Predicting the evolutionary and structural compensation in Tat ARM region across HIV-1 groups using machine learning approach.

Trans-activator of transcription (Tat), a key regulatory protein of HIV-1, binds the TAR RNA element and is essential for early viral pathogenesis. Remarkably, Tat from HIV-1 groups M, N, O, and P shows notable sequence variability, especially in the critical Arginine-Rich Motifs (ARM) required for TAR binding, presenting both challenges and opportunities for therapeutic targeting. Computational analysis and machine learning (ML) approaches were utilized to analyze over 45,000 HIV-1 tat sequences, retrieved from the Los Alamos National Laboratory (LANL) database spanning 1997 to 2021. Host-mediated selection on tat gene ensued distinct group-specific pattern, reflected in non-synonymous vs. synonymous ratio (dn/ds) of 2.20, 2.04, 2.45, and 1.08 for groups M, N, O, and P, with corresponding entropy scores of 0.21, 0.35, 0.14, and 0.01, respectively. Groups M and N exhibited comparable selective pressure, while O and P evolved under distinct environments. Remarkably, group-specific sequence variability was evident in TAR binding region, particularly within the ARM, likely influencing viral pathogenicity. Docking analysis of wild-type Tat-TAR complexes supported this, showing binding affinities (docking scores) of − 313, − 242, − 282, and − 239, for groups M, N, O, and P, respectively. Phylogenetic Dependency Network (PDN) analysis revealed 1375, 36, 135, and 6 co-evolutionary associations within Tat for these groups, including 183, 1, 12, and 0 from the ARM motif. Structural modeling combined with PDN analysis identified 355 associated mutations in group M conferring structural compensation, of which 40 were from the ARM region. By leveraging these structural and co-evolutionary insights, we developed an ML model to classify compensatory mutations. HIV-1 Tat evolves differentially under host-mediated selective pressures, with group M exhibiting the highest adaptability through coevolutionary compensation, reflecting its global infection dynamics. Targeting conserved regions and compensatory mutational pathways in Tat may yield novel strategies for HIV-1 therapeutics.