Ultra-large-scale marine natural product molecular docking method based on heterogeneous many-core architecture

Abstract

The invention relates to a super-large-scale marine natural product molecular docking method based on a heterogeneous many-core architecture, and belongs to the technical field of drug screening, and the method comprises the following steps: constructing a multi-layer parallel scheduling framework based on a heterogeneous supercomputing platform, and designing a molecular docking process of master-slave core asynchronous parallel computing, the invention provides a data access optimization scheme of a molecular docking key algorithm. According to the method, data reading, task scheduling and parallel computing in the molecular docking process are optimized from the aspect of nuclear memory access, the super-large-scale parallel drug virtual screening process is achieved, the computing speed is increased, and meanwhile the drug screening precision is improved. The I / O pressure of the system is reduced, the advantages of the heterogeneous many-core architecture are fully played, and the overall performance of the virtual drug screening system is improved.

Description

technical field [0001] The invention belongs to the technical field of drug screening, in particular to an ultra-large-scale marine natural product molecular docking method based on a heterogeneous many-core architecture. Background technique [0002] Compared with terrestrial sources, marine-derived natural products often have novel structures, unique biological activities or unique biological mechanisms of action, and are currently considered to be the most important source for the discovery of important drug leads and drugs with new biological mechanisms of action in the future. Two-thirds of the new drug entities announced globally in the last four decades are either derived directly from natural products or designed and synthesized derivatives inspired by natural products. [0003] Marine natural products are rich in variety, and have more complex and diverse, novel and peculiar structures and diverse biological activities and mechanisms far beyond the imagination of sc...

AI Technical Summary

Problems solved by technology

[0007] 1. The I / O of the existing molecular docking method usually reads in target files, small molecule files, and configuration files, and then performs docking calculations to generate a result file containing scoring. The degree of parallelism is not high, and the I / O overhead and system high storage overhead

[0008] 2. The existing molecular docking method usually performs the cluster analysis and energy sorting of the generated conformation after performing the conformation search process. The degree of parallelism is low, and the search process takes up a lot of memory, which is not suitable for calculating molecules with large molecular weights.

Images