KR-20260064582-A - Drug Resistance Prediction Method and Device Thereof

KR20260064582AKR 20260064582 AKR20260064582 AKR 20260064582AKR-20260064582-A

Abstract

The present invention provides a method for predicting the occurrence of drug resistance implemented by a processor, comprising the steps of: receiving information about a drug and a protein; determining the binding strength and major binding sites of the drug and the protein using the received information about the drug and the protein; and determining the probability of protein mutation based on the received information about the drug and the protein using a protein mutation prediction module configured to output the probability of protein mutation occurring with the information about the binding strength and major binding sites of the drug and the protein as input. The invention also provides a device and a system for providing information using the same.

Inventors

김창중
박태환
조윤성
모하마드 하산 베이그

Assignees

비엔제이바이오파마 주식회사

Dates

Publication Date: 20260507
Application Date: 20251027
Priority Date: 20241030

Claims (20)

As a method for predicting the occurrence of drug resistance implemented by a processor, Step of receiving information about drugs and proteins; A step of determining the binding strength and major binding sites of the drug and protein using the information on the received drug and protein above; A method comprising the step of determining the likelihood of protein mutation based on received information about a drug and a protein, using a protein mutation prediction module configured to output the likelihood of protein mutation based on information about the binding affinity and major binding sites of the drug and the protein as input. Method for predicting the occurrence of drug resistance.
In Article 1, Information regarding the above drugs and proteins is, The having at least one of genomic information, pathological tissue information, protein structure, and drug-resistance information, Method for predicting the occurrence of drug resistance.
In Article 1, The above protein mutation prediction module is, including an artificial intelligence classification model pre-trained using protein mutation data collected from a database containing protein information, Method for predicting the occurrence of drug resistance.
In Article 1, The step of determining the binding strength and major binding sites of the drug and protein using the information on the received drug and protein above is: A step comprising performing docking simulation and molecular dynamics simulation for drugs and proteins, Method for predicting the occurrence of drug resistance.
In Paragraph 4, The step of determining the binding strength and major binding sites of the above-mentioned drug and protein is, A step further comprising analyzing the major binding and binding strength of major amino acids between the drug and the protein from the above simulation results, Method for predicting the occurrence of drug resistance.
In Article 5, The step of determining the likelihood of protein mutation based on information regarding the received drug and protein, using a protein mutation prediction module configured to output the likelihood of protein mutation based on information regarding the binding affinity and major binding sites of the above-mentioned drug and protein as input, comprises: A step of analyzing the major binding and binding strength of major amino acids between the drug and the protein from the above simulation results; and A step comprising determining the possibility of mutation based on the major binding and binding strength of major amino acids between the analyzed drug and protein, Method for predicting the occurrence of drug resistance.
In Article 1, The step of determining the likelihood of protein mutation based on received information regarding the drug and protein, using a protein mutation prediction module configured to output the likelihood of protein mutation based on information regarding the binding of the above-mentioned drug and protein as input, comprises: A step of predicting the likelihood of protein mutation occurrence and generating a virtual protein mutation; and A step further comprising determining the stability and likelihood of occurrence of the virtual protein mutation generated above, Method for predicting the occurrence of drug resistance.
In Article 7, The stability of the above-mentioned hypothetical protein mutation implies a thermodynamic change in the protein structure, Method for predicting the occurrence of drug resistance.
In Article 8, The stability and likelihood of occurrence of the above protein mutation are determined by measuring the value of any one of the potential energy change, van der Waals energy change, electrostatic energy change, and RMS gradient score, Method for predicting the occurrence of drug resistance.
In Article 9, A method further comprising the step of determining and removing an unstable protein if the measured values of the potential energy change, van der Waals energy change, and electrostatic energy change correspond to any one of a potential energy change of 25% or more, a van der Waals energy change of 25% or more, and an electrostatic energy change of 20% or more. Method for predicting the occurrence of drug resistance.
In Article 10, After the step of determining the stability and occurrence probability of the above-mentioned hypothetical mutant protein, As a step of analyzing the binding affinity between the above-mentioned hypothetical mutant protein and the drug, A step of reanalyzing the binding of the above-generated mutant protein and the drug; and A further step comprising comparing the above reanalysis results with existing drug and protein binding analysis results, Method for predicting the occurrence of drug resistance.
A communication unit configured to receive information about drugs and proteins, and It includes a processor functionally connected to the above-mentioned communication unit, The above processor is, Configured to predict the occurrence of drug resistance using a protein mutation prediction module configured to output the probability of protein mutation occurrence based on information regarding the binding affinity and major binding sites of drugs and proteins as input, Device for predicting the occurrence of drug resistance.
In Article 12, Information regarding the above drugs and proteins is, The having at least one of genomic information, pathological tissue information, protein structure, and drug-resistance information, Device for predicting the occurrence of drug resistance.
In Article 12, The above protein mutation prediction module is, including an artificial intelligence classification model pre-trained using protein mutation data collected from a database containing protein information, Device for predicting the occurrence of drug resistance.
In Article 12, The above processor is, Further configured to perform docking simulations and molecular dynamics simulations for drugs and proteins, Device for predicting the occurrence of drug resistance.
In Article 15, The above processor is, Further configured to analyze the major binding and binding strength of major amino acids between drugs and proteins from the above simulation results, Device for predicting the occurrence of drug resistance.
In Article 16, The above processor is, From the above simulation results, analyze the major binding and binding strength of key amino acids between the drug and the protein, and Further configured to determine the potential for mutation based on the major binding and binding strength of major amino acids between the analyzed drugs and proteins, Device for predicting the occurrence of drug resistance.
In Article 12, A processor configured to determine the probability of protein mutation based on received information regarding the drug and protein, using a protein mutation prediction module configured to output the probability of protein mutation occurrence with information regarding the binding of the above drug and protein as input, Further configured to generate virtual protein mutations by determining the probability of protein mutation occurrence, Device for predicting the occurrence of drug resistance.
In Article 18, A processor configured to generate a virtual protein mutation by determining the probability of the above-mentioned protein mutation occurring, Further configured to determine the stability and likelihood of occurrence of the virtual protein mutation generated above, Device for predicting the occurrence of drug resistance.
In Article 19, The stability of the above-mentioned hypothetical protein mutation implies a thermodynamic change in the protein structure, Device for predicting the occurrence of drug resistance.

Description

Drug Resistance Prediction Method and Device Thereof The present invention relates to a method for predicting the occurrence of drug resistance and an apparatus thereof. Drugs generally exert therapeutic effects by binding to target proteins in the body that cause specific diseases, thereby inhibiting or regulating their activity. The development of drugs targeting specific proteins occupies a significant part of modern cancer treatment, and they are particularly widely used as targeted therapies. Targeted therapies work by suppressing cancer cells through the blocking of specific proteins or pathways that promote tumor growth. These treatments have the advantage of being relatively selective and having fewer side effects compared to conventional chemotherapy, as they have less impact on normal cells. However, this treatment method also has serious limitations, one of which is drug resistance caused by protein mutations. In particular, secondary mutations occurring in various cancers act as a major obstacle for targeted therapies. Due to genetic instability, cancer cells can naturally undergo mutations over time, which causes drugs to no longer function effectively. For example, while cancer cells are initially suppressed by targeted therapies against specific proteins, if these proteins mutate, the binding of the drug weakens or is blocked, eventually leading to a decrease or disappearance of the therapeutic effect. The occurrence of such resistance mutations poses a significant challenge to maintaining the efficacy of anticancer drug treatment in the long term. The development of drug resistance is a frequently observed phenomenon not only in anticancer drugs but also in antibiotic resistance in microorganisms, and the mechanisms of its expression have been elucidated through numerous studies. Various major mechanisms of drug resistance development exist, including changes in drug targets, alterations in drug uptake and efflux, gene amplification, and intracellular pH regulation. Among these, the onset of resistance due to changes in the drug target refers to a case where the drug can no longer effectively act on the protein due to structural modifications in the protein it targets. The development of tolerance due to changes in drug absorption and excretion means cases where the drug's efficacy is not properly manifested due to changes in the amount absorbed when the drug is administered orally, intravenously, intramuscularly, or subcutaneously, or cases where the drug's efficacy is not properly manifested due to breakdown of the drug by the liver, or cases where the drug's efficacy is not properly manifested due to changes in substances in the plasma. Resistance development caused by gene amplification refers to resistance that occurs when the expression of a drug-targeted protein increases excessively, preventing changes in cell behavior even when the drug inhibits the protein's activity. The expression of resistance due to intracellular pH regulation refers to a situation where changes in the concentration of reactive oxygen species within a cell alter the intracellular pH, and the drug's structure is modified by this altered pH, resulting in a failure to exert its proper efficacy. In particular, cancer cells that proliferate rapidly can accumulate mutations, which can lead to secondary mutations in proteins caused by drugs and result in resistance to the drugs. Therefore, among these various mechanisms of resistance expression, there is a continuous need to develop methods to predict the occurrence of drug resistance by predicting changes in the target protein—specifically, structural modifications of the protein targeted by the drug that prevent the drug from acting effectively on the protein. The background description of the invention is provided to facilitate a better understanding of the present invention. The matters described in the background description should not be construed as an acknowledgment that they exist as prior art. FIG. 1 illustrates an exemplary device-based system for predicting drug resistance according to one embodiment of the present invention. FIG. 2a is a block diagram showing the configuration of a user device according to one embodiment of the present invention. FIG. 2b is a block diagram showing the configuration of a server for a drug resistance prediction device according to one embodiment of the present invention. FIG. 3 illustrates the procedure of a drug resistance prediction method according to one embodiment of the present invention. Figure 4 illustrates the results of determining the binding strength and major binding sites between a drug and a protein in a drug resistance prediction method according to one embodiment of the present invention. Figures 5a and 5b illustrate the results of predicting the mutation potential of amino acids in a drug resistance prediction method according to one embodiment of the present invention. Figure 6 illustrates the results of