Search academic texts

Information × Registration Number 0224U033460, (0124U001308) , R & D reports Title Development of an integrated approach to high-throughput modeling of protein complexes, protein structure-function relationships, and prediction of the phenotypic effects of single amino acid variation based on the kinetic aspects of protein association in cellular environment popup.stage_title Розробити інтегрований підхід до високопродуктивного моделювання протеїнових комплексів, структурно-функціональних взаємозв’язків протеїнів, та прогнозування фенотипних ефектів варіації однієї амінокислоти на основі кінетичних аспектів поєднання протеїнів у клітинному середовищі Head Kasianov Pavlo O., д.ф.-м.н. Registration Date 27-12-2024 Organization Educational and Scientific Complex "Institute for Applied System Analysis" of National Technical University of Ukraine "Igor Sikorsky Kyiv Polytechnic Institute" popup.description1 To apply high performance computing and machine learning to the development of an integrated approach to high-throughput modeling of macromolecular interactions for the prediction of kinetic aspects and assembly patterns of molecular mechaniscs in cellular environment. popup.description2  Research object -- models of protein complexes. The aim of the work is to apply high-performance computing and machine learning to develop an integrated approach for modeling macromolecular interactions to predict kinetic aspects and molecular mechanics models in a cellular environment. Research method -- deep learning, artificial intelligence methods. During the study, an examination was conducted in the field of high-performance computing and machine learning, particularly in their application to the development of an integrated approach for high-performance modeling of macromolecular interactions to predict kinetic aspects and molecular assembly schemes in a cellular environment. Innovative computational approaches have been developed for the structural characterization of protein complexes at realistic macromolecular concentrations, taking into account the latest advances in the application of artificial intelligence and deep learning for predicting the structure of proteins and protein assemblies. An idea for further optimization of computations has been presented. The results of the research obtained during the scientific work have been implemented in a number of training courses: "Reinforcement Learning." As part of the project, one PhD dissertation has been prepared for defense. One article is being prepared for publication in a journal indexed in the Scopus and WoS databases. The results obtained during the research correspond to the world level; they represent new powerful computational approaches that will enable the characterization of cellular mechanisms with atomic precision based on extraordinarily long simulation trajectories, significantly exceeding the capabilities of existing computational methods. Keywords: deep learning, molecular dynamics, process parallelization, machine learning. Product Description popup.authors Levenchuk Liudmyla B. Paliichuk Liliia S. Tytarenko Andrii M. popup.nrat_date 2024-12-27 Close