Collaborative effects of 2019-nCoV-Spike mutants on viral infectivity.

Background: The emerging mutants of the 2019-nCoV coronavirus are posing unprecedented challenges to the pandemic prevention. A thorough, understanding of the mutational characterization responsible for the pathogenic mechanisms of mutations in 2019-nCoV-Spike is indispensable for developing effective drugs and new vaccines. Methods: We employed computational methods and viral infection assays to examine the interaction pattern and binding affinity between ACE2 and both single- and multi-mutants of the Spike proteins. Results: Using data from the CNCB-NGDC databank and analysis of the 2019-nCoV-Spike/ACE2 interface crystal structure, we identified 31 amino acids that may significantly contribute to viral infectivity. Subsequently, we performed molecular dynamics simulations for 589 single-mutants that emerged from the nonsynonymous substitutions of the aforementioned 31 residues. Ultimately, we discovered 8 single-mutants that exhibited significantly higher binding affinities (<-65.00 kcal/mol) to ACE2 compared with the wild-type Spike protein (-55.07 kcal/mol). The random combination of these 8 single-mutants yielded 184 multi-mutants, of which 60 multi-mutants exhibit markedly enhanced binding affinities (<-65.00 kcal/mol). Moreover, the binding free energy analyses of all 773 mutants (including 589 single- and 184 multi-mutants) revealed that Y449R and S494R had a synergistic effect on the binding affinity with other mutants, which were confirmed by virus infection assays of six randomly selected multi-mutants. More importantly, the findings of virus infection assay further validated a strong association between the binding free energy of Spike/ACE2 complex and the viral infectivity. Conclusions: These findings will greatly contribute to the future surveillance of viruses and rational design of therapeutics.

A review of enzyme design in catalytic stability by artificial intelligence.

The design of enzyme catalytic stability is of great significance in medicine and industry. However, traditional methods are time-consuming and costly. Hence, a growing number of complementary computational tools have been developed, e.g. ESMFold, AlphaFold2, Rosetta, RosettaFold, FireProt, ProteinMPNN. They are proposed for algorithm-driven and data-driven enzyme design through artificial intelligence (AI) algorithms including natural language processing, machine learning, deep learning, variational autoencoder/generative adversarial network, message passing neural network (MPNN). In addition, the challenges of design of enzyme catalytic stability include insufficient structured data, large sequence search space, inaccurate quantitative prediction, low efficiency in experimental validation and a cumbersome design process. The first principle of the enzyme catalytic stability design is to treat amino acids as the basic element. By designing the sequence of an enzyme, the flexibility and stability of the structure are adjusted, thus controlling the catalytic stability of the enzyme in a specific industrial environment or in an organism. Common indicators of design goals include the change in denaturation energy (ΔΔG), melting temperature (ΔTm), optimal temperature (Topt), optimal pH (pHopt), etc. In this review, we summarized and evaluated the enzyme design in catalytic stability by AI in terms of mechanism, strategy, data, labeling, coding, prediction, testing, unit, integration and prospect.