Table 2.
Computational characterization of highly prevalent SARS‐CoV‐2 mutations, exclusive to the Delta variant.
| Mutation | Secondary structure | Domain/Site | ConSurf grade | No. of protein homologs | Overall predicted change in protein function |
|---|---|---|---|---|---|
| Spike T19R | Loop | N‐terminal domain (Data ref: Spike glycoprotein, 2020) | a | 150 (coronaviruses) | Altered antibody interactions (Data ref: Cerutti et al, 2020) |
| Spike L452R | Strand | Receptor‐binding domain (Data ref: Spike glycoprotein, 2020) | 1 | Potentially increases binding to the ACE2 receptor | |
| Spike T478K | Strand | 1 | |||
| Spike P681R | Loop | Proximal to furin cleavage site (Data ref: Spike glycoprotein, 2020) | 1 | Altered cleavage by host furin (Hoffmann et al, 2020) | |
| Nucleocapsid R203 M | Loop | Proximal to phosphorylation site (SR‐rich domain) (Tung & Limtung, 2020; preprint: Yaron et al, 2020) | 9 | 139 (coronaviruses) | Increased spread of the virus (Syed et al, 2021) and altered interaction with the human 14‐3‐3 protein (Del Veliz et al, 2021) leading to changes in subcellular localization (Surjit et al, 2005) |
| Nucleocapsid D377Y | Loop | – | 1 | Functional impact of the mutation is unclear | |
| Membrane I82T | Helix | Transmembrane domain (Data ref: Membrane protein, 2020) | 7 | 92 (coronaviruses) | Altered glucose binding and uptake |
| NS3 S26L | Helix | Proximal to viroporin transmembrane domain (Data ref: ORF3a protein, 2020) | a | 135 (coronaviruses) | Altered ion channel activity leading to change in NLRP3 inflammasome activation (key component of host antiviral response) (Chen et al, 2019) |
| NS7a V82A | Loop | – | a | 150 (coronaviruses) | Functional impact of the mutation is unclear |
| NS7a T120I | Loop | Proximal to polyubiquitination site (Li et al, 2020a) | 1 | Altered IFN‐I response (Xia et al, 2020) |
The evolutionary conservation of the residues was analyzed using Consurf (Ashkenazy et al, 2016), and graded on a scale of 1 (variable) to 9 (conserved) by the program. Protein sequence homologs were retrieved using one iteration of HMMER (Eddy, 2011) (E‐value ≤ 0.0001) against the UniRef90 database (Suzek et al, 2015), and the multiple sequence alignment was built using MAFFT (Katoh et al, 2002).
a
Unreliable conservation score due to calculations performed on less than six non‐gapped homologous sequences.