Method and system for targeted drug discovery

Abstract

To provide a method, system and non-transitory computer readable medium for targeted drug discovery.SOLUTION: A method comprises: obtaining a receptor panel, where the receptor panel specifies a plurality of targets and a plurality of anti-targets; obtaining a small molecule compound (SMC) seed model; derivatizing a first plurality of candidate SMCs from the SMC seed model; for each of the candidate SMCs in the first plurality of candidate SMCs, simulating first desired interactions between the candidate SMC and each of the targets; for each of the candidate SMCs in the first plurality of candidate SMCs, simulating first undesired interactions between the candidate SMC and each of the anti-targets; obtaining a first SMC interaction score for each of the candidate SMCs in the first plurality of candidate SMCs, on the basis of the first desired interactions and the first undesired interactions; and, on the basis of the first SMC interaction score, determining whether at least the minimum score for a drug is reached.SELECTED DRAWING: Figure 2

Description

Technical Field

[0001] Cross - Reference to Related Applications This application claims priority under 35 U.S.C. § 119(e) to U.S. Provisional Patent Application No. 62 / 590,141, filed on Nov. 22, 2017, having at least one of the same inventors as this application and entitled "METHOD AND SYSTEM FOR DIFFERENTIAL DRUG DISCOVERY". U.S. Provisional Patent Application No. 62 / 590,141 is hereby incorporated by reference herein.

Background Art

[0002] Many diseases have complex biological pathologies that involve multiple pathways. Small - molecule drugs have conventionally been designed against a single protein target but have off - target protein interactions (i.e., polypharmacology) other than the multiple targets. An average small - molecule compound (SMC) is thought to bind to 30 - 300 different proteins in vivo. Usually, drug molecules are intended to bind to only one of them, i.e., the target. The others are unintended off - target interactions that can be beneficial or detrimental and can occur with homologous or non - homologous proteins. In the field of multi - target drug design, one small - molecule drug that targets multiple proteins has the potential to improve the treatment of complex pathologies. However, due to various complexities, it has been difficult to design small - molecule drugs that can bind to multiple targets simultaneously.

Summary of the Invention

[0003] Generally, one or more embodiments are directed to a method of specific drug discovery, comprising obtaining a receptor panel that specifies a plurality of targets and a plurality of anti-targets, obtaining a small molecule compound (SMC) seed model, derivatizing a first plurality of candidate SMCs from the SMC seed model, for each candidate SMC in the first plurality of candidate SMCs, simulating a first desired interaction between the candidate SMC and each of the plurality of targets, for each candidate SMC in the first plurality of candidate SMCs, simulating a first undesired interaction between the candidate SMC and each of the plurality of anti-targets, obtaining a first SMC interaction score for each candidate SMC in the first plurality of candidate SMCs based on the first desired interaction and the first undesired interaction, and determining whether at least a minimum score for a drug has been reached based on the first SMC interaction score.

[0004] Generally, one or more embodiments are directed to a system for specific drug discovery, comprising a derivatization engine configured to derivatize a plurality of candidate SMCs from a small molecule compound (SMC) seed model, a molecular docking simulation engine configured to simulate a desired interaction between a candidate SMC and each of a plurality of targets specified by a receptor panel and to simulate an undesired interaction between the candidate SMC and each of a plurality of anti-targets specified by the receptor panel for each candidate SMC in the plurality of candidate SMCs, and a scoring engine configured to obtain an SMC interaction score for each candidate SMC in the plurality of candidate SMCs based on the desired interaction and the undesired interaction and to determine whether at least a minimum score for a drug has been reached based on a first SMC interaction score.

[0005] In general, one or more embodiments are non-transitory computer-readable media comprising computer-readable program code for specific drug discovery, the computer-readable program code causing a computer system to obtain a receptor panel that specifies a plurality of targets and a plurality of anti-targets, obtain a small molecule compound (SMC) seed model, derivatize a first plurality of candidate SMCs from the SMC seed model, for each candidate SMC in the first plurality of candidate SMCs, simulate a first desired interaction between the candidate SMC and each of the plurality of targets, for each candidate SMC in the first plurality of candidate SMCs, simulate a first undesired interaction between the candidate SMC and each of the plurality of anti-targets, based on the first desired interaction and the first undesired interaction, obtain a first SMC interaction score for each candidate SMC in the first plurality of candidate SMCs, and based on the first SMC interaction score, determine whether at least a minimum score for the drug has been reached.

[0006] Other aspects of the embodiments will become apparent from the following description and the appended claims.

[0007] This embodiment is shown by way of example and is not intended to be limited by the figures of the accompanying drawings.

Brief Description of the Drawings

[0008]

Figure 1

Figure 2

Figure 3A

Figure 3B

Figure 3C

Figure 4

Figure 5A

Figure 5B

Best Mode for Carrying Out the Invention

[0009] For the specific embodiments disclosed herein, reference will be made to the accompanying drawings and described in detail below. Similar elements in the various figures may be denoted by similar reference numerals and / or similar names for the sake of consistency.

[0010] The following detailed description is merely exemplary in nature and is not intended to limit the embodiments disclosed herein or the use and applications of the embodiments disclosed herein. Furthermore, it is not intended to be bound by any theory, whether explicit or implicit, presented in the above technical field, background art, brief summary, or the following detailed description.

[0011] In the following detailed description of some embodiments disclosed herein, many specific details are set forth in order to provide a more thorough understanding of the various embodiments disclosed herein. However, it will be apparent to those skilled in the art that the embodiments may be practiced without these specific details. In other instances, well-known functions have not been described in detail to avoid unnecessarily complicating the description.

[0012] Throughout this application, ordinal numbers (e.g., first, second, third, etc.) may be used as adjectives of elements (i.e., any noun in this application). The use of ordinal numbers does not imply or create any particular order of elements or limit any element to being only a single element, unless explicitly disclosed by the use of terms such as "before," "after," "single," etc. Rather, the use of ordinal numbers is to distinguish elements. As an example, a first element is different from a second element, the first element includes a plurality of elements, and may follow (or precede) the second element in the ordering of elements.

[0013] Since many diseases are known to involve multiple proteins, compound pharmacological drugs that interact with multiple targets (e.g., proteins) may be particularly valuable. Therefore, drugs that can target multiple proteins associated with a disease may be more effective than drugs specific to only one protein.

[0014] However, finding small molecule compounds (SMCs) that interact with multiple targets can be difficult for various reasons. Specifically, for example, the pockets (or more generally, interaction sites) of target proteins can have different geometric shapes and / or physicochemical compositions. Additionally, the structure of SMCs tends to be dynamic, and for example, many structures of the same SMC may exist. Therefore, predicting the compatibility of SMC targets can be a non - trivial task. Specifically, similar pockets are likely to bind a common SMC, but pocket similarity is not strictly necessary for compatibility. Instead, due to conformational changes, an SMC may be able to connect to quite different pockets.

[0015] In one or more embodiments, a multi-target drug design (MTDD) strategy is used to identify small molecule compounds (SMCs) having desired characteristics. The MTDD strategy, discussed later, can identify pharmaceutical therapies that target multiple targets simultaneously. Thus, the MTDD strategy can result in the identification of combination drugs that affect a disease network rather than a single target.

[0016] More specifically, the methods and systems according to one or more embodiments utilize combinatorics to reduce the risk of target incompatibility. That is, according to one or more embodiments, the SMCs are not designed for any one particular target. Instead, a receptor panel that may include some of both targets and anti-targets is used. In this way, the MTDD strategy of one or more embodiments increases the likelihood of finding SMCs that can interact with at least one or some of the targets on the receptor panel while avoiding interaction with anti-targets. Thus, a single drug having multiple affinities or a combination of drugs used as a combination therapy may be developed using the MTDD strategy according to one or more embodiments.

[0017] The MTDD strategy may be based on a novel design of SMCs using a computational approach in which candidate drugs are designed in silico from building blocks such as molecular fragments, atoms, etc.

[0018] Referring to FIG. 1, a system (100) for specific drug discovery is shown according to one or more embodiments. The system (100) includes a differential drug discovery engine (150). Inputs to the differential drug discovery engine (150) include a receptor panel (102) and a small molecule compound (SMC) seed model (104). Outputs generated by the differential drug discovery engine (150) include one or more discovered SMCs (190). More specifically, the outputs include one or more models (formulas, descriptions) of the discovered SMCs. Each of these components will be described later.

[0019] According to one or more embodiments, the receptor panel (112) may specify a target (104) and an anti-target (108). The target (104) may be any cellular component within a cell of any species that is regulated by the SMC. The target may be a macromolecule, such as a protein. When treating a medical condition, modulation of the target by a drug (SMC) may result in a beneficial outcome. In contrast, the anti-target (108) may be a macromolecule with which the drug being developed is not thought to interact. Modulation of the anti-target may not result in an undesirable or adverse pharmacology, such as a known effect or toxicity. Thus, by specifying a target (104) with which the SMC is assumed to interact and an anti-target (108) with which the SMC is not assumed to interact, the receptor panel (112) establishes the suitability objective for developing the SMC based on these desired and undesired interactions. The receptor panel (112) may be established based on a specific objective, such as treating or curing a disease, or more generally, based on the objective of affecting an organism in a desired manner. Based on this objective, the receptor panel (102) may be constructed based on the target (104) and the anti-target (108). The construction of the receptor panel is discussed in the flowchart of FIG. 2 and examples are provided in FIGS. 3A, 3B, and 3C.

[0020] The target may have a plurality of interaction sites that enable interaction with a drug target. Each of these interaction sites may be represented in the receptor panel (102) by one or more models of the interaction site (106). Many models of the interaction site, for example, hundreds or thousands of models of the interaction site, may be included in the receptor panel (102). Those skilled in the art will understand that the interaction site may be any kind of structure or region of a protein that enables interaction with a protein (e.g., a binding site such as a pocket). Furthermore, the interaction between the SMC and the protein is not limited to a ligand pocket binding interaction between the SMC and the protein. Instead, any kind of interaction between the SMC and the protein is within the scope of the present invention. The plurality of models of the interaction site may be within the receptor panel and correspond to a plurality of conformations. As a result, the receptor panel (102) may be based on any number of targets (104), each of which may have any number of interaction sites that may be included as models of the interaction site (106) in the receptor panel (102). Furthermore, even a single interaction site may be modeled using a plurality of models to represent different conformational configurations. Some or all of the known models may be included.

[0021] According to one or more embodiments, the presence of a plurality of targets (104) provided as models of the interaction site (106) is assumed to increase the likelihood of discovering SMCs that interact with at least a portion of the interaction site. For example, identifying an SMC that interacts with 3 out of 10 targets is not as difficult as identifying an SMC that interacts with 3 out of 3 targets. Thus, embodiments of the present invention benefit from combinatorics. The combinatorial approach is particularly beneficial in light of the known difficulties associated with systematically predicting the likelihood of interaction based on known interactions between SMCs and proteins. In the case of ligand pocket binding, it is the geometric properties of the pocket substance (such as the volume of the pocket, surface area, size of the mouth, etc.). However, due to the large number of possible molecular conformations, the fact that a first ligand binds well to a pocket does not necessarily imply that a second ligand of a quite different shape will not bind to the same pocket. Similarly, a first ligand may bind well to a pocket, but a second ligand that differs only slightly from the first ligand may not bind well to the same pocket. In light of this potentially insufficient predictability, the availability of a large number of targets that may interact with SMCs increases the likelihood of discovering SMCs with acceptable performance characteristics.

[0022] Continuing with the consideration of the receptor panel (102), the anti-targets (108) may be specified in the same manner. However, the model of the interaction site (110) of the anti-targets is based on proteins previously identified as not being targeted by the SMCs to be developed.

[0023] In one or more embodiments, priority weights may be assigned to the targets (104) and anti-targets (108). These weights indicate the importance of interaction with each target (104) and the importance of avoiding interaction with each anti-target (108).

[0024] Furthermore, a detailed description of how the receptor panel can be established is provided in step 200 of FIG. 2, and examples are provided in FIGS. 3A - 3C.

[0025] The SMC seed model (112) is, according to one or more embodiments, a candidate model of SMC that executes the method of FIG. 2 based on the receptor panel (102) to ultimately discover one or more SM Cs having desired characteristics. As discussed in FIG. 2, the SMC seed model (112) may first be tested for desired characteristics and then modified stepwise until the desired characteristics are discovered. One or more seed models (112) may be provided, and the provided seed models may be based on one or more criteria. The provided seed models may be selected based on prior knowledge. For example, the seed models may be selected based on the knowledge that they interact with one or more of the target interaction sites specified in the receptor panel. The seed models may be further derivatized from a preselected scaffold. Also, the seed models may be selected to avoid interaction with an anti-target. The seed models may represent known SM Cs such as existing drugs or may be represented by SMILES strings. For example, when used as an SMC seed model, the drug aspirin may be represented by the SMILES string "CC(=O)OC1=CC=CC=C1C(=O)O".

[0026] Continuing the discussion of the system (100), the differential drug discovery engine (150) accepts, according to one or more embodiments, the receptor panel (102) and the SMC seed model (112) as inputs and ultimately provides the discovered SMC (190) as an output. The method performed by the differential drug discovery engine (150) according to one or more embodiments aims to obtain the discovered SMC (190) that interacts with multiple targets (106) while avoiding interaction with an anti-target by leveraging the combinations resulting from the multiple targets (104) of the receptor panel (102). The differential drug discovery engine (150) may include a derivatization engine (152), a molecular docking simulation engine (154), and a scoring engine (156).

[0027] The derivatization engine (152) includes a set of machine-readable instructions configured to derive candidate SMCs from a seed model in a first iteration or from previously analyzed candidate SMCs in subsequent iterations. The derivatization of candidate SMCs is described below with respect to steps 204 and 206 of FIG. 2.

[0028] The molecular docking simulation engine (154) includes a set of machine-readable instructions configured to simulate the interaction between SMC candidates and the targets and anti-targets specified in the receptor panel. The simulation is described below with respect to step 208 of FIG. 2.

[0029] The scoring engine (156) includes a set of machine-readable instructions configured to score the simulated interactions of step 208 to obtain a single score for each candidate SMC. The simulation is described below with respect to step 210 of FIG. 2.

[0030] Collectively, the derivatization engine (152), the molecular docking simulation engine (154), and the scoring engine (156) iteratively generate candidate SMCs that may ultimately qualify as the discovered SMCs (190) having the desired properties. Consideration of the iterative execution is provided below with reference to FIG. 2.

[0031] Figure 2 shows a flowchart according to one or more embodiments. In this flowchart, various steps are provided and described in order, but those skilled in the art will understand that some or all of the steps may be executed in a different order, combined, or omitted, or some or all of the steps may be executed in parallel. Further, these steps may be executed actively or passively. For example, some steps may be executed using polling or interrupt-driven according to one or more embodiments. As an example, a decision step may not require the processor to process an instruction unless an interrupt indicating the existence of a condition is received according to one or more embodiments. As another example, a determination step may be executed by performing tests such as checking a data value to test whether the value matches a tested condition according to one or more embodiments.

[0032] The flowchart of Figure 2 shows a method of de novo drug design according to one or more embodiments. The method of de novo drug design is based on a fragment growth strategy (FGS) that computationally optimizes SMC using the target interaction site of the 3D protein structure specified by the receptor panel. FGS may include at least the following steps: finding a "fitting" lead scaffold in the pocket using docking, molecular dynamics (MD) simulation, or a machine learning approach based on characterized SMC and / or characterized interaction sites; modifying, rescoring, and selecting the most suitable new SMC; and repeating to optimize the results as will be discussed later. In all iterations, changes are made to the SMC by either redesigning a part of the SMC fragment, re-derivatizing a part of the molecule, building on the molecule, or removing from the molecule.

[0033] In one or more embodiments, the methods described later optimize SMC for an entire panel of several target and anti-target receptors. In an example where the receptor panel includes 64 different targets, 36 of the 64 targets may include targets that may have a therapeutically positive value, and the remaining 28 targets may include anti-targets that need to be optimized. Using the disclosed methods, a given SMC of this example may be evaluated for all 64 different targets of the receptor panel to determine the predicted interactions and generate a polypharmacology score. The polypharmacology score may be calculated to reward predicted interactions with multiple targets and penalize interactions with anti-targets.

[0034] Referring to this flowchart, at step 200, a receptor panel is obtained. The receptor panel may be obtained in its final format as shown in the example of FIG. 3C, or the receptor panel may be constructed. The construction of the receptor panel may be performed as follows. First, as shown in the example of FIG. 3A, the selection of proteins (or other targets) including targets and anti-targets may be obtained, for example, as a list of proteins. The selection of the provided proteins may be established based on the desired therapeutic effect, for example, when treating a disease. Next, the receptor panel may be compiled by mapping the 3D structure (e.g., a list of atoms that make up the protein with its 3D positions) for each of the proteins on the list. Each of these 3D positions may be an interaction site such as a pocket. Further, for each of the mapped 3D positions, a known configuration (as a result of conformational changes) may be obtained. An example (limited to a single protein) is provided in FIG. 3B. Thereafter, clustering may be performed to reduce the total number of 3D positions. Clustering may be performed using any measure of similarity between 3D structures. A representative may be selected for each of the clusters, and the receptor panel is obtained by compiling representatives of targets and anti-targets. An example of the receptor panel is shown in FIG. 3C.

[0035] In step 202, obtain the SMC seed model. As described above, the SMC seed model may be obtained in SMILES representation.

[0036] In step 204, derivatize the candidate SMC. In the first execution cycle of the method in FIG. 2, step 204 may be skipped, that is, the steps following step 204 may operate directly on the SMC seed model. For subsequent execution cycles, as will be described later, derivatization is performed according to one or more embodiments.

[0037] Each time step 204 is executed, the SMC under consideration may be computationally modified by substituting a functional group with another chemical fragment. Thus, the new SMC may be obtained from the parent SMC, i.e., by modifying the SMC or SMC seed model obtained from the previous execution cycle. More specifically, the SMC may be modified by decomposing the SMC into fragments and by exchanging, adding, and / or removing fragments from the SMC. These operations may be managed by a set of rules to ensure that the major structural features are retained. The set of rules may be based on, for example, Retro - synthetic Combinatorial Procedure (RECAP) or Biologically Relevant Inorganic Chemical Substructures (BRICS). Those skilled in the art will recognize that the present invention is not limited to a particular set of rules and will be able to establish how the SMC is computationally modified. Any method that enables chemically meaningful modification of the SMC may be used. Further, the modification may be of any size, ranging from modification of a single atom to modification of a larger chemical substructure. In one embodiment, fragmentation is performed exhaustively. For example, consider molecule A - B - C. Exhaustive fragmentation may generate fragments A - B, B - C, A, B, and C. The obtained fragments are modified by adding one or more other fragments obtained from a fragment library to obtain new candidate SMCs. Derivatization of the candidate SMCs may be performed randomly, but certain restrictions may be imposed when modifying the SMC. For example, a minimum similarity to the parent SMC may be required, or a portion of the parent SMC may need to be maintained as is, etc.

[0038] Derivatization of step 204 may be performed on all SMCs obtained from the previous execution cycle or on a subset of the SMCs. For example, based on the SMC interaction score, only the top SMCs may be considered. The selection criteria may be changed at the initial (exploration) and later (refinement) stages of optimization.

[0039] In step 206, ineligible candidate SMCs are removed from the candidate SMCs based on screening criteria. The screening criteria include, but are not limited to, requiring candidate SMCs with a minimum similarity to known drugs, requiring SMCs that can be synthesized with merely specific efforts, requiring specific ADMET properties, and / or requiring other desirable calculated properties such as optimal lipophilicity or the absence of labile chemical groups. The selection criteria may be changed in the initial (exploratory) and later (refinement) stages of optimization.

[0040] Furthermore, in one or more embodiments, a clustering algorithm may be used to identify representative SMCs from a set of candidate SMCs using similar multi-pharmacological profiles and perform subsequent rounds of optimization.

[0041] In step 208, according to one or more embodiments, the interactions of the SMCs with the targets and anti-targets in the receptor panel are simulated. As described above, the interaction may be the docking of the SMC into the pocket, or more generally, any kind of interaction between the SMC and the interaction site. The simulation may be performed for all combinations of the SMCs with the interaction sites of the targets and anti-targets. Each of these interactions may be scored to evaluate the degree of the interaction. As will be described later, the simulation may be performed in various ways. After the completion of step 208, based on the obtained scores, the interactions of each of the SMCs with each of the interaction sites (targets and anti-targets) in the target panel may be evaluated.

[0042] In one embodiment, a molecular docking approach is used to simulate the interaction between the SMC and the target (and anti-target). The molecular docking approach may rely on Monte Carlo simulation to minimize the energy associated with the interaction between the SMC and the interaction site. An energy-based score may be obtained based on the pose of the SMC that brings about the interaction.

[0043] In one embodiment, a molecular dynamics approach is used to simulate the interaction between the SMC and the target (and anti-target). A physics engine operating on the SMC and the interaction site may determine whether a binding, or more generally, an interaction occurs. If an interaction is detected, an energy-based score may be obtained for the SMC and the configuration of the interaction site.

[0044] In one embodiment, a machine learning approach is used to simulate the interaction between the SMC and the target (and anti-target) based on the characterized SMC and the characterized target / anti-target. In the machine learning approach, a prediction algorithm such as a random forest, a convolutional neural network, or any other prediction algorithm capable of making quantitative predictions may be used. The prediction may be the binding affinity. The prediction algorithm may have been previously trained using historical data where the interaction (or lack of interaction) between the SMC and the interaction site is known. Thus, the trained prediction algorithm may predict a binding affinity indicating to what extent the SMC under consideration interacts with the interaction site under consideration. The predicted affinity may serve as a score.

[0045] After completion of step 208, a score is available for each of the interactions between the SMC under consideration and the interaction site under consideration.

[0046] In step 210, the SMC interaction score is obtained for the SMC-target / anti-target interaction through the evaluation of the score obtained in step 208. Specifically, one SMC interaction score is obtained for each of the SMCs. The SMC interaction score may indicate to what extent the SMC interacts with the target in the receptor panel while avoiding interaction with the anti-target in the receptor panel. Generally speaking, the SMC interaction score may penalize the interaction with the anti-target in the receptor panel (resulting in a decrease in the SMC interaction score) while rewarding the predicted interaction with the target in the receptor panel (resulting in an increase in the SMC interaction score). The SMC interaction score may be calculated in various ways. For example, a weighted sum of the top 3 (or 5) target interaction scores of the SMC minus (-) the top 3 (or 5) anti-target interaction scores may be used. An example of using the top 3 interaction scores is shown in FIG. 4 and will be described below. Further, in one or more embodiments, the SMC interaction score may be designed to reward combinations of receptors from the same or different biological pathways that are thought to provide synergistic therapeutic results. Different weights may also be applied to different interaction sites to emphasize / de-emphasize the contribution of these interaction sites to the SMC interaction score. For example, a weight of 1.5 may be assigned to the primary interaction site, a weight of 1.0 may be assigned to the secondary interaction site, and a weight of 0.5 may be assigned to the lower interaction site to facilitate the search for highly important interaction sites or targets. Based on the SMC interaction score, the relevant candidate SMCs may be ranked.

[0047] In step 212, the SMC interaction score is evaluated to determine whether one or more of the candidate SMCs are eligible as drugs. The SMC may be considered eligible as a drug if the relevant SMC interaction score reaches or exceeds the minimum score. Other criteria that can be calculated for the SMC, such as molecular weight, solubility, and / or other relevant properties, etc., may also be used to consider the SMC eligible as a drug.

[0048] Using step 214, it is determined whether another iteration needs to be performed or whether the execution of the method needs to be terminated. This determination may be made based on whether at least one of the SMCs is qualified as a drug. This determination may also be made based on convergence. Convergence may be evaluated based on the score obtained in step 208. Convergence may be detected when the score reaches a specific threshold, reaches a steady state (e.g., there is no significant improvement after two iterations), etc. Additionally or alternatively, cost may be a determining factor. The cost may be measured using the CPU time spent during the execution of the method in FIG. 2, and the simulation may end after a specific amount of CPU time has been spent. If another iteration is performed, the execution of the method may proceed to step 204. Or, the execution of the method may end.

[0049] The steps of FIG. 2 may be performed for many candidate SMCs. For example, 10 generations, 100 generations, 1000 generations, or 10000 generations of candidate SMCs may be processed. These candidate SMCs may be derived from a single or multiple SMC seed models.

[0050] Referring to FIGS. 3A, 3B, and 3C, examples for generating a receptor panel according to one or more embodiments are provided. FIG. 3A shows a target list (300). The target list (300) enumerates proteins. The proteins may be selected, for example, based on the desired therapeutic effect when treating a disease. In this example, each of the proteins is identified by a UniProt ID. Additionally, a protein classification is associated with each protein. The classification indicates whether the protein is intended to function as a target or an anti-target.

[0051] Figure 3B shows the compilation of a protein pocket model (310) according to one or more embodiments. In the example of Figure 3B, only the protein pocket model for the protein "Q6PL18" is shown. Although not shown in Figure 3B, when performing step 200 of Figure 2, protein pocket models for all the proteins listed in the target list (300) of Figure 3A are obtained.

[0052] Figure 3C shows an example of a receptor panel (320) according to one or more embodiments. The receptor panel (320) for each of the proteins listed in the target list (300) of Figure 3A includes a set of representative protein pocket models obtained from the compilation of the protein pocket model (310) of Figure 3B.

[0053] Figure 4 shows an example of scored interactions (400) according to one or more embodiments. Figure 4 shows the results for five SMCs (C001 - C005), 16 targets, and 8 anti-targets. The top three targets based on the SMC interaction scores obtained in step 210 are marked, and the top three anti-target interaction scores are also marked. As shown in Figure 4, according to one or more embodiments, each SMC may have a different set of preferred targets and anti-targets.

[0054] Various embodiments have one or more of the following advantages. Embodiments of the present disclosure utilize combinatorics. The use of a relatively large receptor panel, which may further include multiple models of the same protein, increases the likelihood of identifying SMCs that succeed in interacting with at least some targets. A greater number of targets in the receptor panel may have the additional advantage of enabling normalization of interaction scores, adjusting the scores of ligands that are high or low for all targets down or up, respectively, to avoid the selection of non-discriminatory ligands, i.e., those that are generally more sticky to all targets. Also, the use of anti-targets provides additional compounds for normalization and enables optimization for target interactions that may be problematic for specific conditions or SMC scaffolds.

[0055] Pairing the receptor panel with an iterative optimization strategy may enable the simultaneous exploration of chemical space and the target space of the receptor panel. Each SMC may have its own distinct set of targets. In the next generation, it is possible to improve SMC derivatives compared to the same targets or to identify new combinations of targets.

[0056] Embodiments may require only the three-dimensional (3D) structures of the target and anti-target and one or more seed structures. Because 3D structure-based molecular docking simulations are used, the flexibility of the molecules in the ligand and receptor enables the detection of compatible target pairs with different binding site shapes. Experimental target-SMC binding data is not required. The methods and systems of one or more embodiments are computational. That is, the disclosed methods may be executed entirely in silico. However, in vitro experiments may be integrated without departing from the present invention.

[0057] For therapeutic uses, a disease may be treated by modulating multiple targets using the multi-pharmacological drugs obtained using the described methods. The multi-pharmacological drugs may be more effective than a single conventional drug and may significantly reduce the risk of loss of efficacy due to single target mutations. Another important advantage of using a single multi-pharmacological drug rather than a mixture of individual drugs may be that the risk of drug-drug interactions is reduced.

[0058] Embodiments of the present disclosure may be implemented on a computing system. Any combination of mobile, desktop, server, router, switch, embedded device, or other types of hardware may be used. For example, as shown in FIG. 5A, a computing system (500) may include one or more computer processors (502), a non-persistent storage device (504) (e.g., volatile memory such as random access memory (RAM), cache memory), a persistent storage device (506) (e.g., an optical drive such as a hard disk, a compact disc (CD) drive or a digital versatile disc (DVD) drive, flash memory, etc.), a communication interface (512) (e.g., a Bluetooth interface, an infrared interface, a network interface, an optical interface, etc.), as well as a number of other elements and functions.

[0059] The computer processor (502) may be an integrated circuit for processing instructions. For example, the computer processor may be one or more cores or micro-cores of a processor. The computing system (500) may also include one or more input devices (510) such as a touch screen, a keyboard, a mouse, a microphone, a touch pad, an electronic pen, or any other type of input device.

[0060] The communication interface (512) may include an integrated circuit for connecting the computing system (500) to a network (not shown) (e.g., a wide area network (WAN) such as a local area network (LAN), the Internet, a mobile network, or any other type of network) and / or another device such as another computing device.

[0061] Furthermore, the computing system (500) may include one or more output devices (508) such as a screen (e.g., a liquid crystal display (LCD), a plasma display, a touch screen, a cathode ray tube (CRT) monitor, a projector, or other display device), a printer, an external storage device, or any other output device. The one or more output devices may be the same as or different from the input devices. The input and output devices may be connected locally or remotely to the computer processor (502), the non-persistent storage device (504), and the persistent storage device (506). There are many different types of computing systems, and the aforementioned input and output devices may take other forms.

[0062] Software instructions in the form of computer-readable program code for executing embodiments of the present disclosure may be stored, in whole or in part, temporarily or persistently, on a non-transitory computer-readable medium such as a CD, DVD, storage device, diskette, tape, flash memory, physical memory, or other computer-readable storage medium. Specifically, the software instructions may correspond to computer-readable program code configured to execute one or more embodiments of the present disclosure when executed by a processor.

[0063] The computing system (500) of FIG. 5A may be connected to a network or to a part of a network. For example, as shown in FIG. 5B, the network (520) may include a plurality of nodes (e.g., node X (522), node Y (524)). Each node may correspond to a computing system such as the computing system shown in FIG. 5A, or a group of combined nodes may correspond to the computing system shown in FIG. 5A. As an example, embodiments of the present disclosure may be implemented on nodes of a distributed system connected to other nodes. As another example, embodiments of the present disclosure may be implemented on a distributed computing system having a plurality of nodes, and each part of the present disclosure may be arranged on different nodes within the distributed computing system. Further, one or more elements of the aforementioned computing system (500) may be arranged at remote locations and connected to other elements via a network.

[0064] Although not shown in FIG. 5B, a node may correspond to a blade within a server chassis connected to other nodes via a backplane. As another example, a node may correspond to a server within a data center. As another example, a node may correspond to a computer processor or a microcore of a computer processor having shared memory and / or resources.

[0065] Nodes (e.g., node X (522), node Y (524)) within the network (520) may be configured to provide services to a client device (526). For example, a node may be part of a cloud computing system. A node may include a function of receiving a request from the client device (526) and transmitting a response to the client device (526). The client device (526) may be a computing system such as the computing system shown in FIG. 5A. Further, the client device (526) may include and / or execute all or a part of one or more embodiments of the present disclosure.

[0066] The computing system or group of computing systems described in FIGS. 5A and 5B may include the functionality to perform various operations disclosed herein. For example, the computing system may perform communication between processes on the same or different systems. Various mechanisms using some form of active or passive communication may facilitate data exchange between processes on the same device. Examples representing such inter-process communication include, but are not limited to, the implementation of files, signals, sockets, message queues, pipelines, semaphores, shared memory, message passing, and memory mapped files. Further details regarding some of these non-limiting examples are provided below.

[0067] Based on the client-server networking model, a socket can function as an interface or a communication channel endpoint to enable two-way data transfer between processes on the same device. First and foremost, according to the client-server networking model, a server process (e.g., a process that provides data) may create a first socket object. Next, the server process binds to the first socket object, thereby associating the first socket object with a unique name and / or address. After creating and binding the first socket object, the server process waits and listens for incoming connection requests from one or more client processes (such as a process that seeks data). At this point, if a client process attempts to obtain data from the server process, the client process starts by creating a second socket object. Next, the client process proceeds to generate a connection request that includes at least the second socket object as well as the unique name and / or address associated with the first socket object. Then, the client process sends the connection request to the server process. Depending on availability, the server process may accept the connection request and establish a communication channel with the client process, or the server process may queue the connection request in a buffer until the server process is ready, being busy with the processing of other operations. The established connection notifies the client process that communication may be initiated. In response, the client process may generate a data request that specifies the data the client process attempts to obtain. Thereafter, the data request is sent to the server process. Upon receiving the data request, the server process analyzes the request and collects the requested data. Finally, the server process then generates a response that includes at least the requested data and sends that response to the client process. The data may be transferred more generally as a datagram or a stream of characters (e.g., bytes).

[0068] Shared memory refers to the allocation of virtual memory space to demonstrate a mechanism by which data may be communicated and / or accessed by multiple processes. In the implementation of shared memory, the initialization process first creates a sharable segment in a persistent or non-persistent storage device. After creation, the initialization process mounts the sharable segment and then maps the sharable segment to the address space associated with the initialization process. Following the mount, the initialization process proceeds to identify one or more permitted processes that may write to or read from the sharable segment and grant access permissions. Changes made to data within the sharable segment by one process may immediately affect other processes that are also linked to the sharable segment. Further, when one of the permitted processes accesses the sharable segment, the sharable segment is mapped to the address space of that permitted process. In many cases, only one permitted process, other than the initialization process, may mount the sharable segment at any given time.

[0069] Other techniques may be used to share data, such as the various data described in this application, between processes without departing from the scope of the present disclosure. The processes may be part of the same or different applications and may be executed on the same or different computing systems.

[0070] Rather than, or in addition to, sharing data between processes, a computing system that executes one or more embodiments of the present disclosure may include the function of receiving data from a user. For example, in one or more embodiments, a user may submit data via a graphical user interface (GUI) on a user device. The user may transmit data via the graphical user interface by selecting one or more graphical user interface widgets, or by using a touchpad, keyboard, mouse, or other input device to insert text and other data into the graphical user interface widgets. In response to the selection of a particular item, information about the particular item may be retrieved by a computer processor from a persistent or non-persistent storage device. When an item is selected by the user, the content of the data retrieved regarding the particular item may be displayed on the user device in response to the user's selection.

[0071] As another example, a request to obtain data regarding a particular item may be sent via a network to a server operably connected to the user device. For example, a user may select a uniform resource locator (URL) link within a web client of the user device, thereby initiating a hypertext transfer protocol (HTTP) or other protocol request that is sent to a network host associated with the URL. In response to the request, the server may extract data regarding the particular selected item and send the data to the device that initiated the request. When the user device receives data regarding a particular item, the content of the received data regarding the particular item may be displayed on the user device in response to the user's selection. In addition to the above example, the data received from the server after selecting the URL link may provide a web page in hypertext markup language (HTML) that is rendered by the web client and displayed on the user device.

[0072] By using the above techniques or when data is retrieved from a memory device, the computing system may extract one or more data items from the retrieved data when executing one or more embodiments of the present disclosure. For example, the extraction may be performed as follows by the computing system of FIG. 5A. First, a data compilation pattern (e.g., grammar, schema, layout) is determined, which may be based on one or more of position (e.g., bit or column position, Nth token in a data stream, etc.), attribute (if the attribute is associated with one or more values), or hierarchical / tree structure (composed of layers of nodes at various levels of detail such as nested packet headers or nested document sections). Next, a stream of raw, unprocessed data symbols is parsed into a stream (or hierarchical structure) of tokens (each token may have an associated "type" of token) in the context of the compilation pattern.

[0073] Next, one or more data items are extracted from the token stream or structure using extraction criteria, which are processed according to the compilation pattern to extract one or more tokens (or nodes from a hierarchical structure). For position-based data, the tokens at the positions identified by the extraction criteria are extracted. For attribute / value-based data, the tokens and / or nodes associated with the attributes that meet the extraction criteria are extracted. For hierarchical / hierarchical data, the tokens associated with the nodes that match the extraction criteria are extracted. The extraction criteria may be as simple as an identifier string or may be a query provided to a structured data repository (the data repository may be compiled according to a database schema or data format such as XML).

[0074] The extracted data may be used for further processing by a computing system. For example, the computing system of FIG. 5A may perform a data comparison while executing one or more embodiments of the present disclosure. The data comparison may be used to compare two or more data values (e.g., A, B). For example, one or more embodiments may determine whether A>B, A=B, A!=B, A<B, etc. This comparison may be performed by submitting an operation code that specifies an operation related to the comparison to A, B, and an arithmetic logic unit (ALU) (i.e., a circuit that performs arithmetic operations and / or bitwise logical operations on two data values). The ALU outputs a numerical result of the operation and / or one or more status flags related to the numerical result. For example, the status flag may indicate whether the numerical result is a positive number, a negative number, zero, etc. The comparison may be performed by selecting an appropriate operation code and then reading the numerical result and / or the status flag. For example, to determine whether A>B, B may be subtracted from A (i.e., A - B), and the status flag may be read to determine whether the result is positive (i.e., if A>B, then A - B>0). In one or more embodiments, B may be regarded as a threshold value, and A may be regarded as satisfying the threshold value if A = B or A>B as determined using the ALU. In one or more embodiments of the present disclosure, A and B may be vectors, and comparing A with B may require comparing the first element of vector A with the first element of vector B, comparing the second element of vector A with the second element of vector B, etc. In one or more embodiments, if A and B are character strings, the binary values of the character strings may be compared.

[0075] The computing system of FIG. 5A may implement and / or connect to a data repository. For example, one type of data repository is a database. A database is a collection of information configured to facilitate data retrieval, modification, rearrangement, and deletion. A database management system (DBMS) is a software application that provides an interface for a user to define, create, query, update, or manage a database.

[0076] A user or software application may submit a statement or query to the DBMS. Next, the DBMS interprets the string. The string is a select statement, update statement, create statement, delete statement, etc. that requests information. Further, the statement may include parameters that specify data, or data containers (such as databases, tables, records, columns, views, etc.), identifiers, conditions (comparison operators), functions (such as join, full join, count, average, etc.), sorting (such as ascending, descending), etc. The DBMS may execute the statement. For example, the DBMS may access a memory buffer, reference, or index file for reading, writing, deleting, or any combination thereof in response to the statement. The DBMS may load data from persistent or non-persistent storage and perform calculations to respond to the query. The DBMS may return the result to the user or software application.

[0077] The computing system of FIG. 5A may include a function of providing raw and / or processed data such as the results of comparison and other processing. For example, providing data may be achieved through various presentation methods. Specifically, the data may be provided via a user interface provided by a computing device. The user interface may include a GUI that displays information on a display device such as a computer monitor or a touch screen on a handheld computer device. The GUI may include various GUI widgets that organize what data is shown and how the data is provided to the user. Further, the GUI may directly provide the user with data, for example, data provided as an actual data value through text, or data rendered into a visual representation of the data through visualization of a data model by a computing device.

[0078] For example, the GUI may first obtain a notification from a software application that requests that a specific data object be provided within the GUI. Next, the GUI may determine the data object type associated with the specific data object, for example, by obtaining data from data attributes within the data object that identify the data object type. Next, the GUI may determine any rules specified for displaying that data object type, for example, rules specified by a software framework for a data object class or rules specified according to local parameters defined by the GUI for presenting that data object type. Finally, the GUI may obtain a data value from the specific data object and render a visual representation of the data value within the display device according to the rules specified for that data object type.

[0079] The data may also be provided by various audio methods. In particular, the data may be rendered in an audio format and provided as sound via one or more speakers operably connected to a computing device.

[0080] The data may also be provided to the user by a haptic method. For example, the haptic method may include vibrations or other physical signals generated by a computing system. For example, the data may be provided to the user using vibrations generated by a handheld computer device for a predetermined duration and strength of vibration to communicate the data.

[0081] In the above description of the functions, only a few examples of the functions executed by the computing system of FIG. 5A and the nodes and / or client devices of FIG. 5B are presented. Other functions may be executed using one or more embodiments of the present disclosure.

[0082] Although the present disclosure has been described with respect to a limited number of embodiments, those skilled in the art having the benefit of this disclosure will understand that other embodiments may be devised that do not depart from the scope of the disclosure disclosed herein. Accordingly, the scope of the present disclosure should be limited only by the appended claims.

[0083] The embodiments and examples described herein are presented to best explain the invention and its particular use, thereby enabling those skilled in the art to make and use the invention. However, those skilled in the art will recognize that the foregoing description and examples are presented for purposes of illustration and example only. The description given is not intended to be exhaustive or to limit the invention to the precise form disclosed.

[0084] Although the present invention has been described with respect to a limited number of embodiments, those of ordinary skill in the art having the benefit of this disclosure will appreciate that other embodiments can be devised that do not depart from the scope of the invention disclosed herein. Accordingly, the scope of the invention should be limited only by the appended claims.

Claims

1. A method for specific drug discovery, comprising: generating a digital protein receptor panel associated with a first protein target, a second protein target, a first protein anti-target, and a pharmaceutical therapy of the second protein anti-target; generating a first set of interaction sites for the first protein target, the first set including a first set of three-dimensional (3D) protein structural models corresponding to molecular conformations; generating a second set of interaction sites for the second protein target, the second set including a second set of 3D protein structural models corresponding to molecular conformations; generating a third set of interaction sites for the first protein anti-target, the third set including a third set of 3D protein structural models corresponding to molecular conformations; generating a fourth set of interaction sites for the second protein anti-target, the fourth set including a fourth set of 3D protein structural models corresponding to molecular conformations; and generating it by deriving a plurality of candidate small molecule compounds (SMCs) from an SMC seed model; utilizing the digital protein receptor panel to generate an SMC treatment interaction score for each candidate SMC in the plurality of candidate SMCs; For a candidate SMC in the plurality of candidate SMCs, utilizing a machine learning model to predict binding affinity between the candidate SMC and the first set of 3D protein structural models from the first set of interaction sites of the digital protein receptor panel corresponding to the first protein target; and simulating a first target interaction between the candidate SMC and the first set of 3D protein structural models from the first set of interaction sites of the digital protein receptor panel corresponding to the first protein target by utilizing weights corresponding to interaction sites from the first set of interaction sites and additional weights corresponding to additional interaction sites from the first set of interaction sites to generate a first target interaction score; For a candidate SMC in the plurality of candidate SMCs, simulating a second target interaction between the candidate SMC and the second set of 3D protein structural models from the second set of interaction sites of the digital protein receptor panel corresponding to the second protein target to generate a second target interaction score; For a candidate SMC in the plurality of candidate SMCs, simulating a first anti-target interaction between the candidate SMC and the third set of 3D protein structural models from the third set of interaction sites of the digital protein receptor panel corresponding to the first protein anti-target to generate a first anti-target interaction score; For a candidate SMC in the plurality of candidate SMCs, simulating a second anti-target interaction between the candidate SMC and the fourth set of 3D protein structural models from the fourth set of interaction sites of the digital protein receptor panel corresponding to the second protein anti-target to generate a second anti-target interaction score; generating an SMC therapeutic interaction score by combining the first target interaction score, the second target interaction score, the first anti-target interaction score, and the second anti-target interaction score, wherein the SMC therapeutic interaction score comprises a single score indicative of the function of the candidate SMC interacting with the first protein target, the second protein target, the first protein anti-target, and the second protein anti-target corresponding to the pharmaceutical therapy; and generating it by determining the polypharmacological drug by selecting SMCs from the candidate SMCs based on specific SMC therapeutic interaction scores corresponding to the SMCs that satisfy an SMC interaction score threshold; synthesizing the determined polypharmacological drug; A method comprising:

2. deriving a plurality of additional candidate SMCs from said plurality of candidate SMCs; utilizing the digital protein receptor panel to generate an additional SMC treatment interaction score for each additional candidate SMC in the plurality of additional candidate SMCs; For an additional candidate SMC in the plurality of additional candidate SMCs, simulating a first additional target interaction between the additional candidate SMC and the first set of 3D protein structural models from the first set of interaction sites of the digital protein receptor panel corresponding to the first protein target to generate a first additional target interaction score; For the additional candidate SMC in the plurality of additional candidate SMCs, simulating a second additional target interaction between the additional candidate SMC and the second set of 3D protein structural models from the second set of interaction sites of the digital protein receptor panel corresponding to the second protein target to generate a second additional target interaction score; For the additional candidate SMC in the plurality of additional candidate SMCs, simulating a first additional anti-target interaction between the additional candidate SMC and the third set of 3D protein structural models from the third set of interaction sites of the digital protein receptor panel corresponding to the first protein anti-target to generate a first additional anti-target interaction score; For the additional candidate SMC in the plurality of additional candidate SMCs, simulating a second additional anti-target interaction between the additional candidate SMC and the fourth set of 3D protein structural models from the fourth set of interaction sites of the digital protein receptor panel corresponding to the second protein anti-target to generate a second additional anti-target interaction score; generating an additional SMC treatment interaction score by combining the first additional target interaction score, the second additional target interaction score, the first additional anti-target interaction score, and the second additional anti-target interaction score; and generating it by determining the polypharmacological drug by selecting an additional SMC from the additional candidate SMCs based on comparing the additional SMC therapeutic interaction score to the SMC interaction score threshold; The method of claim 1 further comprising:

3. 10. The method of claim 1, further comprising, prior to simulating the first target interaction, updating the plurality of candidate SMCs by removing a subset of candidate SMCs based on a screening criterion.

4. 4. The method of claim 3, wherein the screening criteria include at least one selected from the group consisting of ADMET properties, synthetic feasibility, and similarity to known drugs.

5. The method of claim 1 , wherein deriving the plurality of candidate SMCs from the SMC seed model comprises generating a plurality of molecular combinations from the SMC seed model.

6. generating the SMC treatment interaction score, increasing the SMC therapeutic interaction score utilizing the first target interaction score and the second target interaction score; reducing the SMC therapeutic interaction score utilizing the first anti-target interaction score and the second anti-target interaction score; The method of claim 1 , comprising:

7. 2. The method of claim 1, further comprising simulating a target interaction between the candidate SMC and the first protein target by utilizing a machine learning model having a feature representation of the candidate SMC and a plurality of feature representations for each of the first protein targets to generate binding affinity predictions between the candidate SMC and the first set of 3D protein structural models from the first set of interaction sites of the digital protein receptor panel corresponding to the first protein target.

8. A method for specific drug discovery, comprising: generating a digital protein receptor panel associated with a first protein target, a second protein target, a first protein anti-target, and a pharmaceutical therapy of the second protein anti-target; generating a first set of interaction sites for the first protein target, the first set including a first set of three-dimensional (3D) protein structural models corresponding to molecular conformations; generating a second set of interaction sites for the second protein target, the second set including a second set of 3D protein structural models corresponding to molecular conformations; generating a third set of interaction sites for the first protein anti-target, the third set including a third set of 3D protein structural models corresponding to molecular conformations; generating a fourth set of interaction sites for the second protein anti-target, the fourth set including a fourth set of 3D protein structural models corresponding to molecular conformations; and generating it by utilizing the digital protein receptor panel to generate an SMC treatment interaction score for a candidate SMC; For the candidate SMC, utilizing a machine learning model to predict binding affinity between the candidate SMC and the first set of 3D protein structural models from the first set of interaction sites of the digital protein receptor panel corresponding to the first protein target; and simulating a first target interaction between the candidate SMC and the first set of 3D protein structural models from the first set of interaction sites of the digital protein receptor panel corresponding to the first protein target by utilizing weights corresponding to interaction sites from the first set of interaction sites and additional weights corresponding to additional interaction sites from the first set of interaction sites to generate a first target interaction score; For the candidate SMC, simulating second target interactions between the candidate SMC and the second set of 3D protein structural models from the second set of interaction sites of the digital protein receptor panel corresponding to the second protein target to generate second target interaction scores; For the candidate SMC, simulating a first anti-target interaction between the candidate SMC and the third set of 3D protein structural models from the third set of interaction sites of the digital protein receptor panel corresponding to the first protein anti-target to generate a first anti-target interaction score; For the candidate SMC, simulating second anti-target interactions between the candidate SMC and the fourth set of 3D protein structural models from the fourth set of interaction sites of the digital protein receptor panel corresponding to the second protein anti-target to generate a second anti-target interaction score; generating an SMC therapeutic interaction score by combining the first target interaction score, the second target interaction score, the first anti-target interaction score, and the second anti-target interaction score, wherein the SMC therapeutic interaction score comprises a single score indicative of the function of the candidate SMC interacting with the first protein target, the second protein target, the first protein anti-target, and the second protein anti-target corresponding to the pharmaceutical therapy; and generating it by synthesizing a polypharmacological drug from the candidate SMCs based on the SMC therapeutic interaction scores corresponding to the SMCs that satisfy an SMC interaction score threshold; A method comprising:

9. A method for specific drug discovery, comprising: generating a digital protein receptor panel associated with a first protein target, a second protein target, a first protein anti-target, and a pharmaceutical therapy of the second protein anti-target; generating a first set of interaction sites for the first protein target, the first set including a first set of three-dimensional (3D) protein structural models corresponding to molecular conformations; generating a second set of interaction sites for the second protein target, the second set including a second set of 3D protein structural models corresponding to molecular conformations; generating a third set of interaction sites for the first protein anti-target, the third set including a third set of 3D protein structural models corresponding to molecular conformations; generating a fourth set of interaction sites for the second protein anti-target, the fourth set including a fourth set of 3D protein structural models corresponding to molecular conformations; and generating it by Identifying a plurality of candidate small molecule compounds (SMCs); utilizing the digital protein receptor panel to generate an SMC treatment interaction score for each candidate SMC in the plurality of candidate SMCs; For a candidate SMC in the plurality of candidate SMCs, utilizing a machine learning model to predict binding affinity between the candidate SMC and the first set of 3D protein structural models from the first set of interaction sites of the digital protein receptor panel corresponding to the first protein target; and simulating a first target interaction between the candidate SMC and the first set of 3D protein structural models from the first set of interaction sites of the digital protein receptor panel corresponding to the first protein target by utilizing weights corresponding to interaction sites from the first set of interaction sites and additional weights corresponding to additional interaction sites from the first set of interaction sites to generate a first target interaction score; For a candidate SMC in the plurality of candidate SMCs, simulating a second target interaction between the candidate SMC and the second set of 3D protein structural models from the second set of interaction sites of the digital protein receptor panel corresponding to the second protein target to generate a second target interaction score; For a candidate SMC in the plurality of candidate SMCs, simulating a first anti-target interaction between the candidate SMC and the third set of 3D protein structural models from the third set of interaction sites of the digital protein receptor panel corresponding to the first protein anti-target to generate a first anti-target interaction score; For a candidate SMC in the plurality of candidate SMCs, simulating a second anti-target interaction between the candidate SMC and the fourth set of 3D protein structural models from the fourth set of interaction sites of the digital protein receptor panel corresponding to the second protein anti-target to generate a second anti-target interaction score; generating an SMC therapeutic interaction score by combining the first target interaction score, the second target interaction score, the first anti-target interaction score, and the second anti-target interaction score, wherein the SMC therapeutic interaction score comprises a single score indicative of the function of the candidate SMC interacting with the first protein target, the second protein target, the first protein anti-target, and the second protein anti-target corresponding to the pharmaceutical therapy; and generating it by determining the polypharmacological drug by selecting SMCs from the candidate SMCs based on specific SMC therapeutic interaction scores corresponding to SMCs that satisfy an SMC interaction score threshold; synthesizing the determined polypharmacological drug; A method comprising:

10. deriving a plurality of additional candidate SMCs from said plurality of candidate SMCs; utilizing the digital protein receptor panel to generate an additional SMC treatment interaction score for each additional candidate SMC in the plurality of additional candidate SMCs; For an additional candidate SMC in the plurality of additional candidate SMCs, simulating additional target interactions between the additional candidate SMC and the first set of interaction sites or the second set of interaction sites of the digital protein receptor panel corresponding to the first protein target or the second protein target to generate an additional target interaction score; For the additional candidate SMC in the plurality of additional candidate SMCs, simulating additional anti-target interactions between the additional candidate SMC and the third set of interaction sites or the fourth set of interaction sites of the digital protein receptor panel corresponding to the first protein anti-target or the second protein anti-target to generate additional anti-target interaction scores; generating an additional SMC therapeutic interaction score by combining the additional target interaction score and the additional anti-target interaction score; and generating it by determining the polypharmacological drug by selecting an additional SMC from the additional candidate SMCs based on comparing the additional SMC therapeutic interaction score to the SMC interaction score threshold; 10. The method of claim 9, further comprising:

11. 10. The method of claim 1, further comprising using said polypharmacological drug to target one or more proteins in a treatment.

12. 9. The method of claim 8, wherein the first set of interaction sites comprises a plurality of 3D protein structural models for the first protein target.

13. increasing the SMC therapeutic interaction score utilizing the first target interaction score and the second target interaction score; reducing the SMC therapeutic interaction score utilizing the first anti-target interaction score and the second anti-target interaction score; generating the SMC treatment interaction score by The method of claim 8 further comprising:

14. 10. The method of claim 8, further comprising simulating the first target interaction by utilizing molecular docking between the first set of 3D protein structural models from the first set of interaction sites of the digital protein receptor panel corresponding to the first protein target.

15. The method of claim 8, further comprising generating an additional SMC treatment interaction score for each additional candidate SMC in the plurality of additional candidate SMCs based on a second additional target interaction and a second additional anti-target interaction in the digital protein receptor panel.

16. 9. The method of claim 8, further comprising deriving the candidate SMC from an SMC seed model.

17. 10. The method of claim 9, further comprising updating the plurality of candidate SMCs by removing a subset of candidate SMCs based on a screening criterion, the screening criterion being determined utilizing ADMET characteristics of the plurality of candidate SMCs.

18. increasing the SMC therapeutic interaction score utilizing the first target interaction score and the second target interaction score; reducing the SMC therapeutic interaction score utilizing the first anti-target interaction score and the second anti-target interaction score; generating the SMC treatment interaction score by 10. The method of claim 9, further comprising:

19. 10. The method of claim 9, wherein the first set of interaction sites comprises a plurality of 3D protein structural models for the first protein target, and the third set of interaction sites comprises a plurality of additional 3D protein structural models for the first protein anti-target.

20. 10. The method of claim 9, further comprising simulating a target interaction between the candidate SMC and the first protein target by utilizing the machine learning model having a feature representation of the candidate SMC and a plurality of feature representations for each of the first protein targets to generate binding affinity predictions between the candidate SMC and the first set of 3D protein structural models from the first set of interaction sites of the digital protein receptor panel that correspond to the first protein target.

Images