Vaccine rollouts have started in many countries around the world to fight against the ongoing COVID19 pandemic. The COVID19 infection is caused by the severe acute respiratory syndrome coronavirus-2 (SARS-CoV-2 virus). However, new emerging strains of the virus have been undermining the efforts of the governments to eliminate the virus. Experts have said that this might be due to their partial resistance to antibodies created by the vaccines or increased ability of transmission. Now, a team of experts has discovered a computational tool, which can produce all possible single amino acid alternates in the SARS-CoV-2 virus and can easily predict their effects. The new study has been done by experts from the UK. Experts have compared their outcomes with probable findings that have been based on the observed strain frequency and past studies to find out more common variants of the virus, which have clinically large effects. They have been able to identify variants, which might affect the efficacy of the antibodies. However, these strains might not be able to impact other aspects of viral biology. The findings of the new study have been released as a preprint on the BioRxiv server.
The variants, which have been identified in the UK, South Africa, and Brazil, are among the strains, which have become a huge concern around the world. Experts have noted that the majority of common mutations are alternates of one nucleotide for another. These mutations affect both viral RNA formation and function and protein sequence as well. A new study has focused on the protein sequence of the variants. They have found that alterations in protein sequence are able to change the basic structure, stability, and action of the protein. Therefore, it affects its biological function. Experts have said that the E484K and N439K mutations affect antibody binding and affinity for the ACE2 receptor of the cell for the virus in the human body. The new computational tool assesses the potential effects of these mutations with the help of protein sequence and structure. If a strain is found at the same positions in different species, it means it is going to become a fixed mutation. It is the basis of tools such as SIFT4G and EVCouplings. On the other hand, the predictions of FoldX and Rosetta are based on protein structure. These tools replicate the changes, which are caused by the alternates of the energetics of protein. However, experts have stopped using these predictive tools due to some technical issues and high costs. The basis of the new study is the Mutfunc web service, which has been developed to offer an edge where predictions for all variants of human beings, the housefly, and baker’s yeast genomes have been calculated.
The new tool has been used to observe the SARS-CoV-2 proteins along with a blend of factors, which can affect the predicted effects. These factors consist of the preservation of sequences across variants, the shape of the protein, known protein-to-protein contact, phosphorylation sites, and variant frequencies. The new study tests the validity of the predictions made with the help of this tool to identify the variants of concern, their effects on the emergence of new strains, and antibody-triggered elimination of the virus. Experts have compared the predictions done by SIFT4G for variants, which are known to persist at high frequencies. They have seen that scores have been low for uncommon variants. In FoldX predictions, the scores have been low for more frequently found variants. With the help of Deep Mutational Scanning (DMS), experts have found that the SARS-CoV-2 spike variants have shown that harmful mutations contain lower viral fitness and size.