Location:Home >> Research >> Research Progress

Researcher in CNIC Makes Progress in Machine Learning-Assisted Quantum Chemical Calculations with Optimized Load-Balancing

Date: Mar 22, 2023
Researchers in CNIC proposed a machine learning-assisted load balancing calculation scheme, which was applied to general quantum chemical calculations. The test results show that, for a computing scale of one hundred of nodes, this scheme has an efficiency improvement of more than 50% compared with the default scheduling and queuing scheme (Figure 1). Combined with this scheme, we have carried out preliminary quantum chemical calculations of the interactions between the important amino acid residues in the SARS-CoV-2 (Wild-type and its Omicron variant) and human angiotensin-converting enzyme 2 (hACE2). The calculated interaction energies can be used to analyze the change in the affinity energy of SARS-CoV-2 (Wild-type) and its Omicron variant  when binding to human proteins (Figure 2). 

Figure 1:Machine Learning-Assisted Quantum Chemical Calculations with Optimized Load-Balancing


Figure 2:Important residuals in SARS-CoV-2(upper), binding energies calculations(bottom

The research results have been published by the Journal of Computational Chemistry (SCI, JCR Q2/CAS Q3), an international journal in the field of computational chemistry. The first author of the paper is Associate Researcher Ma Yingjin from the High Performance Computing Department of the Center, and the cooperative units include Shenzhen Bay Laboratory, Institute of Chemistry, Chinese Academy of Sciences, Wenzhou University, and Northwest University; the corresponding authors of the paper are Associate Professor Ma Yingjin and Professor Suo Bingbing (Northwest University), and Professor Jin Zhong. The optimization method proposed in this study has been granted an invention patent (Patent No. 202010403157.4) at the same time. The research work has been supported by the National Key Research and Development Program, the National Natural Science Foundation of China, Youth Innovation Promotion Association of the Chinese Academy of Sciences, and the Informatization Plan of Chinese Academy of Sciences.

For details, please contact Ma Yingjin  (yingjin.ma@cnic.cn)



Yingjin Ma*, Zhiying Li, Xin Chen, Bowen Ding, Teng Lu, Baohua Zhang, Bingbing Suo*, Zhong Jin*,

Machine-Learning assisted Scheduling Optimization and Its Application in Quantum Chemical Calculations,

Journal of Computational Chemistry 2023, 1. https://doi.org/10.1002/jcc.27075