Machine learning (ML) model for active ingredient (AI) and excipient combinations, and for building a phase boundary

Abstract

The following relates generally to determining miscibility of an active ingredients (AIs) with excipients, and more particularly relates to building solid-liquid equilibrium (SLE) and liquid-liquid equilibrium (LLE) phase boundaries for formulations of one or more AIs mixed with one or more excipients. In some examples, the AI may be any number of (macro)molecular species (e.g., a small molecule, such as a solvent, drug, etc.). In some embodiments, one or more processors: receive information of an AI; receive information of an excipient, wherein the excipient comprises a polymer, a lipid, a surfactant, or a plasticizer; produce at least one output parameter by inputting the information of the AI and the information of the excipient into at least one machine learning model; build a SLE or LLE phase boundary based on the at least one output parameter; and rank formulations of AIs and excipients.

Description

MACHINE LEARNING (ML) MODEL FOR ACTIVE INGREDIENT (Al) AND EXCIPIENT COMBINATIONS, AND FOR BUILDING A PHASE BOUNDARYCROSS REFERENCE TO RELATED APPLICATIONS

[0001] This application claims the benefit of U.S. Provisional Application No. 63 / 534,378, entitled “Machine Learning (ML) Model for Active Ingredient (Al) and Excipient Combinations” (filed August 24, 2024), the entirety of which is incorporated by reference herein.FIELD

[0002] The present disclosure generally relates to determining miscibility of an active ingredients (AIs) with excipients, and more particularly relates to computing phase boundaries for solid-liquid equilibrium (SLE) and liquid-liquid equilibrium (LLE) for formulations including one or more AIs mixed with one or more excipients. In some examples, these thermodynamic boundaries may be either binodal or spinodal boundaries for one or more components. The following also relates generally to ranking formulations (e.g., combinations of AIs and excipients).

[0003] In this regard, the present disclosure generally relates to determining the Gibbs free energy change for mixing chemical substances and subsequent construction of SLE and / or LLE equilibrium phase boundaries. More particularly, it relates to mixing higher molecular weight components such as pharmaceutical (polymeric) excipients with active ingredients (Al) and / or solvents. Among other use-cases, an example application is excipient and solvent screening for the formulation development of pharmaceutical (amorphous) solid dispersions.BACKGROUND

[0004] Evaluation of the miscibility of two or more chemical components is a common problem during the development of a multi-component formulation. Homogenous, stable mixtures can be achieved when interactions among individual components are favorable, while unfavorable interactions can cause problems such as phase separation in the form of precipitation and / or crystallization of one or more components.

[0005] Here, manufacturers of poorly soluble active ingredients (AIs) often must find an excipient (or multiple excipients) to mix with an AL For example, when a drug manufacturerfinds a new Al, the drug manufacturer may desire to also find an excipient(s) to mix with the Al to produce a suitable, stable pharmaceutical dosage form. However, finding a suitable excipient to mix with the Al may be challenging, time-consuming, and / or cumbersome.

[0006] 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

[0007] In one aspect, a computer-implemented method for building a phase boundary may be provided. For instance, in one example, the method may include: 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; producing, via the one or more processors, at least one output parameter by inputting the information of the first chemical component and the information of the second chemical component into at least one machine learning model; and building, via the one or more processors, a phase boundary based on the at least one output parameter.

[0008] In another aspect, a computer system for building a phase boundary may be provided. For example, in one instance, the computer system may include one or more processors configured to: receive information of first chemical component; receive information of a second chemical component; produce at least one output parameter by inputting the information of the first chemical component and the information of the second chemical component into at least one machine learning model; and build a phase boundary based on the at least one output parameter.

[0009] In yet another aspect, a computer device for building a phase boundary may be provided. For instance, in one example, the computer device may include: one or more processors; and / or one or more memories coupled to the one or more processors. The one or more memories including computer executable instructions stored therein that, when executed by the one or more processors, may cause the one or more processors to: receive information of a first chemical component; receive information of a second chemical component; produce at least one output parameter by inputting the information of the first chemical component and theinformation of the second chemical component into at least one machine learning model; and build a phase boundary based on the at least one output parameter.

[0010] In yet another aspect, a computer-implemented method for ranking formulations may be provided. For instance, in one example, the method may include: receiving, via one or more processors, information of a first chemical component; receiving, via the one or more processors, information of a plurality of second chemical components; creating, via the one or more processors, respective formulations for the first chemical component paired with respective second chemical components of the second chemical components; producing, via the one or more processors, an output parameter for each respective formulation by inputting the information of the first chemical component and information of the respective second chemical components into at least one machine learning model; and ranking, via the one or more processors, the respective formulations based on the produced output parameters.

[0011] In yet another aspect, a computer system for ranking formulations may be provided. For example, in one instance, the computer system may include one or more processors configured to: receive information of a first chemical component; receive information of a plurality of second chemical components; create respective formulations for the first chemical component paired with respective second chemical components of the second chemical components; produce an output parameter for each respective formulation by inputting the information of the first chemical component and information of the respective second chemical components into at least one machine learning model; and rank the respective formulations based on the produced output parameters.

[0012] In yet another aspect, a computer device for ranking formulations may be provided. For instance, in one example, the computer device may include: one or more processors; and / or one or more memories coupled to the one or more processors. The one or more memories including computer executable instructions stored therein that, when executed by the one or more processors, may cause the one or more processors to: receive information of a first chemical component; receive information of a plurality of second chemical components; create respective formulations for the first chemical component paired with respective second chemical components of the second chemical components; produce an output parameter for each respective formulation by inputting the information of the first chemical component andinformation of the respective second chemical components into at least one machine learning model; and rank the respective formulations based on the produced output parameters.BRIEF DESCRIPTION OF THE DRAWINGS

[0013] 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.

[0014] 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.

[0015] Figure 1 depicts an exemplary computer system for building a phase boundary.

[0016] Figure 2A depicts an example phase diagram.

[0017] Figure 2B depicts example phase diagram including a liquid-liquid equilibrium (LLE) boundary.

[0018] Figure 3 depicts an example flow diagram of building one or more phase boundaries.

[0019] Figure 4 depicts an example screen into which a user may enter the information of theAl.

[0020] Figure 5A depicts an example screen into which a user may enter the information of the excipient.

[0021] Figure 5B depicts an example screen into which a user may enter experimental melting point depression data.

[0022] Figure 6A depicts an example graph of an example free energy of mixing value.

[0023] Figure 6B depicts free energy of mixing values for multiple monomers.

[0024] Figure 6C depicts an example graph illustrating dependence of the parameter value on mixture composition (active ingredient weight fraction) at different temperatures.

[0025] Figure 6D depicts an example graph illustrating dependence of the parameter value on temperature for different mixture compositions (active ingredient weight fraction).

[0026] Figure 6E depicts an example graph illustrating a 3-dimentional (3D) plot of the dependence of the parameter value on temperature and composition of the mixture (active ingredient weight fraction).

[0027] Figure 7 illustrates a block diagram of an exemplary machine learning modeling method for training and evaluating a machine learning model.

[0028] Figure 8 illustrates an exemplary table of historical information that may be used to train an artificial intelligence or ML algorithm.

[0029] Figure 9 depicts an example screen relating to ranking information.

[0030] Figures 10 A- 10C depict graphical representations of the output parameters shown for ranking in Figure 9. More specifically, Figure 10A depicts an example graph illustrating glass transition Tgboundary, and SLE boundary. Figure 10B depicts an example graph illustrating: a first free energy of mixing value Xi boundary; a second free energy of mixing value2boundary; and a combined free energy of mixing value X combined boundary. Figure 10C depicts an example graph illustrating glass transition Tgboundary; and SLE boundary.

[0031] Figure 11 illustrates a comparison between calculation of a free energy of mixing value via the ML model with calculation of the free energy of mixing value via a melting point depression method.

[0032] Figure 12A illustrates a comparison between experimental melting points and predicted melting points.

[0033] Figure 12B illustrates a summary of the accuracy when predicting the rank order of a number of ALpolymer pairs.DETAILED DESCRIPTION

[0034] Drug manufacturers often must find an excipient to mix with an Al. For instance, when a drug manufacturer finds a new Al, the drug manufacturer must also find an excipient(s)to mix with the Al to produce a pill. Yet, finding a suitable excipient to mix with the Al may be challenging, time-consuming, and / or cumbersome.

[0035] To solve this problem and others, some techniques described herein advantageously help in the selection of chemical component(s), such as excipients, by computing the Gibbs free energy change for mixing of the components and building a phase diagram of a formulation (e.g., a solid dispersion, etc.) including a first chemical component (e.g., one or more AIs) and a second chemical component (e.g., one or more excipients). The phase diagram may include, among other things, phase boundaries (e.g., a solid-liquid equilibrium (SLE) phase boundary, liquid-liquid equilibrium (LLE) phase boundary, a glass transition temperature phase boundary, spinodal boundaries, etc.).

[0036] Moreover, the systems and methods described herein represent an improvement to existing technologies, namely technologies for selecting an excipient(s) to formulate with an Al. In prior systems, laborious and cumbersome experimentation is required to determine the properties of a formulation of an Al / excipient mix. As will be seen, by first using an ML algorithm to produce an output parameter, and subsequently using the output parameter in a thermodynamic equation to determine a phase boundary, such as an SLE phase boundary, the techniques described herein advantageously eliminate the need for the laborious and cumbersome experimentation.Example System for Building Phase Boundaries

[0037] To this end, Figure 1 illustrates an exemplary computer system 100 for building a phase boundary 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.

[0038] 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 phase boundary application 124, ranking application 125, and / or artificial intelligence or machine learning (ML) training application 126.

[0039] In operation, the phase boundary application 124 may determine a phase boundary and / or build phase diagrams of formulations. The formulations may include, for example, one or more AIs and one or more excipients. Examples of the AIs may include: naproxen, nifedipine, sulfadimidine, sulfamerazine, sulfamethoxazole, sulfathiazole, tadalafil, etc.

[0040] 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.

[0041] 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.

[0042] Examples of block polymers include: poloxamer 188, and poloxamer 407.

[0043] Examples of random copolymers include: copovidone, and poly(lactic-co-glycolic acid).

[0044] Examples of graft polymers include: soluplus.

[0045] Examples of other excipients include: polyoxyl 40 castor oil, water, sodium lauryl sulfate, colloidal silica, and organic solvents.

[0046] Figure 2A depicts an example phase diagram 200 of an example formulation, which may be created by the phase boundary application 124. The example phase diagram 200 may include a solid-liquid equilibrium (SLE) phase boundary 202, which may include, for example, an x-axis of Al weight fraction (of the formulation, e.g., with the remaining weight fraction of the formulation being of the excipient) and / or a y-axis of temperature. Broadly speaking, the SLE curve 202 may show that to the right of / below the curve, the formulation (e.g., soliddispersion) may be a solid in equilibrium with a mixed liquid or amorphous glass; whereas, to the left of / above the curve, the formulation (e.g., solid dispersion) may be a liquid if above the Tgboundary or an amorphous glass below the Tgboundary. Generally, when a manufacturing company is searching for an excipient to pair with an Al, it is advantageous to find an excipient that combines with the Al to form an amorphous formulation with a SLE curve that is closer to the lower right of the phase diagram. This allows for a higher, thermodynamically stable weight fraction of the Al being used in combination with the excipient.

[0047] In addition, the example phase diagram 200 also illustrates experimental data (e.g., values measured by experiments on actual chemicals rather than computer-simulated values) including experimental data indicating the physical state of the mixture (amorphous or showing signs of crystallization).

[0048] The phase boundary application 124 may also build a glass transition temperature Tgcurve 204, as in the example phase diagram 204. The glass transition temperature curve 204 may show a transition boundary for the formulation between a glassy state and a viscous or rubbery state.

[0049] Figure 2B depicts example phase diagram 250, which also may be created by the phase boundary application 124, and which includes a liquid-liquid equilibrium boundary 256. It may be noted that in some examples, the spinodal boundary (e.g., the LLE boundary) is a specific thermodynamic boundary that denotes the limit of metastability. For instance, in some cases, this is the boundary at which there may no longer be a mixture of Al and excipient being metastable to demixing. That is, inside this boundary the mixture should spontaneously demix (from a thermodynamic perspective). Kinetically speaking, this demixing might progress slowly. This is contrary to being outside the boundary, where there would be either a stable mixture or a metastable mixture. The closer to the boundary, the more likely there is to be a metastable region. Rigorously speaking, the metastable region would be bounded above (if in the temperature-composition plane) by the binodal boundary. The binodal boundary may be referred to herein as LLE phase boundary. Advantageously, the spinodal boundary is computationally less intensive, and thus quicker to compute than the binodal boundary (LLE boundary).

[0050] In addition, the example phase diagram 250 also illustrates SLE phase boundary 252, and glass transition temperature curve 254, which may show a transition boundary for the formulation between a glassy state and a viscous or rubbery state.

[0051] In some embodiments, the phase boundary application 124 may use an artificial intelligence or machine learning ML algorithm to (wholly or partially) determine a phase boundary and / or build a phase diagram. 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.

[0052] In some examples, the phase boundary application 124 receives a request to make a phase diagram from the Al manufacturer device 150. For example, the Al manufacturer device 150 may be associated with a drug company that is considering producing a drug with the Al; in this regard, the drug company may be searching for an excipient to mix with the Al (e.g., to produce a drug with the Al in a solid dosage form).

[0053] The Al 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 Al 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 Al manufacturer device 150 to provide access to the computer-readable instructions stored thereon. In particular, the computer-readable instructions stored on the memory 157 may include instructions for executing various applications.

[0054] Additionally or alternatively to sending the request to make the phase diagram, the Al manufacturer device 150 may also send information of the Al (e.g., possibly including a glasstransition temperature of the Al, a melting point temperature of the Al, enthalpy of fusion of the Al, a density of the Al, molecular weight (MW) of the Al, etc., which, in some embodiments,may be stored in the Al database 158) and / or identifying information of the Al, which the phase boundary application 124 may advantageously use to aide in determining phase boundaries. Additionally or alternatively, the phase boundary application 124 may request and / or receive information (e.g., information of the Al) from the internal database 118 and / or the external database 180. For example, the phase boundary application 124 may query the internal database 118 and / or the external database 180 for information of the Al based on identifying information of the Al received from the Al manufacturer device 150.

[0055] 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, Al manufacturers, Al databases, etc.).Example Methods for Building Phase Boundaries

[0056] Figure 3 depicts an example flow diagram 300 of building one or more phase boundaries.

[0057] With reference thereto, at block 305, the one or more processors 120 may receive information of an Al (e.g., a first chemical component). The information of the Al (e.g., information of the first chemical component) may comprise any suitable information.Examples of the information of the Al include a glass-transition temperature of the Ala melting point temperature of the Al (Tm o), enthalpy of fusion of the Ala density of the Al, molecular weight (MW) of the Al, a molar volume of the Al (vm) , etc.

[0058] The one or more processors 120 may receive information of the Al in a simplified molecular input line entry system (SMILES) form. Additionally or alternatively, the information may be received as a 2D or 3D structure, which is in turn used to derive the SMILES form. Additionally or alternatively, the information may be received in chemical abstracts service (CAS) number form (referenced in Figure 3 as CAS#). For instance, a user may enter the information of the Al in SMILES form into a display of the computing device 102 and / or Al manufacturer device 150. Additionally or alternatively, a user may enteridentifying information of the Al (e.g., a name, chemical abstract service registry number, etc.) into the display of the computing device 102 and / or Al manufacturer device 150, and the one or more processors 120 may use the identifying information of the Al to retrieve the information of the Al e.g., at block 315).

[0059] Figure 4 depicts an example screen 400 (e.g., of the computing device 102 and / or Al manufacturer device 150) into which a user may enter the information of the Al. For instance, a user may use the dropdown menu 405 to select the Al. The example screen 400 may also be used to enter any other information of the Al, such as: melting point, glass transition temperature, enthalpy of fusion, density of Al, molecular properties of the Al, molar weight of the Al, molar volume of the Al, and Hansen solubility parameters (e.g., contributions from dispersion forces, contribution from dipolar intermolecular force, and contribution for hydrogen bonds).

[0060] At block 310, the one or more processors 120 may receive information of an excipient (e.g., a second chemical component). The information of the excipient (e.g., information of the second chemical component) may comprise any suitable information. Examples of the information of the excipient may include a glass-transition temperature of the excipienta molecular weight (MW) of the excipient, a molar volume of the excipient (vm), an architecture of the excipient, monomers comprised in the excipient, etc.

[0061] For example, a user of the computing device 102 and / or Al manufacturer device may select an excipient from a dropdown menu, and the one or more processors 120 may then retrieve the information of the excipient (e.g., from the internal database 118, the external database 180, etc.). In another example, the one or more processors 120 may receive information of the excipient in a SMILES form. For instance, a user may enter the information of the excipient in SMILES form into a display of the computing device 102 and / or Al manufacturer device 150. Additionally or alternatively, a user may enter identifying information of the excipient (e.g., a name, etc.) into the display of the computing device 102 and / or Al manufacturer device 150, and the one or more processors 120 may use the identifying information of the excipient to retrieve the information of the excipient (e.g., at block 315).

[0062] Figure 5A depicts an example screen 500 (e.g., of the computing device 102 and / or Al manufacturer device 150) into which a user may enter the information of the excipient. In theillustrated example, the excipient comprises a polymer; however, it should be appreciated that the excipient may additionally or alternatively comprise a lipid, surfactant, or a plasticizer. As illustrated, a user may use the dropdown menu 505 to select the excipient. The example screen 500 may also be used to enter any other information of the excipient, such as: glass transition temperature, density of the excipient, molar weight of the excipient, molar volume of the excipient, and Hansen solubility parameters (e.g., contributions from dispersion forces, contribution from dipolar intermolecular force, and contribution for hydrogen bonds).

[0063] Figure 5B depicts an example screen 520 (e.g., of the computing device 102 and / or Al manufacturer device 150) into which a user may enter experimental melting point depression data (e.g., to refine the prediction for the excess free energy of mixing prediction, etc.).

[0064] Returning to Figure 3, at block 312, the one or more processors 120 may receive information of historical excess free energy data with other molecules. Examples of the historical excess free energy data with other molecules include melting point depression data and the interaction parameter Xi,j for the Gibbs free energy change for mixing.

[0065] As mentioned above, at box 315, if any information of the Al and / or information of the excipient has not been received by the one or more processors 120, the one or more processors 120 may attempt to look up the missing information (e.g., by querying the internal database 118 and / or the external database 180). Furthermore, the one or more processors 120 may attempt to look up a parameter(s) that quantifies the excess free energies of mixing as functions of temperature, formulation composition ({%}), excipient(s) molecular weight (M), and excipient(s) specific chemical composition, { / }. (It should be appreciated that the brackets { } denote a set of compositions). These excess free energies of mixing may be denoted asamix(.{x}> T> M, { / }) and will often refer to a specific free energy of mixing (chi ) function, which denotes a specific parametric form: a ^ix= xAxBxAB{x], T, M, { / }).

[0066] At box 320, the one or more processors 120 may use an ML algorithm(s) to output an output parameter. For example, if the one or more processors 120 were not able to retrieve any missing information (e.g., information of the Al and / or information of the excipient) at block 315, the one or more processors 120 may use the ML algorithm(s) to “predict” the information (e.g., output an output parameter of the missing information). For instance, if a free energy of mixing x value of the formulation is missing, an ML algorithm may produce an outputparameter of the free energy of mixing value. The training of the ML algorithm will be described elsewhere herein.

[0067] In some examples, the ML algorithm produces more than one output parameter. For example, if the excipient comprises a polymer comprising a plurality of monomers, the ML algorithm may produce a free energy of mixing value (e.g., an output parameter) for each monomer. In some such examples, the free energy of mixing values are combined (using suitable weighting equations for the contribution of individual output parameter values) before input into the thermodynamic equation(s). In another example, the ML algorithm(s) may produce, as output parameters, both the enthalpy of fusion, neat melting point, and molecular volume of the Al, the molecular volume of the monomers in the excipient(s), and the free energy of mixing value(s) between all monomer(s) and Al in the formulation.

[0068] In some examples, different ML algorithms produce different output parameter(s).For example, a first machine learning model 351 may determine an Al melting point; a second machine learning model 352 may determine an Al enthalpy of fusion; a third machine learning model 353 may determine a glass transition temperature of an Al, excipient and / or formulation; a fourth machine learning model 354 may determine free energy of mixing of a formulation; and / or a fifth machine learning model 355 may determine molar volume of an Al and / or excipient.

[0069] At block 325, the one or more processors 120 apply thermodynamic equation(s) (e.g., by using the output parameter(s) along with information of the Al and / or information of the excipient in the thermodynamic equation(s)) to thereby produce outputs, such as phase boundaries (e.g., as illustrated by block 330).

[0070] The thermodynamic equations may include any property derived from the free energy of mixing, amix, which is defined has both an ideal, a^fxl, and excess. a^ix, contributions:amix= amixl+amix([x}’ T> M, { / }). Some examples of properties may include: (1) binodal boundaries involving the first derivatives of amixwith respect to composition, (2) spinodal boundaries involving the second derivatives of amixwith respect to composition, and (3) critical points involving the third derivatives of amixwith respect to composition.

[0071] The glass transition equations may include, for example, (1) Fox-equation, (2) Fox- Flory equation, (3) Ogawa equation, (4) the Fox-Loshaek equation (5) the Kelley-Bueche equation, among others.

[0072] As illustrated in the example of Figure 3, examples of the outputs include phase boundaries, such as SLE phase boundaries, LLE phase boundaries, spinodal boundaries, glasstransition boundaries, etc.

[0073] The phase boundaries may then be used to produce a phase diagram, such as the example phase diagram 200 of Figure 2A. In some examples, plots of Xmonomer-Ai (T, M, {x}) interactions may also be produced, as listed in the example of Figure 3.

[0074] As described above, the one or more processors 120 may also determine free energy of mixing value(s), for example, as part of determining phase boundaries. In this regard, the one or more processors 120 may also build graphs of the free energy of mixing value(s), and Figure 6A depicts an example graph 600 of an example free energy of mixing value. In some examples, monomer- API values will be combined into excipient- API free energy of mixing value Xcombined may be graphed (e.g. , free energy of mixing x values for individual monomers are combined to produce Xcombinec etc.), such as in the examples of Figures 6C-6E.

[0075] Figure 6B depicts free energy of mixing values for multiple monomers. More specifically, Figure 6B depicts example graph 610 for the prediction of the interaction parameter for an Al with SOLUPLUS, which is a tri-block terpolymer comprised of ethylene oxide (EO), vinyl acetate (VAc) and vinylcaprolactam (VC) monomers. It may be noted that total XAB(T> fso> fvAc> fvc> { / ; / })■ The illustrated example uses the monomer-monomer x parameters XEO-VAC, XEO-VC, and XVAC-VC)-

[0076] More particularly, example graph 620 of Figure 6C shows dependence of the x parameter value on mixture composition (active ingredient weight fraction) at different temperatures.

[0077] Example graph 640 of Figure 6D shows dependence of the x parameter value on temperature for different mixture compositions (active ingredient weight fraction).

[0078] Example graph 660 of Figure 6E shows a 3-dimentional (3D) plot of the dependence of the x parameter value on temperature and composition of the mixture (active ingredient weight fraction).Exemplary Al and / or ML Techniques

[0079] Broadly speaking, the artificial intelligence or ML training application 126 may train an artificial intelligence or ML algorithm to, for example, determine one or more output parameters (e.g., to be used to input to the thermodynamic equation(s) at block 325 of Figure 3, etc.). Although the following discussion refers to an ML algorithm, it should be appreciated that it applies equally to ML algorithms and / or artificial intelligence algorithms.

[0080] In some examples, different ML algorithms are trained to determine different output parameters. For example, a first machine learning model may be trained to determine an Al melting point, a second machine learning model may be trained to determine an Al enthalpy of fusion, a third machine learning model may be trained to determine a glass transition temperature of an Al, excipient and / or formulation, a fourth machine learning model may be trained to determine free energy of mixing of a formulation, and / or a fifth machine learning model may be trained to determine molar volume of an Al and / or excipient

[0081] Figure 7 is a block diagram of an exemplary machine learning modeling method 700 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 7 may apply to any machine learning algorithm discussed herein.

[0082] At a high level, the machine learning modeling method 700 includes a block 710 to prepare the data, a block 720 to build and train the model, and a block 730 to run the model.

[0083] Block 710 may include sub-blocks 712 and 716. At block 712, 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 (i)historical information of historical AIs, historical excipients, and / or historical formulations, and (ii) historical output parameters.

[0084] In some embodiments, the ML algorithm may be trained using the above (i) as inputs to the machine learning model (e.g., also referred to as independent variables, or explanatory variables), and the above (ii) 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) (e.g., the historical information of historical AIs, historical excipients, and / or historical formulations) may have an impact on (ii) (e.g., the historical output parameters); and the ML algorithm may be trained to find this impact.

[0085] Broadly speaking, examples of the historical information include: (i) independent variables comprising (a) historical glass transition temperatures of historical AIs, (b) historical melting points of the historical AIs, (c) historical enthalpy of fusions of the historical AIs, (d) historical densities of the historical AIs, (e) historical molar weights of the historical AIs, (f) historical molar volumes of the historical AIs, (g) historical glass transition temperatures of historical excipients, (h) historical molar weights of the historical excipients, (i) historical molar volumes of the historical excipients, and / or (j) historical densities of the historical excipients, (k) historical excess free energies of mixing data (e.g., historical free energy of mixing data (e.g., chi x data), such as historical: monomer-monomer, monomer- Al, or combined excipient- AI free energy of mixing values, etc.), and / or (ii) dependent variables comprising historical output parameters.

[0086] In some embodiments, the historical information may be held in the form of a table, such as the exemplary tables 800, 820, 840 illustrated in the examples of Figure 8A-8C. 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. It may be noted that the workflow for Al parameters and excipient parameters is different, as the excipient(s) can have different sub-structures / architectures that should be accounted for. Thus, advantageously, in some embodiments, the historical information for AIs and excipients may be input via different tables (e.g., a table for Al historical data, such as in the example of Figure 8A; and a table for excipient, such as in the example of Figure 8B).

[0087] To this end, Figure 8A depicts an example table 800 of historical data including parameters used for ML models that predict pure component properties for the AIs. The example table 800 includes independent variables 805, and further includes independent / dependent variables 810 (e.g., variables that can be used as either independent or dependent variables). When inputting the example table 800 into a ML model for training, it would be specified if each of the variables is an independent or dependent variable.

[0088] In some embodiments, as in the example of Figure 8A, the independent variables include chemical descriptors (e.g., generalized chemical descriptors that are computed from the SMILES). Examples of the chemical descriptors include: atom partial charges; molecular weight; heavy atom count; number of valence electrons; molecular fingerprints (Morgan, extended-connectivity, etc.); atom connectivity; (macro)molecular structure; molecule conformers; octanol-water partitioning; molecule solubilities; number of rotatable bonds; radius of gyration; asphericity; number of unique functional groups; highest occupied molecular orbital (HOMO); lowest unoccupied molecular orbital (LUMO); and polarizability.

[0089] Figure 8B depicts an example table 820 of historical data including parameters used for ML models that predict pure component properties for the AIs. The example table 820 includes independent variables 825, and further includes independent / dependent variables 830 (e.g., variables that can be used as either independent or dependent variables). When inputting the example table 820 into a ML model for training, it would be specified if each of the variables is an independent or dependent variable.

[0090] Figure 8C depicts an example table 840 of historical data including parameters used for ML models that predict pure component properties for a formulation (e.g., one or more AIs mixed with one or more excipients, etc.). The example table 840 includes independent variables 845, and further includes independent / dependent variables 850 (e.g., variables that can be used as either independent or dependent variables). When inputting the example table 840 into a ML model for training, it would be specified if each of the variables is an independent or dependent variable.

[0091] Any of the historical information discussed above (e.g., in the table 800, etc.) may come from any suitable source. For example, the historical information may be received (e.g.,by the one or more processors 120) from the internal database 118, the external database 180, and / or the Al manufacturer device 150.

[0092] 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.

[0093] Returning now to Figure 7, At block 716 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 716, 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.

[0094] Block 720 may include sub-blocks 722 and 726. At block 722, the machine learning (ML) model is trained (e.g. based upon the data received from block 710).

[0095] At block 726, 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.

[0096] Further regarding block 726, 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 the actual 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.

[0097] 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 ofresults, cross-validation with subject matter experts, comparisons between predicted results and actual results, etc.

[0098] At block 730, the artificial intelligence or ML training application 126 runs the ML model. For example, information received at blocks 305 and / or 310 of Figure 3 may be input into the ML model to determine an output parameter.

[0099] 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 7, following block 730, the example method 700 may return to block 710. For example, following an initial training phase and running of the ML model, the one or more processors 120 may receive a user-input output parameter corresponding to a formulation of the Al and the excipient (e.g., that the ML model has been run on, etc.); and, during a subsequent training phase, the one or more processors 120 may further train the machine learning model by inputting second historical information into the machine learning model, with the second historical information comprising a dependent variable comprising the user-input output parameter. In some variations, the second historical information may additionally or alternatively include: (a) a glass transition temperature of the Al; (b) a melting point of the Al; (c) an enthalpy of fusion of the Al; (d) a density of the Al, (e) a molar weight of the Al; (f) a molar volume of the Al; (e) a glass transition temperature of the excipient; (h) a molar weight of the excipient; (i) a molar volume of the excipient; and / or (j) excess free energies of mixing data (e.g., historical free energy of mixing data (e.g., chi data), such as historical: monomer-monomer, monomer- Al, or combined excipient-AI free energy of mixing values, etc.). In some alternate variations, rather than use a user-input output parameter corresponding to a formulation of the Al and the excipient, the output parameter generated by the ML algorithm is fed back into block 710 to further update the ML algorithm.

[0100] It should be understood that not all blocks and / or events of the exemplary signal diagrams and / or flowcharts 300, 700 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 / orflowchart). The exemplary signal diagrams and / or flowcharts may include additional, less, or alternate functionality, including that discussed elsewhere herein.Ranking Formulations

[0101] Additionally or alternatively to building a phase boundary, some embodiments may include ranking formulations. For example, formulations may be ranked based on any of (i) glass transition temperature (or difference of glass transition temperature to reference temperature), (ii) combined free energy of mixing value, and / or (iii) stable ALload (e.g., a difference the SLE boundary and the 0.0 AP weight fraction at a reference temperature). In some embodiments, the ranking is performed by the ranking application 125. It should be appreciated that any of the principles discussed elsewhere herein apply to embodiments that rank formulations, and / or chemical components. For example, the machine learning algorithms discussed elsewhere herein may be used to determine output parameter(s) of formulations, and the ranking may be based at least in part on the determined output parameter(s).

[0102] In some embodiments, a user wholly or partially controls the ranking. For example, Figure 9 depicts an example screen 900 (e.g., of the computing device 102 and / or Al manufacturer device 150) which a user may use to control the ranking. More particularly, Figure 9 illustrates an example where a first chemical component comprises an Al, and a plurality of second chemical components comprises a plurality of excipients. In the example of Figure 9, a user may enter an Al weight fraction into box 910, and a reference temperature into box 920. The user may further enter excipients into the plurality of excipients boxes 930. Additionally or alternatively, the excipients may be prepopulated into the boxes 930 (e.g., with recommendations made by the computing device 102, such as based on output parameters determined by machine learning algorithms described elsewhere herein). It should be appreciated that each listed excipient forms a formulation with the Al.

[0103] The user may further select how to rank the formulations / excipients. For example, the user may select: box 941 to rank the formulations / excipients by glass transition temperatures; box 942 to rank the formulations / excipients by difference from glass transition temperature to reference temperature; box 943 to rank the formulations / excipients by combined free energy of mixing value Xcombmed'- and box 944 to rank the formulations / excipients by stable Al-load.Moreover, the buttons 941, 942, 943, 944 may be used to set the ranking as ascending or descending.

[0104] In addition, the example screen 900 displays: numerical values for the formulation’s glass transition temperature at the reference temperature in column 951 ; numerical values for the formulation’s difference from glass transition temperature to the reference temperature in column 952; numerical values for the formulation’s combined free energy of mixing value Xcombined incolumn 953; and numerical values for the formulation’s stable Al-loads in column 954.

[0105] Furthermore, information related to the ranking may be graphically depicted. In one example, Figure 10A depicts an example graph 1000 illustrating glass transition Tgboundary 1002, and SLE boundary 1004. The example graph 1000 further illustrates indication 1006, which indicates a difference between the glass transition Tgboundary 1002 and the reference temperature (e.g. , the difference between the glass transition Tgboundary 1002 and the SLE boundary 1004 at the reference temperature).

[0106] In another example, Figure 10B depicts example graph 1020 illustrating: a first free energy of mixing value Xi boundary 1022; a second free energy of mixing value2boundary 1024; and a combined free energy of mixing value Xcombined boundary 1026. The example graph 1020 further illustrates indication 1030 indicating the combined free energy of mixing value Xcombinedatthe reference temperature.

[0107] In yet another example, Figure 10C depicts an example graph 1040 illustrating glass transition Tgboundary 1042; and SLE boundary 1044. The example graph 1040 further illustrates indication 1046 indicating the formulation’s stable ALload (e.g. , an indication with a double arrow illustrating the difference the SLE boundary 1044 and the 0.0 AP weight fraction at the reference temperature).

[0108] It should be appreciated that graphical depictions, such as the example graphs 1000, 1020, 1040, improve the technical functioning of the system. For example, such graphs facilitate and streamline user selection of which AIs, excipients, and / or formulations to rank and / or select. In one illustrative example, the user views example graph 1000, and based on her opinion of the example graph 1000, she decides to list an excipient of the example graph 1000in excipient boxes 930 of Figure 9 to thereby facilitate ranking of the excipient and / or formulations with the excipient. It should be appreciated that this streamlines and improves the process for determining formulations. Furthermore, once the selected excipient has been ranked against other excipients / formulations (e.g., as illustrated in example screen 900), the user is able to select, from the ranked list, a single excipient / formulation, thereby even further streamlining and improving the process.Additional Data - Free Energy of Mixing Value

[0109] Further improved technical functioning of the system is shown, for example, by the highly accurate determinations of free energy of mixing value. For example, Figure 11 illustrates a comparison between calculation of a free energy of mixing value via the ML model 1110 (e.g., according to the techniques described herein) with calculation of the free energy of mixing value via a melting point depression method 1120. As can be seen, the techniques described herein are highly accurate.Additional Data - Melting Point Value

[0110] Further improved technical functioning of the system is shown, for example, by the highly accurate determinations of melting points values. For example, Figure 12A illustrates a comparison between experimental melting points (e.g., x-axis of graphs 1200, 1210) (e.g., data gathered from melting point depression experiments) and predicted melting points (e.g., y-axis of graphs 1200, 1210) (e.g., predicted according to the techniques described herein). Graph 1200 illustrates a comparison with an ideal line (e.g., a line that accounts for entropic effects, but not enthalpic effects), whereas graph 1210 illustrates a comparison with an SLE line (e.g., a line that also accounts for enthalpic effects, not just entropic effects). Furthermore, with the data illustrated in Figure 12A, the prediction of the SLE line has an average error of + / - 5.8 °C compared to the experimental values. As can be seen, the techniques described herein are highly accurate.

[0111] Figure 12B illustrates a summary of the accuracy when predicting the rank order of a number of Al -polymer pairs. The closer the Spearman coefficient is to (positive) 1.00, the higher the agreement among the predicted rank order versus the actual observed rank order of the ALpolymer pairs.Additional Exemplary Embodiments - Building a Phase Boundary

[0112] Aspect 1. A computer-implemented method for building a phase boundary, the 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; producing, via the one or more processors, at least one output parameter by inputting the information of the first chemical component and the information of the second chemical component into at least one machine learning model; and building, via the one or more processors, a phase boundary based on the at least one output parameter.

[0113] Aspect 1A. The computer-implemented method of aspect 1, wherein the phase boundary is a solid-liquid equilibrium (SLE) phase boundary, and / or a liquid-liquid equilibrium (LLE) phase boundary.

[0114] Aspect 2. The computer-implemented method of aspect 1, wherein: the excipient comprises the polymer, and the polymer comprises a first monomer and a second monomer; the at least one output parameter comprises a first free energy of mixing value Xi ■> and a second free energy of mixing value2; the method further comprises determining, via the one or more processors, a combined free energy of mixing value Xcombmed based on the first free energy of mixing valueand the second free energy of mixing value2; and the building of the phase boundary comprises, via the one or more processors, building the phase boundary according to a thermodynamic equation, wherein the thermodynamic equation includes the combined free energy of mixing value Xcombmedas avariable.

[0115] Aspect 3. The computer-implemented method of any one of aspects 1-2, further comprising: building, via the one or more processors, a free energy parameter plot of the combined free energy of mixing value Xcombined-’ wherein the free energy parameter plot conveys a relationship between the combined free energy of mixing value Xcombined and temperature; and displaying, via the one or more processors, the free energy parameter plot.

[0116] Aspect 4. The computer-implemented method of any one of aspects 1-3, wherein: the building of the phase boundary further comprises, via the one or more processors, building the phase boundary according to a thermodynamic equation.

[0117] Aspect 5. The computer-implemented method of any one of aspects 1-4, further comprising: building, via the one or more processors, a glass-transition boundary based on the at least one output parameter.

[0118] Aspect 6. The computer-implemented method of any one of aspects 1-5, further comprising: building, via the one or more processors, a solid-liquid equilibrium (SLE) phase boundary and / or a liquid-liquid equilibrium (LLE) phase boundary based on the at least one output parameter.

[0119] Aspect 7. The computer-implemented method of any one of aspects 1-6, further comprising displaying, via the one or more processors, on a display a phase diagram of a formulation, wherein the phase diagram of the formulation includes an indication of the phase boundary.

[0120] Aspect 8. The computer-implemented method of any one of aspects 1-7, further comprising displaying, via the one or more processors, on a display a phase diagram of a formulation, and wherein the phase diagram of the formulation includes: an indication of the phase boundary; a first axis of Al weight fraction; and a second axis of temperature.

[0121] Aspect 9. The computer-implemented method of any one of aspects 1-8, wherein the second chemical component comprises an excipient comprising a polymer, a lipid, and / or a plasticizer, and wherein the method further comprises presenting, via the one or more processors, on a display, a dropdown selection of a plurality excipients including the excipient; and wherein the information of the excipient is received according to a selection of the excipient from the dropdown menu.

[0122] Aspect 10. The computer-implemented method of any one of aspects 1-9, wherein: the first chemical component comprises an active ingredient (Al), and the information of the first chemical component comprises information of the Al; the received information of the Al includes a user-input glass transition temperature of the Al; and the inputting the information of the Al and the information of the excipient into the at least one machine learning model includes inputting the user-input glass transition temperature of the Al into the at least one machine learning model.

[0123] Aspect 11. The computer-implemented method of any one of aspects 1-10, wherein: the first chemical component comprises an active ingredient (Al), and the information of the first chemical component comprises information of the Al; the received information of the Al includes a user-input melting point temperature of the Al; and the inputting the information of the Al and the information of the excipient into the at least one machine learning model includes inputting the user-input melting point temperature of the Al into the at least one machine learning model.

[0124] Aspect 12. The computer-implemented method of any one of aspects 1-11, wherein: the first chemical component comprises an active ingredient (Al), and the information of the first chemical component comprises information of the Al; the received information of the Al includes a user-input enthalpy of fusion of the Al; and the inputting the information of the Al and the information of the excipient into the at least one machine learning model includes inputting the user-input enthalpy of fusion of the Al into the at least one machine learning model.

[0125] Aspect 13. The computer-implemented method of any one of aspects 1-12, wherein: the first chemical component comprises an active ingredient (Al), and the information of the first chemical component comprises information of the Al; the received information of the Al includes a user-input density of the Al; and the inputting the information of the Al and the information of the excipient into the at least one machine learning model includes inputting the user-input density of the Al into the at least one machine learning model.

[0126] Aspect 14. The computer-implemented method of any one of aspects 1-13, wherein: the first chemical component comprises an active ingredient (Al), and the information of the first chemical component comprises information of the Al; the received information of the Al includes a molecular weight of the Al; and the inputting the information of the Al and the information of the excipient into the at least one machine learning model includes inputting the molecular weight of the Al into the at least one machine learning model.

[0127] Aspect 15. The computer-implemented method of any one of aspects 1-14, wherein: the second chemical component comprises an excipient comprising a polymer, a lipid, and / or a plasticizer, and the received information of the second chemical component comprises information of the excipient; the received information of the excipient includes a user-input glass transition temperature of the excipient; and the inputting the information of the Al and the information of the excipient into the at least one machine learning model includes inputting the user-input glass transition temperature of the excipient into the at least one machine learning model.

[0128] Aspect 15A. The computer-implemented method of any one of aspects 1-15, wherein the received information of the excipient includes a user-input : a molecular weight of the excipient; an architecture of the excipient; a molecular weight Polydispersity of the excipient; sequence dispersity information (e.g., blockiness of the polymer) of the excipient; side chain distributions of the excipient; impurities in the excipient;Hansen solubility parameters of the excipient; historical melting point depression experiments with other AIs of the excipient; other free energy of mixing values (e.g., x values) between the excipient and other solvents, AIs, or excipients themselves; a glass transition temperature of the excipient; degree of crystallinity of the excipient; loss and storage modulus information on the excipient;synthesis information (e.g., how the excipient was created) of the excipient; and / or indicating how the excipient was processed post synthesis.

[0129] Aspect 16. The computer-implemented method of any one of aspects 1-15, wherein the second chemical component comprises an excipient comprising a polymer, a lipid, and / or a plasticizer, and wherein the method further comprises: receiving, via the one or more processors, identifying information of the excipient; and retrieving, via the one or more processors, from a database, and based on the identifying information of the excipient: a glass transition temperature of the excipient; and / or a density of the excipient; and wherein the inputting the information of the Al and the information of the excipient into the at least one machine learning model includes inputting the retrieved glass transition temperature and / or density of the excipient into the at least one machine learning model.

[0130] Aspect 17. The computer-implemented method of any one of aspects 1-16, wherein the second chemical component comprises an excipient comprising a polymer, and the polymer comprises: (i) a homopolymer, (ii) a block polymer, (ii) a random copolymer, or (iv) a triblock polymer, (v) a graft polymer, (vi) a bottlebrush, or a (vii) star polymer.

[0131] Aspect 18. The computer-implemented method of any one of aspects 1-17, wherein: the first chemical component comprises an active ingredient (Al), and the information of the first chemical component comprises information of the Al; the information of an Al is received in a simplified molecular input line entry system (SMILES) form; and the information of the excipient is received in the SMILES form.

[0132] Aspect 19. The computer-implemented method of any one of aspects 1-18, wherein:(i) the first chemical component comprises an active ingredient (Al), and (ii) the second chemical component comprises an excipient comprising a polymer, a lipid, and / or a plasticizer: the at least one machine learning model comprises: a first machine learning model trained to determine an Al melting point, a second machine learning model trained to determine an Al enthalpy of fusion, a third machine learning model trained to determine a glass transition temperature of an Al, excipient and / or formulation, a fourth machine learning model trained todetermine free energy of mixing of a formulation, and / or a fifth machine learning model trained to determine molar volume of an Al and / or excipient; and wherein the method further comprises: training, via the one or more processors, the first, second, third, fourth, and / or fifth machine learning models by inputting historical information into the first, second, third, fourth, and / or fifth machine learning models, the historical information comprising: (i) independent variables comprising (a) historical glass transition temperatures of historical AIs, (b) historical melting points of the historical AIs, (c) historical enthalpy of fusions of the historical AIs, (d) historical densities of the historical AIs, (e) historical molar weights of the historical AIs, (f) historical molar volumes of the historical AIs, (g) historical glass transition temperatures of historical excipients, (h) historical molar weights of the historical excipients, (i) historical molar volumes of the historical excipients, (j) historical densities of the historical excipients, and / or (k) historical excess free energy of mixing values, and / or (ii) dependent variables comprising historical output parameters.

[0133] Aspect 20. The computer-implemented method of aspect 19, wherein the historical output parameters comprise historical free energy of mixing values.

[0134] Aspect 21. The computer-implemented method of any one of aspects 1-20, wherein:(i) the first chemical component comprises an active ingredient (Al), and (ii) the second chemical component comprises an excipient comprising a polymer, a lipid, and / or a plasticizer; the at least one machine learning model comprises: a first machine learning model trained to determine an Al melting point, a second machine learning model trained to determine an Al enthalpy of fusion, a third machine learning model trained to determine a glass transition temperature of an Al, excipient and / or formulation, a fourth machine learning model trained to determine free energy of mixing of a formulation, and / or a fifth machine learning model trained to determine molar volume of an Al and / or excipient; and the method further comprises: during an initial training phase, training, via the one or more processors, the first, second, third, fourth, and / or fifth machine learning models by inputting first historical information into the first, second, third, fourth, and / or fifth machine learning models, the first historical information comprising: (i) independent variables comprising (a) historical glass transition temperatures of historical AIs, (b) historical melting points of the historical AIs, (c) historicalenthalpy of fusions of the historical AIs, (d) historical densities of the historical AIs, (e) historical molar weights of the historical AIs, (f) historical molar volumes of the historical AIs,(g) historical glass transition temperatures of historical excipients, (h) historical molar weights of the historical excipients, (i) historical molar volumes of the historical excipients, (j) historical densities of the historical excipients, (k) historical excess free energy of mixing values, and / or (ii) dependent variables comprising historical output parameters; receiving, via the one or more processors, a user-input output parameter corresponding to a formulation of the Al and the excipient; and during a subsequent training phase, training, via the one or more processors, the machine learning model further by inputting second historical information into the first, second, third, fourth, and / or fifth machine learning models, the second historical information comprising a dependent variable comprising the user-input output parameter.

[0135] Aspect 22. The computer-implemented method of aspect 21, wherein the second historical information further comprises independent variables comprising:(a) a glass transition temperature of the Al;(b) a melting point of the Al;(c) an enthalpy of fusion of the Al;(d) a density of the Al;(e) a molar weight of the Al;(f) a molar volume of the Al;(e) a glass transition temperature of the excipient;(h) a molar weight of the excipient;(i) a molar volume of the excipient(j) historical density of the excipient; and / or(k) historical excess free energy of mixing values.

[0136] Aspect 23. A computer system for building a phase boundary, the computer system comprising one or more processors configured to: receive information of a first chemical component; receive information of a second chemical component; produce at least one output parameter by inputting the information of the first chemical component and the information of the second chemical component into at least one machinelearning model; and build a phase boundary based on the at least one output parameter.

[0137] Aspect 24. The computer system of aspect 23, wherein: the excipient comprises the polymer, and the polymer comprises a first monomer and a second monomer; the at least one output parameter comprises a first free energy of mixing value Xi , and a second free energy of mixing value2; the one or more processors are further configured to determine a combined free energy of mixing value Xcombined based on the first free energy of mixing value Xi, and the second free energy of mixing value2; and the one or more processors are further configured to build the phase boundary by building the phase boundary according to a thermodynamic equation, wherein the thermodynamic equation includes the combined free energy of mixing value Xcombined as a variable.

[0138] Aspect 25. The computer system of any one of aspects 23-24, wherein the one or more processors are further configured to: build a free energy parameter plot of the combined free energy of mixing value Xcombined - wherein the free energy parameter plot conveys a relationship between the combined free energy of mixing value Xcombined and temperature; and display the free energy parameter plot.

[0139] Aspect 26. The computer system of any one of aspects 23-25, wherein the one or more processors are further configured to: build the phase boundary further by building the phase boundary according to a thermodynamic equation.

[0140] Aspect 27. The computer system of any one of aspects 23-26, wherein the one or more processors are further configured to: build a glass-transition boundary based on the at least one output parameter.

[0141] Aspect 28. The computer system of any one of aspects 23-27, wherein the one or more processors are further configured to: build a solid-liquid equilibrium (SLE) phase boundary and / or a liquid-liquid equilibrium (LLE) phase boundary based on the at least one output parameter.

[0142] Aspect 29. The computer system of any one of aspects 23-28, wherein the one or more processors are further configured to display, on a display, a phase diagram of a formulation, wherein the phase diagram of the formulation includes an indication of the phase boundary.

[0143] Aspect 30. The computer system of any one of aspects 23-29, wherein the one or more processors are further configured to display, on a display, a phase diagram of a formulation, and wherein the phase diagram of the formulation includes: an indication of the phase boundary; a first axis of Al weight fraction; and a second axis of temperature.

[0144] Aspect 31. The computer system of any one of aspects 23-30, wherein the second chemical component comprises an excipient comprising a polymer, a lipid, and / or a plasticizer, and wherein the method, and wherein the one or more processors are further configured to present, on a display, a dropdown selection of a plurality excipients including the excipient; and wherein the information of the excipient is received according to a selection of the excipient from the dropdown menu.

[0145] Aspect 32. The computer system of any one of aspects 23-31, wherein: the first chemical component comprises an active ingredient (Al), and the information of the first chemical component comprises information of the Al; the received information of the Al includes a user-input glass transition temperature of the Al; and the inputting the information of the Al and the information of the excipient into the at least one machine learning model includes inputting the user-input glass transition temperature of the Al into the at least one machine learning model.

[0146] Aspect 33. The computer system of any one of aspects 23-32, wherein: the first chemical component comprises an active ingredient (Al), and the information of the first chemical component comprises information of the Al; the received information of the Al includes a user-input melting point temperature of the Al; and the inputting the information of the Al and the information of the excipient into the atleast one machine learning model includes inputting the user-input melting point temperature of the Al into the at least one machine learning model.

[0147] Aspect 34. The computer system of any one of aspects 23-33, wherein: the second chemical component comprises an excipient comprising a polymer, a lipid, and / or a plasticizer, and the received information of the second chemical component comprises information of the excipient; the received information of the Al includes a user-input enthalpy of fusion of the Al; and the inputting the information of the Al and the information of the excipient into the at least one machine learning model includes inputting the user-input enthalpy of fusion of the Al into the at least one machine learning model.

[0148] Aspect 35. The computer system of any one of aspects 23-34, wherein: the received information of the Al includes a user-input density of the Al; and the inputting the information of the Al and the information of the excipient into the at least one machine learning model includes inputting the user-input density of the Al into the at least one machine learning model.

[0149] Aspect 36. The computer system of any one of aspects 23-35, wherein: the received information of the Al includes a molecular weight of the Al; and the inputting the information of the Al and the information of the excipient into the at least one machine learning model includes inputting the molecular weight of the Al into the at least one machine learning model.

[0150] Aspect 37. The computer system of any one of aspects 23-36, wherein: the received information of the excipient includes a user-input glass transition temperature of the excipient; and the inputting the information of the Al and the information of the excipient into the at least one machine learning model includes inputting the user-input glass transition temperature of the excipient into the at least one machine learning model.

[0151] Aspect 38. The computer system of any one of aspects 23-37, wherein the second chemical component comprises an excipient comprising a polymer, a lipid, and / or a plasticizer, and wherein the one or more processors are further configured to: receive identifying information of the excipient; andretrieve, from a database, and based on the identifying information of the excipient: a glass transition temperature of the excipient; and / or a density of the excipient; and wherein the inputting the information of the Al and the information of the excipient into the at least one machine learning model includes inputting the retrieved glass transition temperature and / or density of the excipient into the at least one machine learning model.

[0152] Aspect 39. The computer system of any one of aspects 23-38, wherein the second chemical component comprises an excipient comprising a polymer, and the polymer comprises: (i) a homopolymer, (ii) a block polymer, (ii) a random copolymer, or (iv) a triblock polymer, (v) a graft polymer, (vi) a bottlebrush, or a (vii) star polymer.

[0153] Aspect 40. The computer system of any one of aspects 23-39, wherein: the first chemical component comprises an active ingredient (Al), and the information of the first chemical component comprises information of the Al; the information of an Al is received in a simplified molecular input line entry system (SMILES) form; and the information of the excipient is received in the SMILES form.

[0154] Aspect 41. The computer system of any one of aspects 23-40, wherein: (i) the first chemical component comprises an active ingredient (Al), and (ii) the second chemical component comprises an excipient comprising a polymer, a lipid, and / or a plasticizer; the at least one machine learning model comprises: a first machine learning model trained to determine an Al melting point, a second machine learning model trained to determine an Al enthalpy of fusion, a third machine learning model trained to determine a glass transition temperature of an Al, excipient and / or formulation, a fourth machine learning model trained to determine free energy of mixing of a formulation, and / or a fifth machine learning model trained to determine molar volume of an Al and / or excipient; and wherein the one or more processors are further configured to: train the first, second, third, fourth, and / or fifth machine learning models by inputting historical information into the first, second, third, fourth, and / or fifth machine learning models, the historical information comprising: (i) independent variables comprising (a) historical glass transition temperatures of historical AIs, (b) historical melting points of the historical AIs, (c)historical enthalpy of fusions of the historical AIs, (d) historical densities of the historical AIs, (e) historical molar weights of the historical AIs, (f) historical molar volumes of the historical AIs, (g) historical glass transition temperatures of historical excipients, (h) historical molar weights of the historical excipients, (i) historical molar volumes of the historical excipients, (j) historical densities of the historical excipients, (k) historical excess free energy of mixing values, and / or (ii) dependent variables comprising historical output parameters.

[0155] Aspect 42. The computer system of aspect 41, wherein the historical output parameters comprise historical free energy of mixing values.

[0156] Aspect 43. The computer system of any one of aspects 23-42, wherein:(i) the first chemical component comprises an active ingredient (Al), and (ii) the second chemical component comprises an excipient comprising a polymer, a lipid, and / or a plasticizer, the at least one machine learning model comprises: a first machine learning model trained to determine an Al melting point, a second machine learning model trained to determine an Al enthalpy of fusion, a third machine learning model trained to determine a glass transition temperature of an Al, excipient and / or formulation, a fourth machine learning model trained to determine free energy of mixing of a formulation, and / or a fifth machine learning model trained to determine molar volume of an Al and / or excipient; and wherein the one or more processors are further configured to: during an initial training phase, train the first, second, third, fourth, and / or fifth machine learning models by inputting first historical information into the first, second, third, fourth, and / or fifth machine learning models, the first historical information comprising: (i) independent variables comprising (a) historical glass transition temperatures of historical AIs, (b) historical melting points of the historical AIs, (c) historical enthalpy of fusions of the historical AIs, (d) historical densities of the historical AIs, (e) historical molar weights of the historical AIs, (f) historical molar volumes of the historical AIs, (g) historical glass transition temperatures of historical excipients, (h) historical molar weights of the historical excipients, (i) historical molar volumes of the historical excipients, (j) historical densities of the historical excipients, (k) historical excess free energy of mixing values, and / or (ii) dependent variables comprising historical output parameters; receive a user-input output parameter corresponding to a formulation of the Al and the excipient; andduring a subsequent training phase, train the machine learning model further by inputting second historical information into the machine learning model, the second historical information comprising a dependent variable comprising the user-input output parameter.

[0157] Aspect 44. The computer system of aspect 43, wherein the second historical information further comprises independent variables comprising:(a) a glass transition temperature of the Al;(b) a melting point of the Al;(c) an enthalpy of fusion of the Al;(d) a density of the Al;(e) a molar weight of the Al;(f) a molar volume of the Al;(e) a glass transition temperature of the excipient;(h) a molar weight of the excipient; and / or(i) a molar volume of the excipient;(j) historical density of the excipient; and / or(k) historical excess free energy of mixing values.

[0158] Aspect 45. A computer device for building a phase boundary, the computer device comprising: one or more processors; and one or more memories, the one or more memories having stored thereon computerexecutable 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; produce at least one output parameter by inputting the information of the first chemical component and the information of the second chemical component into at least one at least one machine learning model; and build a phase boundary based on the at least one output parameter.

[0159] Aspect 46. The computer device of aspect 45, the one or more memories having stored thereon computer executable instructions that, when executed by the one or moreprocessors, cause the one or more processors to: build a free energy parameter plot of the combined free energy of mixing value Xcombined^ wherein the free energy parameter plot conveys a relationship between the combined free energy of mixing value X combined and temperature; and display the free energy parameter plot.Additional Exemplary Embodiments - Ranking Formulations

[0160] Aspect 100. A computer-implemented method for ranking formulations, the method comprising: receiving, via one or more processors, information of a first chemical component; receiving, via the one or more processors, information of a plurality of second chemical components; creating, via the one or more processors, respective formulations for the first chemical component paired with respective second chemical components of the second chemical components; producing, via the one or more processors, an output parameter for each respective formulation by inputting the information of the first chemical component and information of the respective second chemical components into at least one machine learning model; and ranking, via the one or more processors, the respective formulations based on the produced output parameters.

[0161] Aspect 102. The computer-implemented method of aspect 1, further comprising determining, via the one or more processors, based on the produced output parameters, respective glass transition temperatures of the respective formulations; and wherein the ranking the respective formulations further comprises ranking, via the one or more processors, the respective formulations based on the respective glass transition temperatures.

[0162] Aspect 103. The computer-implemented method of aspect 102, further comprising: receiving, via the one or more processors, a user-input reference temperature; determining, via the one or more processors, respective differences between the reference temperature and the respective glass transition temperatures; anddisplaying, via the one or more processors, on a display: (i) the respective formulations ranked according to the ranking, and (ii) the determined respective differences between the reference temperature and the respective glass transition temperatures.

[0163] Aspect 104. The computer-implemented method of any one of aspects 100-103, further comprising: receiving, via the one or more processors, a user-input reference temperature; building, via the one or more processors, based on the at least one output parameter a glass transition phase boundary of a first formulation of the respective formulations; and displaying, via the one or more processors, on a display, a graph including the glass transition phase boundary and an indication of a difference between the glass transition phase boundary, and the reference temperature.

[0164] Aspect 105. The computer-implemented method of any one of aspects 100-104, further comprising determining, via the one or more processors, based on the produced output parameters, respective combined free energy of mixing values of the respective formulations; and wherein the ranking the respective formulations further comprises ranking, via the one or more processors, the respective formulations based on the respective combined free energy of mixing values.

[0165] Aspect 106. The computer-implemented method of any one of aspects 100-105, wherein: the second chemical components include an excipient comprising a polymer, and the polymer comprises a first monomer and a second monomer; the output parameter of the formulation corresponding to the excipient comprises a first free energy of mixing value Xi and the method further comprises: determining, via the one or more processors, a combined free energy of mixing value Xcombined based on the first free energy of mixing value Xi, and a second free energy of mixing value X2 ■> wherein the second free energy of mixing value2is determined by inputting the information of the first chemical component and information of the excipient into the at least one machine learning model;building, via the one or more processors, a free energy parameter plot of the combined free energy of mixing value Xcombined - wherein the free energy parameter plot conveys a relationship between the combined free energy of mixing value Xcombined and temperature; receiving, via the one or more processors, a user-input reference temperature; and displaying, via the one or more processors: (i) the free energy parameter plot, and (ii) an indication of the combined free energy of mixing value Xcombinedatthe reference temperature.

[0166] Aspect 107. The computer-implemented method of any one of aspects 100-106, further comprising: building, via the one or more processors, a phase boundary according to a thermodynamic equation including the combined free energy of mixing value Xcombinedas avariable; and displaying, via the one or more processors, on a display, the phase boundary.

[0167] Aspect 108. The computer-implemented method of any one of aspects 100-107, further comprising determining, via the one or more processors, based on the produced output parameters, respective stable active ingredient (Al)-loads of the respective formulations; and wherein the ranking the respective formulations further comprises ranking, via the one or more processors, the respective formulations based on the respective stable Al-loads.

[0168] Aspect 109. The computer-implemented method of any one of aspects 100-107, further comprising: receiving, via the one or more processors, a user-input reference temperature; building, via the one or more processors, based on the at least one output parameter a solid-liquid equilibrium (SLE) phase boundary of a first formulation of the respective formulations; and displaying, via the one or more processors, on a display, a graph including the SLE phase boundary and an indication of a stable ALload of the first formulation at the reference temperature.

[0169] Aspect 110. The computer-implemented method of any one of aspects 100-109, wherein: (i) the first chemical component comprises an active ingredient (Al), and (ii) the plurality of second chemical components include an excipient comprising a polymer, a lipid, and / or a plasticizer.

[0170] Aspect 111. The computer-implemented method of any one of aspects 100-110, further comprising displaying, via the one or more processors, on a display, the ranked respective formulations according to the ranking.

[0171] Aspect 112. The computer-implemented method of any one of aspects 100-111, further comprising; presenting, via the one or more processors, on a display: (i) a first option to rank the respective formulations according to glass transition temperature, (ii) a second option to rank the respective formations according to combined free energy of mixing, and (iii) a third option to rank the respective formulations according to stable active ingredient (Al)-load; and receiving, via the one or more processors, a selection of the first option, the second option, or the third option; and wherein the ranking the respective formulations further comprises ranking, via the one or more processors, the respective formulations according to the selection; and wherein the method further comprising displaying, via the one or more processors, the ranked respective formulations according to the ranking, and the selection.

[0172] Aspect 113. The computer-implemented method of any one of aspects 100-112, wherein:(i) the first chemical component comprises an active ingredient (Al), and (ii) the plurality of second chemical components include an excipient comprising a polymer, a lipid, and / or a plasticizer; the at least one machine learning model comprises: a first machine learning model trained to determine an Al melting point, a second machine learning model trained to determine an Al enthalpy of fusion, a third machine learning model trained to determine a glass transition temperature of an Al, excipient and / or formulation, a fourth machine learning model trained to determine free energy of mixing of a formulation, and / or a fifth machine learning model trained to determine molar volume of an Al and / or excipient; and wherein the method further comprises: training, via the one or more processors, the first, second, third, fourth, and / or fifth machine learning models by inputting historical information into the first, second, third, fourth, and / or fifth machine learning models, the historical information comprising: (i) independent variables comprising (a) historical glass transition temperatures of historical AIs, (b) historicalmelting points of the historical AIs, (c) historical enthalpy of fusions of the historical AIs, (d) historical densities of the historical AIs, (e) historical molar weights of the historical AIs, (f) historical molar volumes of the historical AIs, (g) historical glass transition temperatures of historical excipients, (h) historical molar weights of the historical excipients, (i) historical molar volumes of the historical excipients, (j) historical densities of the historical excipients, and / or (k) historical excess free energy of mixing values, and / or (ii) dependent variables comprising historical output parameters.

[0173] Aspect 114. A computer system for ranking formulations, the computer system comprising one or more processors configured to: receive information of a first chemical component; receive information of a plurality of second chemical components; create respective formulations for the first chemical component paired with respective second chemical components of the second chemical components; produce an output parameter for each respective formulation by inputting the information of the first chemical component and information of the respective second chemical components into at least one machine learning model; and rank the respective formulations based on the produced output parameters.

[0174] Aspect 115. The computer system of aspect 114, wherein the one or more processors are further configured to: determine, based on the produced output parameters, respective glass transition temperatures of the respective formulations; and rank the respective formulations by ranking the respective formulations based on the respective glass transition temperatures.

[0175] Aspect 116. The computer system of aspect 115, further comprising a display, and wherein the one or more processors are further configured to: receive a user-input reference temperature; determine respective differences between the reference temperature and the respective glass transition temperatures; and display, on the display: (i) the respective formulations ranked according to the ranking,and (ii) the determined respective differences between the reference temperature and the respective glass transition temperatures.

[0176] Aspect 117. The computer system of any one of aspects 114-116, wherein: (i) the first chemical component comprises an active ingredient (Al), and (ii) the plurality of second chemical components include an excipient comprising a polymer, a lipid, and / or a plasticizer.

[0177] Aspect 118. A computer device for ranking formulations, the computer device comprising: one or more processors; and one or more memories, the one or more memories having stored thereon computerexecutable 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 plurality of second chemical components; create respective formulations for the first chemical component paired with respective second chemical components of the second chemical components; produce an output parameter for each respective formulation by inputting the information of the first chemical component and information of the respective second chemical components into at least one machine learning model; and rank the respective formulations based on the produced output parameters.

[0178] Aspect 119. The computer device of aspect 118, the one or more memories having stored thereon computer executable instructions that, when executed by the one or more processors, cause the one or more processors to: determine, based on the produced output parameters, respective combined free energy of mixing values of the respective formulations; and rank the respective formulations by ranking the respective formulations based on the respective combined free energy of mixing values.

[0179] Aspect 120. The computer device of any one of aspects 118-119, the one or more memories having stored thereon computer executable instructions that, when executed by the one or more processors, cause the one or more processors to:determine, based on the produced output parameters, respective stable active ingredient (Al)-loads of the respective formulations; and rank the respective formulations by ranking the respective formulations based on the respective stable active ingredient (Al)-loads.OTHER MATTERS

[0180] 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.

[0181] 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.

[0182] 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.

[0183] 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.

[0184] 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 gate array (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.

[0185] 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.

[0186] 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).

[0187] 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.

[0188] 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.

[0189] 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 transformsdata 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.

[0190] 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.

[0191] 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 with each other, but yet still co-operate or interact with each other. The embodiments are not limited in this context.

[0192] 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).

[0193] 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.

[0194] 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 andcomponents 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.

[0195] 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.

[0196] 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.

[0197] 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.

[0198] 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

WHAT IS CLAIMED:

1. A computer-implemented method for building a phase boundary, the 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; producing, via the one or more processors, at least one output parameter by inputting the information of the first chemical component and the information of the second chemical component into at least one machine learning model; and building, via the one or more processors, a phase boundary based on the at least one output parameter.

2. The computer-implemented method of claim 1, wherein: the second chemical component comprises an excipient comprising a polymer, and the polymer comprises a first monomer and a second monomer; the at least one output parameter comprises a first free energy of mixing value Xi , and a second free energy of mixing value2; the method further comprises determining, via the one or more processors, a combined free energy of mixing value X combined based on the first free energy of mixing value Xi, and the second free energy of mixing value2; and the building of the phase boundary comprises, via the one or more processors, building the phase boundary according to a thermodynamic equation, wherein the thermodynamic equation includes the combined free energy of mixing value Xcombined as a variable.

3. The computer-implemented method of claim 2, further comprising: building, via the one or more processors, a free energy parameter plot of the combined free energy of mixing value Xcombined’ wherein the free energy parameter plot conveys a relationship between the combined free energy of mixing value Xcombined and temperature; and displaying, via the one or more processors, the free energy parameter plot.

4. The computer-implemented method of claim 1, wherein:the building of the phase boundary further comprises, via the one or more processors, building the phase boundary according to a thermodynamic equation.

5. The computer-implemented method of claim 1, further comprising: building, via the one or more processors, a glass-transition boundary based on the at least one output parameter.

6. The computer-implemented method of claim 1, further comprising: building, via the one or more processors, a solid-liquid equilibrium (SLE) phase boundary and / or a liquid-liquid equilibrium (LLE) phase boundary based on the at least one output parameter.

7. The computer-implemented method of claim 1, further comprising displaying, via the one or more processors, on a display, a phase diagram of a formulation, wherein the phase diagram of the formulation includes an indication of the phase boundary.

8. The computer-implemented method of claim 1, further comprising displaying, via the one or more processors, on a display, a phase diagram of a formulation, and wherein the phase diagram of the formulation includes: an indication of the phase boundary; a first axis of Al weight fraction; and a second axis of temperature.

9. The computer-implemented method of claim 1, wherein the second chemical component comprises an excipient comprising a polymer, a lipid, and / or a plasticizer, and wherein the method further comprises presenting, via the one or more processors, on a display, a dropdown selection of a plurality excipients including the excipient; and wherein the information of the excipient is received according to a selection of the excipient from the dropdown menu.

10. The computer-implemented method of claim 1, wherein:the first chemical component comprises an active ingredient (Al), and the information of the first chemical component comprises information of the Al; the received information of the Al includes a user-input glass transition temperature of the Al; and the inputting the information of the Al and the information of the excipient into the at least one machine learning model includes inputting the user-input glass transition temperature of the Al into the at least one machine learning model.

11. The computer-implemented method of claim 1, wherein: the first chemical component comprises an active ingredient (Al), and the information of the first chemical component comprises information of the Al; the received information of the Al includes a user-input melting point temperature of the Al; and the inputting the information of the Al and the information of the excipient into the at least one machine learning model includes inputting the user-input melting point temperature of the Al into the at least one machine learning model.

12. The computer-implemented method of claim 1, wherein: the first chemical component comprises an active ingredient (Al), and the information of the first chemical component comprises information of the Al; the received information of the Al includes a user-input enthalpy of fusion of the Al; and the inputting the information of the Al and the information of the excipient into the at least one machine learning model includes inputting the user-input enthalpy of fusion of the Al into the at least one machine learning model.

13. The computer-implemented method of claim 1, wherein: the first chemical component comprises an active ingredient (Al), and the information of the first chemical component comprises information of the Al; the received information of the Al includes a user-input density of the Al; andthe inputting the information of the Al and the information of the excipient into the at least one machine learning model includes inputting the user-input density of the Al into the at least one machine learning model.

14. The computer-implemented method of claim 1, wherein: the first chemical component comprises an active ingredient (Al), and the information of the first chemical component comprises information of the Al; the received information of the Al includes a molecular weight of the Al; and the inputting the information of the Al and the information of the excipient into the at least one machine learning model includes inputting the molecular weight of the Al into the at least one machine learning model.

15. The computer-implemented method of claim 1, wherein: the second chemical component comprises an excipient comprising a polymer, a lipid, and / or a plasticizer, and the received information of the second chemical component comprises information of the excipient; the received information of the excipient includes a user-input glass transition temperature of the excipient; and the inputting the information of the Al and the information of the excipient into the at least one machine learning model includes inputting the user-input glass transition temperature of the excipient into the at least one machine learning model.

16. The computer-implemented method of claim 1, wherein the second chemical component comprises an excipient comprising a polymer, a lipid, and / or a plasticizer, and wherein the method further comprise: receiving, via the one or more processors, identifying information of the excipient; and retrieving, via the one or more processors, from a database, and based on the identifying information of the excipient: a glass transition temperature of the excipient; and / or a density of the excipient; andwherein the inputting the information of the Al and the information of the excipient into the at least one machine learning model includes inputting the retrieved glass transition temperature and / or density of the excipient into the at least one machine learning model.

17. The computer-implemented method of claim 1, wherein the second chemical component comprises an excipient comprising a polymer, and the polymer comprises: (i) a homopolymer, (ii) a block polymer, (ii) a random copolymer, or (iv) a triblock polymer, (v) a graft polymer, (vi) a bottlebrush, or a (vii) star polymer.

18. The computer-implemented method of claim 1, wherein: the first chemical component comprises an active ingredient (Al), and the information of the first chemical component comprises information of the Al; the information of an Al is received in a simplified molecular input line entry system (SMILES) form; and the information of the excipient is received in the SMILES form.

19. The computer-implemented method of claim 1, wherein:(i) the first chemical component comprises an active ingredient (Al), and (ii) the second chemical component comprises an excipient comprising a polymer, a lipid, and / or a plasticizer; the at least one machine learning model comprises: a first machine learning model trained to determine an Al melting point, a second machine learning model trained to determine an Al enthalpy of fusion, a third machine learning model trained to determine a glass transition temperature of an Al, excipient and / or formulation, a fourth machine learning model trained to determine free energy of mixing of a formulation, and / or a fifth machine learning model trained to determine molar volume of an Al and / or excipient; and wherein the method further comprises: training, via the one or more processors, the first, second, third, fourth, and / or fifth machine learning models by inputting historical information into the first, second, third, fourth, and / or fifth machine learning models, the historical information comprising: (i) independent variables comprising (a) historical glass transition temperatures of historical AIs, (b) historical melting points of the historical AIs, (c) historical enthalpy of fusions of the historical AIs, (d)historical densities of the historical AIs, I historical molar weights of the historical AIs, (f) historical molar volumes of the historical AIs, (g) historical glass transition temperatures of historical excipients, (h) historical molar weights of the historical excipients, (i) historical molar volumes of the historical excipients, (j) historical densities of the historical excipients, and / or (k) historical excess free energy of mixing values, and / or (ii) dependent variables comprising historical output parameters.

20. The computer-implemented method of claim 19, wherein the historical output parameters comprise historical free energy of mixing values.

21. The computer-implemented method of claim 1, wherein:(i) the first chemical component comprises an active ingredient (Al), and (ii) the second chemical component comprises an excipient comprising a polymer, a lipid, and / or a plasticizer; the at least one machine learning model comprises: a first machine learning model trained to determine an Al melting point, a second machine learning model trained to determine an Al enthalpy of fusion, a third machine learning model trained to determine a glass transition temperature of an Al, excipient and / or formulation, a fourth machine learning model trained to determine free energy of mixing of a formulation, and / or a fifth machine learning model trained to determine molar volume of an Al and / or excipient; and the method further comprises: during an initial training phase, training, via the one or more processors, the first, second, third, fourth, and / or fifth machine learning models by inputting first historical information into the first, second, third, fourth, and / or fifth machine learning models, the first historical information comprising: (i) independent variables comprising (a) historical glass transition temperatures of historical AIs, (b) historical melting points of the historical AIs, (c) historical enthalpy of fusions of the historical AIs, (d) historical densities of the historical AII(e) historical molar weights of the historical AIs, (f) historical molar volumes of the historical AIs, (g) historical glass transition temperatures of historical excipients, (h) historical molar weights of the historical excipients, (i) historical molar volumes of the historical excipients, (j) historical densities of the historical excipients, (k) historical excess free energy of mixing values, and / or (ii) dependent variables comprising historical output parameters;receiving, via the one or more processors, a user-input output parameter corresponding to a formulation of the Al and the excipient; and during a subsequent training phase, training, via the one or more processors, the machine learning model further by inputting second historical information into the first, second, third, fourth, and / or fifth machine learning models, the second historical information comprising a dependent variable comprising the user-input output parameter.

22. The computer-implemented method of claim 21, wherein the second historical information further comprises independent variables comprising:(a) a glass transition temperature of the Al;(b) a melting point of the Al;(c) an enthalpy of fusion of the Al;(d) a density of the Al;(e) a molar weight of the Al;(f) a molar volume of the Al;(e) a glass transition temperature of the excipient;(h) a molar weight of the excipient;(i) a molar volume of the excipient;(j) historical density of the excipient; and / or(k) historical excess free energy of mixing values.

23. A computer system for building a phase boundary, the computer system comprising one or more processors configured to: receive information of an active ingredient (Al); receive information of an excipient, wherein the excipient comprises a polymer, a lipid, or a plasticizer; produce at least one output parameter by inputting the information of the first chemical component and the information of the second chemical component into at least one machine learning model; and build a phase boundary based on the at least one output parameter.

24. A computer device for building a phase boundary, the computer device comprising: one or more processors; and one or more memories, the one or more memories having stored thereon computerexecutable 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, wherein the excipient comprises a polymer, a lipid, or a plasticizer; produce at least one output parameter by inputting the information of the first chemical component and the information of the second chemical component into at least one machine learning model; and build phase boundary based on the at least one output parameter.

25. The computer device of claim 24, wherein the phase boundary is a solid-liquid equilibrium (SLE) phase boundary, and / or a liquid-liquid equilibrium (LLE) phase boundary.

26. The computer device of claim 24, wherein: (i) the first chemical component comprises an active ingredient (Al), and (ii) the second chemical component comprises an excipient comprising a polymer, a lipid, and / or a plasticizer.

Images