Modeling and optimization techniques for dialytic toxin replacement agent compounds

Abstract

Systems, methods, and / or apparatuses can be operated to perform dialysis procedures including a displacement agent infusion process. In one embodiment, a method for determining a displacement agent compound may include constructing a plurality of quantitative structure-activity relationship (QSAR) models of a target protein, each of a plurality of binding sites corresponding to one target protein QSAR model; analyzing a set of candidate compound models using the plurality of QSAR models to determine a set of at least one potential compound having an affinity for binding to each of the plurality of binding sites; and selecting at least one displacement agent compound from the set including the at least one potential compound. Other embodiments are also described.

Description

[0001] Cross-reference to related applications

[0002] This application claims priority to U.S. Provisional Patent Application Serial No. 62 / 967,738, filed January 30, 2020, pursuant to 35 U.S. SC § 119(e), the entire contents of which are incorporated herein by reference as if fully set forth herein. Technical Field

[0003] This disclosure generally relates to a dialysis system including techniques for removing target substances from a patient's blood using a replacement agent compound during dialysis treatment, and more specifically, to processes for modeling, optimizing, and / or otherwise determining a replacement agent compound, target compound, treatment regimen, and / or the like for a particular patient. Background Technology

[0004] Dialysis machines are known for treating kidney disease. The two main dialysis methods are hemodialysis (HD) and peritoneal dialysis (PD). During dialysis treatment, various unwanted substances may be removed from a patient's blood, including waste products (e.g., urea), toxins, and foreign substances (e.g., prescription drug molecules). The dialysis process may not be as effective at removing protein-bound substances as free or unbound substances, because only the unbound fraction of the substance can pass through the dialyzer membrane. Therefore, protein-bound substances often require longer dialysis times and / or specific treatment methods to be effectively removed from a patient's blood during dialysis treatment.

[0005] Various techniques have been used to attempt to remove protein-bound uremic toxins. For example, activated charcoal suspended in dialysate can help maintain the diffusion gradient of protein-bound uremic toxins and increase their removal rate. However, this approach is non-selective and logically cumbersome. Attempts have been made to address this issue by increasing certain aspects of toxin removal, such as increasing dialysis duration, dialyzer size, dialysate flow rate, dialysis frequency and duration, hemofiltration and hemodiafiltration, dialysis membrane pore size or surface area, fractionated plasma separation, and increased convection. These methods may have limited potential in increasing the removal of protein-bound uremic toxins; however, they all suffer from one or more major drawbacks, such as high cost, cumbersome implementation, potential adverse side effects, unknown feasibility or clinical applicability, and inability to be implemented on a large scale in dialysis centers or other medical facilities.

[0006] It is precisely because of these and other considerations that this improvement may be useful. Summary of the Invention

[0007] The present invention is provided to introduce the selection of concepts in a simplified form, which will be further described in the detailed embodiments below. The present invention is not intended to necessarily identify key or essential features of the claimed subject matter, nor is it intended to help determine the scope of the claimed subject matter.

[0008] This disclosure generally relates to methods, apparatus, and systems for a displacement agent determination process, which operates to determine a displacement agent for performing a displacement dialysis procedure to remove protein-bound substances from a patient's blood. The displacement agent can be used during dialysis treatment to bind to proteins to displace target substances bound to proteins (i.e., toxic substances to be removed via dialysis treatment). In some embodiments, the displacement agent determination process may include determining displacement agent properties, such as displacement agent-protein binding sites, suitability for dialysis patients, and / or similar characteristics. In various embodiments, quantitative structure-activity relationship (QSAR) models can be developed to screen candidates based on displacement agent properties. QSAR models can be used to assess the binding affinity of candidates to proteins, etc., to determine the displacement agent.

[0009] In one embodiment, a method for determining a displacer compound may include constructing at least one QSAR model based on at least one displacer property for analyzing candidate compounds for binding to a target protein.

[0010] In one embodiment, a method for determining a displacement agent compound for binding a target protein having a plurality of binding sites may include constructing a plurality of target protein QSAR models, each of the plurality of binding sites corresponding to one target protein QSAR model, determining a set of result compounds having an affinity for binding to each of the plurality of binding sites, and selecting at least one displacement agent compound from the set of result compounds.

[0011] In one embodiment, a method for removing a target substance from a patient's blood during a dialysis procedure may include infusing the patient with a displacing agent configured to displace the target substance bound to a target protein to increase the amount of free target substance in the patient's blood. The displacing agent is determined via a displacing agent determination process, which may include selecting the displacing agent using at least one QSAR model configured to model multiple binding sites of the target protein.

[0012] In one embodiment, an apparatus may include a storage device for storing instructions and a logic unit coupled to the storage device, the logic unit responding to the execution instructions to construct a plurality of target protein QSAR models, each of the plurality of binding sites corresponding to a target protein QSAR model, determine a set of result compounds having an affinity for binding to each of the plurality of binding sites, and select at least one displacement agent compound from the set of result compounds.

[0013] In one embodiment, an apparatus may include a storage device for storing instructions and a logic unit coupled to the storage device, the logic unit responding to the execution instructions to infuse a displacer configured to displace a target substance bound to a target protein to increase the amount of free target substance in the patient's blood, the displacer being determined via a displacer determination process that may include selecting the displacer using at least one QSAR model configured to model multiple binding sites of the target protein.

[0014] In various embodiments, the at least one QSAR model can be configured to predict the binding affinity of the displacer to the target protein. In some embodiments, the at least one QSAR model includes at least one albumin-binding model and at least one plasma protein-binding (PPB) model.

[0015] In some embodiments, the target protein may include multiple binding sites. In various embodiments, the at least one QSAR model may include one model for each of the multiple binding sites.

[0016] In some embodiments, the target protein may include albumin. In various embodiments, the target protein may include albumin, and the plurality of binding sites may include binding site I and binding site II. In some embodiments, the at least one QSAR model may include a binding site I model for predicting the binding affinity of a candidate to binding site I of albumin and a binding site II model for predicting the binding affinity of a candidate to binding site II of albumin. In some embodiments, the displacing agent compound determined via the binding site I model may have a binding affinity of log K (log[%PPB / (101-%PPB)]). In an exemplary embodiment, the displacing agent compound determined via the binding site II model may have a binding affinity of log K.

[0017] The replacement agent determination process according to some embodiments, and the dialysis procedure using a replacement agent selected via the replacement agent determination process, can have several technical advantages over conventional techniques. For example, non-limiting technical advantages may include determining the replacement agent compound with improved accuracy, which may be more effective for the patient. In another example, non-limiting technical advantages may include improving the patient's quality of life and treatment experience by providing more treatment options and reducing disease and treatment complications. Furthermore, the replacement agent determination process can be integrated into multiple practical applications, such as providing dialysis treatment recommendations using the selected replacement agent and performing dialysis using the selected replacement agent.

[0018] Furthermore, the introduction of a replacement agent may lead to undesirable side effects (e.g., with the use of ibuprofen, tryptophan, and / or other known replacement agents). For example, the formation of a free target substance in the blood may cause the amount of the substance to increase to unsafe levels. In another example, it may be necessary to control the amount of replacement agent in a patient's blood, for example, to prevent it from reaching unsafe or unhealthy levels that would otherwise affect the patient in ways other than binding to the target protein. However, according to some embodiments, selecting a replacement agent based on a replacement agent determination process can allow researchers and / or healthcare professionals to determine, for example, the optimal replacement agent that is safer and / or controllable to achieve better patient outcomes (e.g., requiring as many doses as possible, and / or the like). Furthermore, by determining one or more replacement agents according to some embodiments, patient healthcare can select replacement agents that are more effective, less costly, more readily available, have fewer side effects, are non-toxic or less toxic, are personalized for the patient, and / or otherwise have other beneficial qualities compared to conventional compounds.

[0019] In view of this disclosure, those skilled in the art will recognize the additional technical advantages and practical applications. Attached Figure Description

[0020] As an example, a specific embodiment of the disclosed machine will now be described with reference to the accompanying drawings, in which:

[0021] Figure 1 An example of a first operating environment that may represent some embodiments of the present disclosure is shown;

[0022] Figure 2 An example of a second operating environment that may represent some embodiments of this disclosure is shown;

[0023] Figure 3 Exemplary graphs are shown of dialysate removal of target substances during various dialysis procedures according to this disclosure;

[0024] Figure 4 Exemplary diagrams of infusion configurations according to some embodiments of this disclosure are shown;

[0025] Figure 5 The first logical flow according to this disclosure is shown;

[0026] Figure 6 The first logical flow according to this disclosure is shown;

[0027] Figure 7 An exemplary hemodialysis system is shown; and

[0028] Figure 8 An embodiment of a computing architecture according to this disclosure is shown. Detailed Implementation

[0029] This embodiment will now be described more fully below with reference to the accompanying drawings, in which several exemplary embodiments are illustrated. However, the subject matter of this disclosure may be implemented in many different forms and should not be construed as limited to the embodiments set forth herein. Rather, these embodiments are provided so that this disclosure will be thorough and complete, and are intended to convey the scope of the subject matter to those skilled in the art. In the drawings, the same numerals consistently refer to the same or similar elements.

[0030] Various embodiments are generally directed to systems, methods, and / or apparatuses for identifying a displacement agent compound that can be used to remove protein-bound target compounds from a patient's blood during an exchange dialysis procedure. Non-limiting examples of target compounds may include protein-bound uremic toxins (PBUTs). The exchange dialysis procedure may introduce a displacement agent during dialysis treatment, which operates to displace the target substance from a protein in its protein-bound form to generate a free substance. Typically, the target substance is an unwanted substance removed via dialysis, which can be released by applying the displacement agent. A non-limiting example of an exchange dialysis procedure may include a process identical or similar to that described in U.S. Patent No. 8,206,591 entitled "Method of Removing Protein-Bound Deleterious Substances During Extracorporeal Renal Replacement Treatment," which is incorporated herein by reference as if fully set forth in this disclosure.

[0031] The generation of free target substances in the blood can cause the amount of the substance to increase to unsafe levels. Furthermore, it may be necessary to control the amount of replacement agent in the patient's blood, for example, to prevent it from reaching unsafe levels that would otherwise affect the patient in ways beyond binding to the target protein. Therefore, a replacement dialysis process according to some embodiments can be operated to balance the introduction of sufficient amounts of replacement agent to effectively generate free substances from protein-bound material for removal via dialysis, without resulting in unsafe levels of the substance in the blood during treatment. Thus, a replacement dialysis process according to some embodiments may include determining a replacement agent infusion process or configuration that operates to manage the infusion of replacement agent into the patient's blood during dialysis to facilitate effective removal of the substance while preventing the substance from reaching unsafe levels. A non-limiting example of a replacement dialysis process using an infusion configuration may include a process identical or similar to that described in U.S. Patent Application Publication No. 2019 / 0321537 entitled "Techniques for Removing Bound Target Substances during Dialysis," which is incorporated herein by reference as if fully set forth in this disclosure.

[0032] Removing protein-bound uremic toxins (PBUTs) during hemodialysis and convection-based hemodiafiltration is challenging. Due to their high affinity for albumin, only low levels of free PBUT are available, resulting in a small concentration gradient between the blood and dialysate streams and consequently, limited removal. In some embodiments, a method may include infusing a binding competitor that competes with PBUT for the same binding sites on albumin into the arterial line of the extracorporeal circuit to increase the free fraction, thereby enhancing PBUT removal. Essentially, binding competitor-enhanced hemodialysis is a reactive separation technique, a paradigm shift from passive diffusion-based hemodialysis decades ago. Competitive binding has proven effective for PBUT removal in vitro, in silico, and in a proof-of-concept clinical study involving 18 patients. In some embodiments, mathematical models used for simulating optimal infusion configurations may be employed. In some embodiments, the replacement agent optimization process may use quantitative structure-activity relationship (QSAR) models with machine learning approaches to screen for better replacement agents. Some embodiments may utilize a database of FDA-approved drugs for potential candidate replacement agents that can target multiple binding sites on albumin. In some embodiments, the replacement agent optimization process can identify compounds using a QSAR model of candidate binding competitors and test selected binding competitors in vitro, as well as their long-term effects on predialysis PBUT concentrations and patient-reported outcomes (PROs).

[0033] Uremic toxins are broadly classified into three categories: (1) small toxins (<500 Da), (2) medium and large uremic toxins (>500 Da), and (3) protein-bound uremic toxins (PBUTs). Removal of PBUTs is particularly poor during routine hemodialysis (HD). Convection-based hemodiafiltration or membrane adsorption offers only slightly better results than HD in removing these toxins. In extracorporeal renal replacement therapy, PBUT removal is poor because most toxins are protein-bound; only a small fraction is available for transfer across the dialyzer membrane. Numerous clinical studies have shown that PBUTs have many detrimental effects, including increased mortality in patients with ESRD; enhanced removal of them may improve patient outcomes. Although PBUTs, and in particular indophenol sulfate and p-cresol sulfate, are used as examples in this specification, the examples are not as limited as other types of target compounds contemplated herein.

[0034] Figure 1 An example of an operating environment 100, which may represent some embodiments, is shown. For example... Figure 1 As shown, the operating environment 100 may include an interface with the dialysis machine 160 (e.g., see [link to documentation]). Figure 2 The dialysis system 105 is associated with the dialysis machine 160. In various embodiments, the dialysis system 105 may include a computing device 110 communicatively coupled to the dialysis machine 160. The computing device 110 may operate to manage processes such as replacement agent determination according to some embodiments. In various embodiments, the computing device 110 may operate to manage dialysis processes (e.g., HD processes) and / or replacement agent infusion processes of the dialysis machine 160.

[0035] Despite Figure 1 Only one computing device 110 and dialysis machine 160 are shown in the figures, but the embodiments are not limited thereto. In various embodiments, the functions, operations, configurations, data storage functions, applications, logic units, and / or such described with respect to computing device 110 may be executed and / or stored therein by one or more other computing devices (not shown), for example, coupled to computing device 110 via network 170. A single computing device 110 and dialysis machine 160 are shown for illustrative purposes only to simplify the figures. The embodiments are not limited to this context.

[0036] The computing device 110 may include processor circuitry 120 communicatively coupled to memory unit 130. According to some embodiments, the processing circuitry 120 may include and / or have access to various logic units for performing processes. For example, the processor circuitry 120 may include and / or have access to a displacement determination logic unit 122 and / or a dialysis logic unit 124. The processing circuitry 120, the displacement determination logic unit 122 and / or the dialysis logic unit 124 and / or portions thereof may be implemented in hardware, software, or a combination thereof. As used herein, the terms “logic,” “component,” “layer,” “system,” “circuit,” “decoder,” “encoder,” and / or “module” are intended to refer to computer-related entities, either hardware, a combination of hardware and software, software, or software in execution, examples of which are provided by the exemplary computing architecture 800. For example, a logic unit, circuit, or module can be and / or may include, but is not limited to, a process running on a processor, a processor, a hard disk drive, multiple storage drives (of optical and / or magnetic storage media), an object, an executable file, a thread of execution, a program, a computer, a hardware circuit, an integrated circuit, an application-specific integrated circuit (ASIC), a programmable logic device (PLD), a digital signal processor (DSP), a field-programmable gate array (FPGA), a system-on-a-chip (SoC), a memory unit, a logic gate, a register, a semiconductor device, a chip, a microchip, a chipset, a software component, a program, an application program, firmware, a software module, computer code, any combination of the foregoing, and / or such.

[0037] Despite Figure 1 The displacement agent determination logic unit 122 and the dialysis logic unit 124 are shown as being within the processor circuitry 120, but the embodiment is not limited thereto. For example, the displacement agent determination logic unit 122, the dialysis logic unit 124, and / or any component thereof may be located within an accelerator, a processor core, an interface, a separate processor die, and may be fully implemented as a software application (e.g., displacement agent determination application 140) and / or the like.

[0038] Memory cell 130 may include various types of computer-readable storage media and / or systems in the form of one or more higher-speed memory cells, such as read-only memory (ROM), random access memory (RAM), dynamic RAM (DRAM), dual data rate DRAM (DDRAM), synchronous DRAM (SDRAM), static RAM (SRAM), programmable ROM (PROM), erasable programmable ROM (EPROM), electrically erasable programmable ROM (EEPROM), flash memory, polymer memory such as ferroelectric polymer memory, austenite memory, phase change or ferroelectric memory, silicon oxide-nitride-oxide The storage unit 130 may include SONOS memory, magnetic or optical cards, arrays of devices such as RAID drives, solid-state storage devices (e.g., USB storage, SSDs), and any other type of storage medium suitable for storing information. Furthermore, the storage unit 130 may include various types of computer-readable storage media in the form of one or more low-speed storage units, including internal (or external) hard disk drives (HDDs), floppy disk drives (FDDs), and optical disc drives for reading from or writing to removable optical discs (e.g., CD-ROMs or DVDs), SSDs, and / or the like.

[0039] Memory unit 130 may store a displacement agent determination application 140, which may operate alone or in combination with displacement agent determination logic unit 122 and / or dialysis logic unit 124 to determine displacement agent compounds from a pool of candidate compounds. For example, according to some embodiments, displacement agent determination application 140 may be operable to perform a displacement agent determination process. In another example, dialysis logic unit 124 may be operable to perform a dialysis process (e.g., HD process) via dialysis machine 160.

[0040] In some embodiments, memory unit 130 may store displacement agent information 132. In various embodiments, displacement agent information 132 may include information associated with potential displacement agent candidates. For example, displacement agent information may include candidate pools from compound lists and / or databases, such as the Food and Drug Administration (FDA) database, DrugBank (e.g., version 5.1.4), literature-based compound sets (e.g., Zsila et al., “Evaluation of drug–human serumalbumin binding interactions with support vector machine aided onlineautomated docking,” Bioinformatics 27(13), 1806–1813 (2011) and Zhu et al., “The use of pseudo-equilibrium constant affords improved QSAR models of human plasmaprotein binding,” Pharm. Res. 30(7): 1790–8 (2013)), and / or the like.

[0041] In some embodiments, replacement agent information 132 may include replacement agent requirements, such as the characteristics or properties of compounds that can be selected as replacement agent compounds. For example, a replacement agent compound may be selected as a molecule that can be used long-term for dialysis and end-stage renal disease (ESRD) populations. Non-limiting examples of the properties of replacement agent candidates and / or selected replacement agent compounds may include one or more of the following: sharing the same binding site (e.g., for albumin, sharing one or more albumin binding sites with most identified target compounds, primarily sites I and II); having a high affinity that can compete with the albumin binding affinity of PBUT, or having a dose that is approved (e.g., FDA approved) for administration in humans that will demonstrate a replacement effect; being suitable for the specified delivery procedure (e.g., administration via intravenous (IV) or via another desired procedure is safe and feasible); not being dependent on renal excretion; and being safe for use in the target patient or patient population (e.g., the end-stage renal disease population).

[0042] In some embodiments, replacement agent information 132 may include replacement agent compounds determined via a replacement agent determination process according to some embodiments. In some embodiments, as part of dialysis treatment, dialysis professionals and / or healthcare professionals may consult replacement agent information regarding replacement agent compounds to be used on a patient according to various embodiments.

[0043] In some embodiments, computational model 134 may include a model used by a displacer determination process to determine the displacer compound. In various embodiments, computational model 134 may be or may include one or more quantitative structure-activity relationship (QSAR) models. In some embodiments, a QSAR model may be developed for a target protein (e.g., albumin). In various embodiments, a QSAR model for plasma protein binding (PPB) may be developed. QSAR models according to some embodiments may be constructed to predict the binding affinity of candidate compounds to target proteins.

[0044] In some embodiments, the target protein may have multiple binding sites. For example, the target protein may be albumin, which has two primary binding sites (binding site I and binding site II). A QSAR model can be constructed for each binding site. For example, various embodiments may include a QSAR model for binding site I and a QSAR model for binding site II of albumin. Thus, in some embodiments, separate QSAR models can be used to screen candidates for binding sites I and II and to obtain their affinity for each site. In some embodiments, for example, QSAR models can be developed for both albumin binding and PPB to predict the binding affinity of screened candidates.

[0045] In some embodiments, dialysis information 136 may include patient health information, dialysis treatment or prescription information, target substance information, binding protein information, replacement agent information, configuration information, and / or similar information. In various embodiments, dialysis information 136 may include a replacement agent compound to be used or recommended for use during dialysis. In some embodiments, dialysis information 136 may include an infusion configuration for the replacement agent during dialysis.

[0046] Figure 2An example of an operating environment 200, which may represent some embodiments, is shown. Drug-drug interactions are a well-known phenomenon in pharmacokinetics. Compounds having the same binding site on albumin can interfere with each other's free fraction in plasma, thereby increasing or decreasing drug effects, or even causing adverse reactions. While drug-drug interactions are generally something clinicians try to avoid, binding competition mechanisms can be used to increase the free fraction of PUBT. Higher free fractions of PBUT can be achieved by introducing competitors or a combination of a few competitors in the extracorporeal loop hemodialysis unit. Therefore, higher dialysis removal rates can be expected for PBUT. This treatment method may include three components: a replacement agent or a replacement agent cocktail solution, a programmable infusion pump, and an infusion configuration that can be personalized for each patient. Some embodiments can be operated to address functional areas of toxin removal and secretion, as well as to design target quantities and / or maintain small clearances / reductions (lower blood concentrations) of protein-bound uremic toxins. Figure 2 Competing combinations in a dialysis process according to some embodiments are illustrated. Generally speaking, Figure 2 This diagram illustrates the binding competition between protein-bound uremic toxins and competing drugs for the same binding site. The infusion of either a replacement agent or a competing drug before the dialyzer increases the concentration of free toxins, thereby improving dialysis removal.

[0047] like Figure 2 As shown, the dialysis machine 205 can be operated to cause dialysate inflow of dialysate fluid 204 and dialysate outflow of dialysate fluid along with unwanted substances 206 (see, for example, see...). Figure 7 The patient's blood 202 may include target substances (e.g., phenytoin) 210 bound to a target protein (e.g., albumin) 220 and free or unbound target substances 210. Unbound target substances 210 can pass through the dialysis membrane 250 and are removed as unwanted substance 206 with the dialysate. Bound target substances 210 cannot pass through the dialysis membrane 250 and therefore cannot be removed as unwanted substance 206 with the dialysate.

[0048] In some embodiments, the dialysis machine 205 may include a replacement agent container 240 or be in fluid communication with the replacement agent container 240, the replacement agent container 240 being operated to facilitate the flow of replacement agent 230 via the patient's blood into the infusion of the replacement agent 230 into the patient's bloodstream 202. Figure 2 As shown, the replacement agent 230 can compete for binding sites on albumin 220, resulting in a reduction (or even elimination) of the bound target substance 210 and an increase in the free target substance 210. The increase in the free target substance 210 can promote removal from the patient's blood 202, or remove a greater amount of phenytoin 210 from the patient's blood 202 than would be in the absence of the replacement agent 230.

[0049] In some embodiments, the replacement agent 230 may be selected based on the replacement agent determination process according to this disclosure. In various embodiments, the replacement agent 230 may be infused into the patient based on the infusion configuration according to this disclosure.

[0050] The replacement dialysis method according to some embodiments has been tested in vitro, ex vivo, and in at least one clinical study. Among the many compounds tested in in vitro studies, ibuprofen has been used as a prototype contender in a proof-of-concept clinical study in 18 HD patients. For example, ibuprofen (800 mg) was infused from the 20th to the 40th minute during routine high-flux hemodialysis. Dialysate clearance was observed to increase from 6.6 to 20 mL / min for indophenol sulfate (IS) and from 4.4 to 14.9 mL / min for p-cresol sulfate (pCS) (two prominent examples of PBUT). Ibuprofen infusion was well tolerated in all patients. Figure 3 Exemplary graphs showing dialysate clearance rates of target substances during various dialysis processes according to this disclosure are provided. Generally, Figure 3 Figures 310-315 show information related to dialysis clearance for different solutes. A significant increase in the clearance of IS and pCS was observed between the pre-infusion and infusion phases, but no increase in the clearance of urea and creatinine (non-protein-bound solutes).

[0051] In some embodiments, one or more mathematical models, calibrated and / or validated, may be used, for example, by using clinical data. These models can be used to develop infusion configurations to maximize the efficacy of the replacement agent and / or minimize its dosage. Optimized infusion configurations can be used to control the infusion pump during treatment. For example, if the replacement agent is well tolerated but requires a high dose during hemorrhage (HD) due to its weak binding affinity, the infusion configuration can guide the infusion pump to achieve maximum replacement during HD. On the other hand, according to some embodiments, the efficacy of some replacement agents can be improved by using infusion configurations, thereby minimizing the residual blood concentration of certain replacement agents. Figure 4 Exemplary diagrams of infusion configurations according to some embodiments of this disclosure are shown. In particular, Figure 4 Figures 410-413 illustrate an example infusion configuration 414, along with simulated concentrations of indophenol sulfate in plasma and interstitial pools 411-413. In some embodiments, according to certain examples, such an infusion configuration can be used to develop and / or control hardware and / or software solutions for programming and / or controlling the infusion pump during dialysis procedures.

[0052] Some embodiments may provide a displacement determination process that operates using QSAR modeling. For example, one or more QSAR models may be generated to determine displacement agents, such as compounds that bind to a target protein. A non-limiting example of a target protein may be albumin. In one embodiment with an albumin target protein, a QSAR model may be used to screen compounds for site I and site II binding and to obtain their affinity for each site. In some embodiments, QSAR models may be developed for both albumin binding and plasma protein binding (PPB) to predict the binding affinity of screened drugs. In one case study, a set of 1240 chemicals was used along with known plasma protein binding data from the literature (e.g., see Zhu et al.) to construct an aPPB model. This model can be tested on a set of candidates with known albumin binding data. Both models can evaluate the binding affinity of the tested candidate compounds.

[0053] Additional classification QSAR models (e.g., using machine learning (ML) and / or artificial intelligence (AI) methods) can be developed for binding site I and binding site II respectively to distinguish the binding site for each compound. For the binding site I model, the obtained equilibrium accuracy is likely to be in the range of 86% to 91%, and for the binding site II model, it is likely to be in the range of 79% to 88%. Candidates can be accessed from various sources. In a non-limiting example, candidates may include FDA-approved drugs with reported intravenous (IV) administration routes (e.g., extracted from the DrugBank database). In one example, 169 binding site I binders and 149 drug site II binders were identified. They were tested against PPB and human serum albumin (HAS) regression QSAR models to estimate the binding affinity of these drugs. The results include a list of binding site I binders and a list of binding site II binders, where log K(log[%PPB / (101-%PPB)]) and log K are used as indicators of binding affinity, respectively. The overlap of these two lists represents those compounds that bind to both sites I and II.

[0054] Solutions according to some embodiments are not limited to improving mortality rates in patients with ESRD. For example, processes according to some embodiments can improve patients' quality of life (QoL) by providing more treatment options and reducing disease and treatment complications. For example, increased p-cresol sulfate has been reported to be associated with the severity of pruritus as measured by the 5-D pruritus grading system. Uremic pruritus accounts for moderate to severe pruritus in 24% of CKD patients in the United States and is associated with patients reporting decreased quality of life, more severe depressive symptoms, and restless sleep. Hemoperfusion and oral charcoal have been tested as treatment options for pruritus. These previous studies suggest that improving the dialysis removal of protein-bound uremic toxins may help reduce ESRD-related complications, thereby improving patients' quality of life.

[0055] Some embodiments can be operated to improve the model to, for example, distinguish the major binding site of each compound via a protein-ligand docking approach and predict site-specific equilibrium constants. In some embodiments, for example, a final list of top displacement agent candidates can be screened in a laboratory setting to confirm their displacement capacity for multiple known PBUTs. Based on laboratory results, one or more of the best candidates can be selected for clinical studies.

[0056] In some embodiments, candidates may be determined based on machine learning (ML) and / or AI models. In various embodiments, such candidate selection and / or optimization models may take into account potential adverse effects. For example, as with many other blood purification techniques that rely on solute diffusion and convection, the proposed solution may non-selectively enhance the removal of toxins and beneficial substances from the blood. The net benefit of the proposed solution may be determined according to models and / or processes of some embodiments, for example, to determine the balance point between gain and loss to select the optimal replacement agent compound. For example, some embodiments may determine whether a replacement agent (alone or in combination with certain pre-selected infusion configurations) can improve the removal of most PBUTs without clinically relevant side effects.

[0057] The procedures described in some embodiments can provide improved clinical outcomes and rapid penetration within current dialysis protocols because the competing combination technology is inexpensive and easy to implement, requiring only very minor modifications to current HD technology. For example, the replacement agent can be infused via a built-in heparin pump that is not typically used (many dialysis providers use rapid heparinization instead). Programmable infusion pumps for infusing replacement agents with optimized infusion configurations are also inexpensive. According to some embodiments, such optimal infusion configurations can be obtained using mathematical models, patient characteristics, the selection of competing drugs, etc.

[0058] Example:

[0059] QSAR models were developed for albumin binding and plasma protein binding (PPB) to predict the binding affinity of chemicals. A PPB model was constructed using a set of 1240 chemicals and literature data on known PPB binding. The model was tested on a set of molecules with known albumin binding data. Comparison of the results from the PPB and albumin binding models revealed no significant difference in performance. Both models can assess the binding affinity of the tested compounds. Better agreement between the two models facilitates easier prediction of the compounds.

[0060] The next step is to distinguish binding by developing QSAR models for binding site 1 (or site I) and binding site 2 (or site II) separately. Classification models ("yes" or "no") have been developed for these two sites.

[0061] Of the 1240 chemicals used to construct the PPB model, 69 with low PPB binding affinity were selected as negative site I binders, and a set of 61 positive albumin site I-binding chemicals from the literature (e.g., Zsila et al.) were used to train the classification QSAR model. The model was validated using a set of 15 known site I binders from individual sources (e.g., Kratochwil et al., “Predicting plasma protein binding of drugs: a new approach,” Biochemical Pharmacology, V.64, No. 9, pp. 1355-1374 (2002)) and 13 negative site I binders selected from the literature (e.g., Zsila et al.). Six QSAR models were developed using three different molecular descriptors and two machine learning methods. The equilibrium accuracy obtained ranged from 86% to 91%.

[0062] Using a similar approach, an eight-site II binding model with four different molecular descriptors and two machine learning methods was constructed. The balanced accuracy achieved for the training set was between 89% and 99%, while for the validation set it was between 79% and 88%.

[0063] The next step was to apply these QSAR models to select potential site I and II binders separately from the FDA-approved drug database. For example, a drug database was downloaded from the DrugBank website (v5.1.4). Only drugs with an intravenous (IV) route of administration from the FDA-approved records were extracted from this database. A total of 514 drugs were tested against two HSA site I classification QSAR models to identify potential binders for HSA site I, and two HSA site II classification QSAR models to identify potential binders for HSA site II. Then, 169 site I binders and 149 drug site II binders were identified and tested against PPB and HSA regression QSAR models to determine the binding affinity of these drugs. The results included a list of site I binders labeled with log K(log[%PPB / (101-%PPB)] as binding affinity and a list of site II binders labeled with log K as binding affinity. The overlap between these two lists represents those compounds that bind to both sites I and II.

[0064] Some embodiments can be used to optimize or otherwise improve the model, for example, by distinguishing the major binding sites of each compound and predicting site-specific equilibrium constants via protein-ligand docking methods.

[0065] The dataset was randomly divided into training (1019 compounds) and test (113 chemicals) sets. The LogK value (LogK = log[%PPB / (101-%PPB]) was used as input.

[0066] After removing duplicates and outliers, the final dataset contained 1132 chemicals. This dataset was randomly divided into training (1019 compounds) and test (113 chemicals) sets. Modeling was run using different descriptors, and a consensus model was built (e.g., without Chemaxon, as it had the lowest accuracy). The developed model achieved reasonable correlations, with R² = 0.75 for the training set and R² = 0.79 for the test set. The model showed R² = 0.74 and RMSE of 0.46 for predicting albumin binding of a set of 120 chemicals.

[0067] This document includes one or more logical flows representing exemplary methods for performing novel aspects of the disclosed architecture. Although, for simplicity of explanation, the one or more methods herein are shown and described as a series of actions, those skilled in the art will understand and recognize that these methods are not limited by the order of actions. Accordingly, some actions may occur in a different order and / or concurrently with other actions shown and described herein. For example, those skilled in the art will understand and recognize that a method may alternatively be represented as a series of interrelated states or events, such as in a state diagram. Furthermore, not all behaviors described in a method are necessary for a new implementation. Blocks designated by dashed lines may be optional blocks of the logical flow.

[0068] The logical flow can be implemented in software, firmware, hardware, or any combination thereof. In software and firmware embodiments, the logical flow can be implemented through computer-executable instructions stored on a non-transitory computer-readable medium or a machine-readable medium. Embodiments are not limited to this context.

[0069] Figure 5 An embodiment of logic flow 500 is illustrated. Logic flow 500 may represent some or all of the operations performed by one or more embodiments described herein, such as computing device 110 and / or its components. In some embodiments, logic flow 500 may represent some or all of the operations of constructing a QSAR model.

[0070] Logic flow 500 can determine a selected dataset 550. For example, in box 530, logic flow 500 can access chemical dataset 502 and perform the removal of mixtures and / or inorganics, in box 532 perform structure transformation cleaning / removal of salts, in box 534 perform standardization for specific chemical types, in box 536 perform treatment of tautomers, in box 538 perform analysis / removal of structural duplications, and / or perform manual inspection in box 540 to obtain a managed dataset 504. In some embodiments, determining the managed dataset 550 may be the same as or similar to the process described in Fourches et al., “Trust, but verify: On the importance of chemical structure curation in cheminformatics and QSAR modeling research,” J. Chem. Inf. Model, 50(7), 1189-1204 (2010).

[0071] In box 510, logic flow 500 can perform descriptor generation. In box 512, logic flow 500 can further split the dataset into training, testing, and / or external validation sets. In box 514, logic flow 500 can apply machine learning techniques. In box 516, logic flow 500 can select a model with high internal and / or high external accuracy. In box 518, logic flow 500 can determine the evaluation of the applicability domain. In box 520, the logic flow can construct and / or select one or more predictive QSAR models.

[0072] Figure 6 An embodiment of logic flow 600 is illustrated. Logic flow 600 may represent some or all of the operations performed by one or more embodiments described herein, such as computing device 110 and / or its components. In some embodiments, logic flow 600 may represent some or all of the operations of determining a displacement agent compound via a displacement agent determination process.

[0073] In block 602, logic flow 600 may identify a target protein. For example, a target protein that binds to a target compound (e.g., PBUT) may be identified. A non-limiting example of a target protein may include albumin. In block 604, logic flow 600 may determine the properties of the displacer. In some embodiments, the properties of the displacer may include desired characteristics of the displacer compound. For example, the properties of the displacer may include binding sites (e.g., binding to one or more binding sites of the target protein, binding to only one binding site), binding affinity (e.g., having minimum and / or maximum binding affinity), patient safety, solubility, binding to other compounds (e.g., not binding to certain other compounds in the blood), and / or the like.

[0074] Non-limiting examples of the properties of replacement agent candidates and / or selected replacement agent compounds may include one or more of the following: sharing the same binding site (e.g., for albumin, sharing albumin binding sites with most identified target compounds, primarily sites I and II); having a high affinity that can compete with the albumin binding affinity of PBUT, or having an approved dose in humans that will demonstrate a replacement effect; being safe and feasible for intravenous (IV) administration; not depending on renal excretion; and being safe for use in end-stage renal disease populations.

[0075] In block 606, logic flow 600 can generate QSAR models. In some embodiments, multiple QSAR models can be generated. Non-limiting examples of QSAR models may include a target protein QSAR model, a PPB model, and / or one or more models for each potential binding site.

[0076] In various embodiments, QSAR models of plasma protein binding (PPB) can be developed. QSAR models based on some embodiments can be constructed to predict the binding affinity of candidate compounds to target proteins.

[0077] In some embodiments, the target protein may have multiple binding sites. For example, the target protein may be albumin, which has two primary binding sites (binding site I and binding site II). A QSAR model can be constructed for each binding site. For example, various embodiments may include a QSAR model for binding site I and a QSAR model for binding site II of albumin. Thus, in some embodiments, separate QSAR models can be used to screen candidates for binding sites I and II and to obtain their affinity for each site. In some embodiments, for example, QSAR models can be developed for both albumin binding and PPB to predict the binding affinity of screened candidates.

[0078] In one example with an albumin target protein, the QSAR model for each binding site can be operated by developing separate QSAR models for binding site 1 (or site I) and binding site 2 (or site II) to distinguish binding. In some embodiments, the QSAR model can operate as a classification (i.e., binary or / or) model, indicating whether a candidate can bind to a specific binding site. In various embodiments, the QSAR model can operate as a quantification model, for example, indicating binding affinity (not just or, but based on binding affinity values, classification) or other characteristics (e.g., solubility, toxicity, and / or the like).

[0079] In some embodiments, for example, a set of K compounds with known binding information can be used to construct a QSAR model. It can be known that a subset of the K compounds does not bind to the target protein (e.g., has negative binding affinity), and it can be known that a subset of the K compounds does bind to the target protein. Compounds with known / unknown binding affinity can be used to train and / or validate the QSAR model.

[0080] In block 608, logic flow 600 can use a QSAR model to analyze candidate compounds. For example, a QSAR model generated according to some embodiments can be used to analyze lists, collections, databases, and / or other sources of candidate compounds. In various embodiments, the output from the QSAR model can indicate binding affinity to one or more binding sites of the target protein. For example, the results of using a QSAR model on binding sites I and II can include a list of site I and a list of site II binders, respectively labeled with binding affinity of log K(log[%PPB / (101-%PPB)] and log K. The overlap of these two lists represents those compounds that bind to both site I and site II.

[0081] In box 610, logic flow 600 may determine a replacement agent compound. For example, selection criteria may be used to select compounds listed based on the analysis performed in box 608. Selection criteria may include availability, toxicity, ability to be integrated into the dialysis process, cost, and / or the like. In some embodiments, compounds determined via the analysis in box 608 may be selected and tested, including in silico, in vivo, in vitro, and / or clinical trials. Examples are not limited to this context.

[0082] In block 612, logic flow 600 may perform treatment. In some embodiments, performing treatment may include providing treatment recommendations using a determined replacement agent compound, infusion configuration, and / or similar methods. In various embodiments, performing treatment may include performing dialysis treatment using a replacement agent compound determined according to a replacement agent determination process according to some embodiments. Performing treatment may include computer-assisted, manual (i.e., performed by a healthcare professional), and / or a combination thereof.

[0083] Figure 7A schematic diagram of an exemplary embodiment of a dialysis system 700 according to the present disclosure is shown. The dialysis system 700 can be configured to provide hemodialysis (HD) treatment to a patient 701. A fluid reservoir 702 can deliver fresh dialysate to a dialyzer 704 via a tube 703, and once the dialysate has passed through the dialyzer 704, a reservoir 706 can receive used dialysate via a tube 705. The hemodialysis operation can filter particles and / or contaminants from the patient's blood through an external filtration device, such as the dialyzer 704. As the dialysate passes through the dialyzer 704, unfiltered patient blood also enters the dialyzer 704 via a tube 707, and filtered blood is returned to the patient 701 via a tube 709. Arterial pressure can be monitored via a pressure sensor 710, inflow pressure via a sensor 718, and venous pressure via a pressure sensor 714. An air trap and detector 716 ensure that air is not introduced into the patient's bloodstream when it is filtered and returned to the patient 701. The flow of blood and dialysate can be controlled via corresponding pumps, including a blood pump 712 and a fluid pump 720. Heparin 722, a blood thinner, can be used in combination with saline 724 to ensure that blood clots do not form or obstruct blood flow through the system.

[0084] In some embodiments, the dialysis system 700 may include a controller 750, which may resemble the computing device 110 and / or components thereof (e.g., processor circuitry 120). The controller 750 may be configured to monitor fluid pressure readings to identify fluctuations in patient parameters such as heart rate and / or respiratory rate. In some embodiments, the patient's heart rate and / or respiratory rate may be determined by fluid pressure in fluid flow lines and fluid bags. The controller 750 may also be operatively connected to or communicate with additional sensors or sensor systems, devices, and / or such devices, although the controller 750 may use any available data regarding the patient's biological function or other patient parameters. For example, according to some embodiments, the controller 750 may send patient data to the computing device 110 for processing.

[0085] Figure 8 An embodiment of an exemplary computing architecture 800 suitable for implementing the various embodiments described above is illustrated. In various embodiments, the computing architecture 800 may include an electronic device or be implemented as part of an electronic device. In some embodiments, the computing architecture 800 may represent, for example, a computing device 802 and / or components thereof. The embodiments are not limited to this context.

[0086] As used in this application, the terms "system," "component," and "module" refer to a computer-related entity, either hardware, a combination of hardware and software, software, or software in execution, examples of which are provided by the exemplary computing architecture 800. For example, a component can be, but is not limited to, a process running on a processor, a processor, a hard disk drive, multiple storage drives (of optical and / or magnetic storage media), an object, an executable file, an execution thread, a program, and / or a computer. As an illustration, an application running on a server and a server can both be components. One or more components may reside in a process and / or an execution thread, and components may be located on a single computer and / or distributed across two or more computers. Furthermore, components can communicatively couple with each other to coordinate operation via various types of communication media. Coordination may involve one-way or two-way exchange of information. For example, components may transmit information in the form of signals transmitted on a communication medium. This information can be implemented as signals assigned to various signal lines. In such an assignment, each message is a signal. However, other embodiments may alternatively use data messages. Such data messages can be sent via various connections. Exemplary connections include parallel interfaces, serial interfaces, and bus interfaces.

[0087] The computing architecture 800 includes various common computing elements such as one or more processors, multi-core processors, coprocessors, memory units, chipsets, controllers, peripherals, interfaces, oscillators, timing devices, video cards, audio cards, multimedia input / output (I / O) components, power supplies, etc. However, embodiments are not limited to those implemented by the computing architecture 800.

[0088] like Figure 8 As shown, the computing architecture 800 includes a processing unit 804, a system memory 806, and a system bus 808. The processing unit 804 can be any of a variety of commercially available processors, including but not limited to... and processor; Application, embedded, and security processors; and and Processors; IBM and Cell processor; Core(2) and Processors; and similar processors. Dual microprocessors, multi-core processors, and other multiprocessor architectures can also be used as processing units 804.

[0089] System bus 808 provides the processing unit 804 with an interface for system components, including but not limited to system memory 806. System bus 808 can be any of a variety of bus structures that can be further interconnected to a memory bus (with or without a memory controller), peripheral bus, and local bus using any of a variety of commercially available bus architectures. Interface adapters can be connected to system bus 808 via a slot architecture. Example slot architectures can include, but are not limited to, Accelerated Graphics Port (AGP), Card Bus, (Extended) Industry Standard Architecture ((E)ISA), Micro Channel Architecture (MCA), NuBus, Peripheral Component Interconnect (Extended) (PCI(X)), PCI Express, PCMCIA, etc.

[0090] System memory 806 may include various types of computer-readable storage media in the form of one or more higher-speed memory cells, such as read-only memory (ROM), random access memory (RAM), dynamic RAM (DRAM), dual data rate DRAM (DDRAM), synchronous DRAM (SDRAM), static RAM (SRAM), programmable ROM (PROM), erasable programmable ROM (EPROM), electrically erasable programmable ROM (EEPROM), flash memory, polymer memory such as ferroelectric polymer memory, austenite memory, phase change or ferroelectric memory, silicon oxide-nitride-oxide-silicon (SONOS) memory, magnetic cards or optical cards, arrays of devices such as redundant array of independent disk drives (RAID), solid-state memory devices (e.g., USB storage, solid-state drives (SSDs), and any other type of storage media suitable for storing information. Figure 8 In the illustrated embodiment, system memory 806 may include non-volatile memory 810 and / or volatile memory 812. The Basic Input / Output System (BIOS) may be stored in non-volatile memory 810.

[0091] Computer 802 may include various types of computer-readable storage media in the form of one or more lower-speed memory cells, including internal (or external) hard disk drive (HDD) 814, magnetic floppy disk drive (FDD) 816 for reading from or writing to removable disk 818, and optical disc drive 820 for reading from or writing to removable optical disc 822 (e.g., CD-ROM or DVD). HDD 814, FDD 816, and optical disc drive 820 may be connected to system bus 808 via HDD interface 824, FDD interface 826, and optical drive interface 828, respectively. HDD interface 824 for external drive implementation may include at least one or both of Universal Serial Bus (USB) and IEEE 1384 interface technologies.

[0092] Drives and associated computer-readable media provide volatile and / or non-volatile storage of data, data structures, computer-executable instructions, etc. For example, numerous program modules may be stored in drive and memory units 810, 812, including an operating system 830, one or more application programs 832, other program modules 834, and program data 836. In one embodiment, the one or more application programs 832, other program modules 834, and program data 836 may include, for example, various applications and / or components of computing device 110.

[0093] Users can input commands and information into computer 802 through one or more wired / wireless input devices, such as keyboard 838 and pointing devices such as mouse 840. Other input devices may include microphones, infrared (IR) remote controls, radio frequency (RF) remote controls, game pads, styluses, card readers, dongles, fingerprint readers, gloves, graphics tablets, joysticks, keyboards, retinal readers, touchscreens (e.g., capacitive, resistive, etc.), trackballs, trackpads, sensors, styluses, etc. These and other input devices are typically connected to processing unit 804 via input device interface 842 coupled to system bus 808, but may also be connected via other interfaces such as parallel ports, IEEE 884 serial ports, game ports, USB ports, IR interfaces, etc.

[0094] Monitor 844 or other types of display devices are also connected to system bus 808 via an interface such as video adapter 846. Monitor 844 can be internal or external to computer 802. In addition to monitor 844, computer typically includes other peripheral output devices such as speakers and printers.

[0095] Computer 802 can operate in a network environment using logical connections to one or more remote computers, such as remote computer 848, via wired and / or wireless communications. Remote computer 848 can be a workstation, server computer, router, personal computer, portable computer, microprocessor-based entertainment device, peer-to-peer device, or other public network node, and typically includes many or all of the elements described relative to computer 802, although for simplicity only memory / storage device 850 is shown. The depicted logical connections include wired / wireless connections to a local area network (LAN) 852 and / or a larger network, such as a wide area network (WAN) 854. Such LAN and WAN networking environments are commonplace in offices and companies and facilitate enterprise-wide computer networks, such as intranets, all of which can connect to global communication networks, such as the Internet.

[0096] When used in a LAN networking environment, computer 802 connects to LAN 852 via a wired and / or wireless communication network interface or adapter 856. Adapter 856 facilitates wired and / or wireless communication to LAN 852, and LAN 852 may also include a wireless access point configured thereon for communicating with the wireless functionality of adapter 856.

[0097] When used in a WAN networking environment, computer 802 may include modem 858, or a communication server connected to WAN 854, or other devices that establish communication via WAN 854, such as via the Internet. Modem 858, which may be internal or external and may be a wired and / or wireless device, is connected to system bus 808 via input device interface 842. In a network environment, program modules relative to computer 802 or partially described herein may be stored in remote memory / storage device 850. It should be understood that the network connections shown are exemplary, and other methods for establishing communication links between computers may be used.

[0098] The computer 802 is operable to communicate with wired and wireless devices or entities using the IEEE 802 family of standards, such as wireless devices operably configured in wireless communication (e.g., IEEE 802.16 air modulation technology). This includes at least Wi-Fi (or Wireless Fidelity), WiMax, and Bluetooth. TM Wireless technologies, etc. Therefore, communication can be a predefined structure like a regular network, or simply simple, self-organizing communication between at least two devices. Wi-Fi networks use radio technology known as IEEE 802.11x (a, b, g, n, etc.) to provide secure, reliable, and fast wireless connectivity. Wi-Fi networks can be used to connect computers to each other, connect to the Internet, and connect to wired networks (which use media and functions associated with IEEE 802.3).

[0099] This document has set forth numerous specific details to provide a thorough understanding of the embodiments. However, those skilled in the art will understand that the embodiments can be practiced without these specific details. In other instances, well-known operations, components, and circuits have not been described in detail so as not to obscure the embodiments. It is understood that the specific structural and functional details disclosed herein may be representative and do not necessarily limit the scope of the embodiments.

[0100] Some embodiments may use the expressions “coupled” and “connected” and their derivatives for description. These terms are not intended to be synonymous with each other. For example, some embodiments may use the terms “connected” and / or “coupled” to indicate that two or more elements are in direct physical or electrical contact with each other. However, the term “coupled” may also indicate that two or more elements are not in direct contact with each other, but still cooperate or interact with each other.

[0101] Unless otherwise expressly stated, terms such as “processing,” “computing,” “determining,” etc., refer to the actions and / or processes of a computer or computing system or similar electronic computing device that manipulate and / or convert data represented as physical quantities (e.g., electrons) within the registers and / or memory of the computing system into other data similarly represented as physical quantities within the memory, registers, or other such information storage, transmission, or display devices of the computing system. Embodiments are not limited to this context.

[0102] It should be noted that the methods described herein need not be performed in the order stated or in any particular order. Furthermore, the various activities described with respect to the methods identified herein can be performed serially or in parallel.

[0103] Although specific embodiments have been shown and described herein, it should be understood that any arrangement calculated to achieve the same purpose may replace the specific embodiments shown. This disclosure is intended to cover any and all modifications or variations of the various embodiments. It should be understood that the above description is illustrative and not restrictive. Combinations of the above embodiments and other embodiments not specifically described herein will be apparent to those skilled in the art upon review of the above description. Therefore, the scope of the various embodiments includes any other applications in which the above-described components, structures, and methods are used.

[0104] Although the subject matter has been described in language specific to structural features and / or methodological actions, it should be understood that the subject matter defined in the appended claims is not necessarily limited to the specific features or actions described above. Rather, the specific features and actions described above are disclosed as exemplary forms for implementing the claims.

Claims

1. A method for determining a displacement agent compound for binding to a target protein having multiple binding sites, the method comprising: Multiple machine learning quantitative structure-activity relationship (QSAR) models were determined for the target albumin. The multiple QSAR models include at least one of a binding site IQSAR model configured to predict the binding affinity of the candidate to binding site I of albumin and a binding site IIQSAR model configured to predict the binding affinity of the candidate to binding site II of albumin. The multiple QSAR models were trained using known albumin binding training data to predict the binding affinity of compounds with unknown binding affinity to binding site I of albumin in QSAR or binding site 2 of albumin in QSAR. The multiple QSAR models are used to analyze a set of candidate compounds to identify a set including at least one potential compound having an affinity for binding to one of the multiple binding sites; as well as At least one displacement agent compound is selected from the set including the at least one potential compound.

2. The method of claim 1, wherein, The multiple QSAR models include at least one albumin-binding model and at least one plasma protein-binding (PPB) model.

3. The method of claim 1 or 2, wherein, The replacement agent compound determined by the binding site I model has a binding affinity of log K = log[%PPB / (101-%PPB)], and the replacement agent compound determined by the binding site II model can have a binding affinity of log K.

4. The method of claim 1 or 2, wherein, The set of candidate compounds is associated with replacement agent properties, wherein the replacement agent properties include at least one of the following: the candidate compound shares the same binding site with the target protein; the candidate compound has a binding affinity exceeding a threshold; the candidate compound is associated with an approved dose that provides the replacement effect; the candidate compound is suitable for a specified delivery procedure; the candidate compound does not require renal excretion; and the candidate compound is safe for the target patient population.

5. An apparatus comprising: At least one processor; A memory coupled to the at least one processor, the memory including instructions that, when executed by the at least one processor, cause the at least one processor to: Multiple machine learning quantitative structure-activity relationship (QSAR) models were determined for the target albumin. The multiple QSAR models include at least one of a binding site IQSAR model configured to predict the binding affinity of the candidate to binding site I of albumin and a binding site IIQSAR model configured to predict the binding affinity of the candidate to binding site II of albumin. The multiple QSAR models were trained using known albumin binding training data to predict the binding affinity of compounds with unknown binding affinity to binding site I of albumin in QSAR or binding site 2 of albumin in QSAR. The multiple QSAR models are used to analyze a set of candidate compounds to identify a set including at least one potential compound having an affinity for binding to one of the multiple binding sites; as well as At least one displacement agent compound is selected from the set including the at least one potential compound.

6. The apparatus of claim 5, wherein, The multiple QSAR models include at least one albumin-binding model and at least one plasma protein-binding (PPB) model.

7. The apparatus of claim 5 or 6, wherein, The replacement agent compound determined by the binding site I model has a binding affinity of log K = log[%PPB / (101-%PPB)], and the replacement agent compound determined by the binding site II model can have a binding affinity of log K.

8. The apparatus of claim 5 or 6, wherein, The set of candidate compounds is associated with replacement agent properties, wherein the replacement agent properties include at least one of the following: the candidate compound shares the same binding site with the target protein; the candidate compound has a binding affinity exceeding a threshold; the candidate compound is associated with an approved dose that provides the replacement effect; the candidate compound is suitable for a specified delivery procedure; the candidate compound does not require renal excretion; and the candidate compound is safe for the target patient population.

Images