is defined by the words of the claims set forth at the end of this patent. The detailed description is to be construed as exemplary only and does not describe every possible embodiment, as describing every possible embodiment would be impractical, if not impossible. One could implement numerous alternate embodiments, using either current technology or technology developed after the filing date of this patent, which would still fall within the scope of the claims.

[0163] It should also be understood that, unless a term is expressly defined in this patent using the sentence “As used herein, the term ‘ ’ is hereby defined to mean. ..” or a similar sentence, there is no intent to limit the meaning of that term, either expressly or by implication, beyond its plain or ordinary meaning, and such term should not be interpreted to be limited in scope based upon any statement made in any section of this patent (other than the language of the claims). To the extent that any term recited in the claims at the end of this disclosure is referred to in this disclosure in a manner consistent with a single meaning, that is done for sake of clarity only so as to not confuse the reader, and it is not intended that such claim term be limited, by implication or otherwise, to that single meaning.

[0164] Throughout this specification, plural instances may implement components, operations, or structures described as a single instance. Although individual operations of one or more methods are illustrated and described as separate operations, one or more of the individual operations may be performed concurrently, and nothing requires that the operations be performed in the order illustrated. Structures and functionality presented as separate components in example configurations may be implemented as a combined structure or component.Similarly, structures and functionality presented as a single component may be implemented as separate components. These and other variations, modifications, additions, and improvements fall within the scope of the subject matter herein.

[0165] Additionally, certain embodiments are described herein as including logic or a number of routines, subroutines, applications, or instructions. These may constitute either software (code embodied on a non-transitory, tangible machine-readable medium) or hardware. In hardware, the routines, etc., are tangible units capable of performing certain operations and may be configured or arranged in a certain manner. In example embodiments, one or more computer systems (e.g., a standalone, client or server computer system) or one or more hardware modules of a computer system e.g., a processor or a group of processors) may be configured by software (e.g., an application or application portion) as a hardware module that operates to perform certain operations as described herein.

[0166] In various embodiments, a hardware module may be implemented mechanically or electronically. For example, a hardware module may comprise dedicated circuitry or logic that is permanently configured (e.g., as a special-purpose processor, such as a field programmable gatearray (FPGA) or an application-specific integrated circuit (ASIC) to perform certain operations). A hardware module may also comprise programmable logic or circuitry (e.g., as encompassed within a general -purpose processor or other programmable processor) that is temporarily configured by software to perform certain operations. It will be appreciated that the decision to implement a hardware module mechanically, in dedicated and permanently configured circuitry, or in temporarily configured circuitry (e.g., configured by software) may be driven by cost and time considerations.

[0167] Accordingly, the term “hardware module” should be understood to encompass a tangible entity, be that an entity that is physically constructed, permanently configured (e.g., hardwired), or temporarily configured (e.g., programmed) to operate in a certain manner or to perform certain operations described herein. Considering embodiments in which hardware modules are temporarily configured (e.g., programmed), each of the hardware modules need not be configured or instantiated at any one instance in time. For example, where the hardware modules comprise a general-purpose processor configured using software, the general-purpose processor may be configured as respective different hardware modules at different times. Software may accordingly configure a processor, for example, to constitute a particular hardware module at one instance of time and to constitute a different hardware module at a different instance of time.

[0168] Hardware modules can provide information to, and receive information from, other hardware modules. Accordingly, the described hardware modules may be regarded as being communicatively coupled. Where multiple of such hardware modules exist contemporaneously, communications may be achieved through signal transmission (e.g., over appropriate circuits and buses) that connect the hardware modules. In embodiments in which multiple hardware modules are configured or instantiated at different times, communications between such hardware modules may be achieved, for example, through the storage and retrieval of information in memory structures to which the multiple hardware modules have access. For example, one hardware module may perform an operation and store the output of that operation in a memory device to which it is communicatively coupled. A further hardware module may then, at a later time, access the memory device to retrieve and process the stored output. Hardware modules may also initiate communications with input or output devices, and can operate on a resource (e.g., a collection of information).

[0169] The various operations of example methods described herein may be performed, at least partially, by one or more processors that are temporarily configured (e.g., by software) or permanently configured to perform the relevant operations. Whether temporarily or permanently configured, such processors may constitute processor-implemented modules that operate to perform one or more operations or functions. The modules referred to herein may, in some example embodiments, comprise processor-implemented modules.

[0170] Similarly, the methods or routines described herein may be at least partially processor- implemented. For example, at least some of the operations of a method may be performed by one or more processors or processor-implemented hardware modules. The performance of certain of the operations may be distributed among the one or more processors, not only residing within a single machine, but deployed across a number of machines. In some example embodiments, the processor or processors may be located in a single location (e.g., within a home environment, an office environment or as a server farm), while in other embodiments the processors may be distributed across a number of geographic locations.

[0171] Unless specifically stated otherwise, discussions herein using words such as “processing,” “computing,” “calculating,” “determining,” “presenting,” “displaying,” or the like may refer to actions or processes of a machine (e.g., a computer) that manipulates or transforms data represented as physical (e.g., electronic, magnetic, or optical) quantities within one or more memories (e.g., volatile memory, non-volatile memory, or a combination thereof), registers, or other machine components that receive, store, transmit, or display information.

[0172] As used herein any reference to “one embodiment” or “an embodiment” means that a particular element, feature, structure, or characteristic described in connection with the embodiment may be included in at least one embodiment. The appearances of the phrase “in one embodiment” in various places in the specification are not necessarily all referring to the same embodiment.

[0173] Some embodiments may be described using the expression “coupled” and “connected” along with their derivatives. For example, some embodiments may be described using the term “coupled” to indicate that two or more elements are in direct physical or electrical contact. The term “coupled,” however, may also mean that two or more elements are not in direct contact witheach other, but yet still co-operate or interact with each other. The embodiments are not limited in this context.

[0174] As used herein, the terms “comprises,” “comprising,” “includes,” “including,” “has,” “having” or any other variation thereof, are intended to cover a non-exclusive inclusion. For example, a process, method, article, or apparatus that comprises a list of elements is not necessarily limited to only those elements but may include other elements not expressly listed or inherent to such process, method, article, or apparatus. Further, unless expressly stated to the contrary, “or” refers to an inclusive or and not to an exclusive or. For example, a condition A or B is satisfied by any one of the following: A is true (or present) and B is false (or not present), A is false (or not present) and B is true (or present), and both A and B are true (or present).

[0175] In addition, use of the “a” or “an” are employed to describe elements and components of the embodiments herein. This is done merely for convenience and to give a general sense of the description. This description, and the claims that follow, should be read to include one or at least one and the singular also includes the plural unless it is obvious that it is meant otherwise.

[0176] Upon reading this disclosure, those of skill in the art will appreciate still additional alternative structural and functional designs for the approaches described herein. Thus, while particular embodiments and applications have been illustrated and described, it is to be understood that the disclosed embodiments are not limited to the precise construction and components disclosed herein. Various modifications, changes and variations, which will be apparent to those skilled in the art, may be made in the arrangement, operation and details of the method and apparatus disclosed herein without departing from the spirit and scope defined in the appended claims.

[0177] The particular features, structures, or characteristics of any specific embodiment may be combined in any suitable manner and in any suitable combination with one or more other embodiments, including the use of selected features without corresponding use of other features. In addition, many modifications may be made to adapt a particular application, situation or material to the essential scope and spirit of the present invention. It is to be understood that other variations and modifications of the embodiments of the present invention described and illustrated herein are possible in light of the teachings herein and are to be considered part of the spirit and scope of the present invention.

[0178] While the preferred embodiments of the invention have been described, it should be understood that the invention is not so limited and modifications may be made without departing from the invention. The scope of the invention is defined by the appended claims, and all devices that come within the meaning of the claims, either literally or by equivalence, are intended to be embraced therein.

[0179] It is therefore intended that the foregoing detailed description be regarded as illustrative rather than limiting, and that it be understood that it is the following claims, including all equivalents, that are intended to define the spirit and scope of this invention.Furthermore, the patent claims at the end of this patent application are not intended to be construed under 35 U.S.C. § 112(f) unless traditional means-plus-function language is expressly recited, such as “means for” or “step for” language being explicitly recited in the claim(s). The systems and methods described herein are directed to an improvement to computer functionality, and improve the functioning of conventional computers.

Claims

Claims1. A computer-implemented method for using artificial intelligence (Al) to analyze chemical structures and / or assess risk, the computer-implemented method comprising: receiving, via one or more processors, information of a first chemical component; receiving, via the one or more processors, information of a second chemical component; inputting, via the one or more processors: (i) the information of the first chemical component and (ii) the information of the second chemical component into an Al algorithm to predict a plurality of potential molecules that could be created by an interaction between the first chemical component and the second chemical component; and filtering, via the one or more processors, the plurality of potential molecules based on a chemical structure type.

2. The computer-implemented method of claim 1, wherein: the first chemical component includes an active ingredient; and the second chemical component includes an excipient.

3. The computer-implemented method of any one of claims 1-2, wherein the chemical structure type includes: a N-nitrosamine structure type; or a glycosylamine structure type.

4. The computer-implemented method of any one of claims 1-3, further including: detecting, via the one or more processors, that the plurality of potential molecules include at least one N-nitrosamine; and in response to the detection that the plurality of potential molecules include at least one N-nitrosamine, performing, via the one or more processors, a cancer risk assessment.

5. The computer-implemented method of claim 4, wherein: the preforming the cancer risk assessment includes determining, via the one or more processors, a recommended maximum amount of the first chemical component per time period.

6. The computer-implemented method of any one of claims 1-5, further including performing a risk assessment based on the fdtered plurality of potential molecules, wherein the risk assessment includes at least one of: a Maillard-based risk assessment; an oxidation-based risk assessment; a hydrolysis-based risk assessment; a complexati on-based risk assessment; a dimerization-based risk assessment; or a salt formation-based risk assessment.

7. The computer-implemented method of any one of claims 1-6, wherein the Al algorithm is an interaction predictor Al algorithm, and the method further includes: inputting, via the one or more processors, the filtered plurality of potential molecules into a likelihood prediction Al algorithm to predict likelihoods of molecules of the filtered plurality of molecules forming.

8. The computer-implemented method of any one of claims 1-7, wherein: the first chemical component includes an active ingredient; the method further includes retrieving, via the one or more processors, an impurity information of an impurity associated with the active ingredient from an impurities database; and the inputting further includes inputting, via the one or more processors, (iii) the impurity information into the Al algorithm to predict the plurality of potential molecules that could be created by the interaction between the first chemical component and the second chemical component.

9. The computer-implemented method of any one of claims 1-8, further including training, via the one or more processors, the Al algorithm by inputting, into the Al algorithm, historical information of: (i) historical first chemical compounds, (ii) historical second chemical compounds, (iii) historical impurities, (iv) historical mixing information, and (v) historical molecules.

10. A computer device for using artificial intelligence (Al) to analyze chemical structures and / or assess risk, the computer device comprising one or more processors configured to: receive information of a first chemical component; receive information of a second chemical component; input: (i) the information of the first chemical component and (ii) the information of the second chemical component into an Al algorithm to predict a plurality of potential molecules that could be created by an interaction between the first chemical component and the second chemical component; and filter the plurality of potential molecules based on a chemical structure type.

11. The computer device of claim 10, wherein: the first chemical component includes an active ingredient; and the second chemical component includes an excipient.

12. The computer device of any one of claims 10-11, wherein the chemical structure type includes: a N-nitrosamine structure type; or a glycosylamine structure type.

13. The computer device of any one of claims 10-12, wherein the one or more processors are further configured to: detect that the plurality of potential molecules include at least one N-nitrosamine; and in response to the detection that the plurality of potential molecules include at least one N-nitrosamine, perform a cancer risk assessment.

14. The computer device of claim 13, wherein the one or more processors are further configured to perform the cancer risk assessment by determining a recommended maximum amount of the first chemical component per time period.

15. The computer device of any one of claims 10-14, wherein the Al algorithm is an interaction predictor Al algorithm, and the one or more processors are further configured to: input the filtered plurality of potential molecules into a likelihood prediction Al algorithm to predict likelihoods of molecules of the filtered plurality of molecules forming.

16. A computer system for using artificial intelligence (Al) to analyze chemical structures and / or assess risk, the computer system comprising: one or more processors; and one or more non-transitory memories, the one or more non-transitory memories having stored thereon computer-executable instructions that, when executed by the one or more processors, cause the one or more processors to: receive information of a first chemical component; receive information of a second chemical component; input: (i) the information of the first chemical component and (ii) the information of the second chemical component into an Al algorithm to predict a plurality of potential molecules that could be created by an interaction between the first chemical component and the second chemical component; and filter the plurality of potential molecules based on a chemical structure type.

17. The computer system of claim 16, wherein: the first chemical component includes an active ingredient; and the second chemical component includes an excipient.

18. The computer system of any one of claims 16-17, wherein the chemical structure type includes: a N-nitrosamine structure type; or a glycosylamine structure type.

19. The computer system of claim 16, wherein the one or more non-transitory memories having stored thereon computer executable instructions that, when executed by the one or more processors, further cause the one or more processors to:detect that the plurality of potential molecules include at least one N-nitrosamine; and in response to the detection that the plurality of potential molecules include at least one N-nitrosamine, perform a cancer risk assessment.

20. The computer system of any one of claims 16-19, wherein the Al algorithm is an interaction predictor Al algorithm, and wherein the one or more non-transitory memories having stored thereon computer executable instructions that, when executed by the one or more processors, further cause the one or more processors to: input the filtered plurality of potential molecules into a likelihood prediction Al algorithm to predict likelihoods of molecules of the filtered plurality of molecules forming.

Images