El Asely A.M., Khalifa M.S., Xu W., Shaheen A.A., et al.
Microorganisms· 2026-02· Michigan State University
The research
Tilapia lake virus (TiLV) has emerged as a serious threat to Nile tilapia aquaculture worldwide since its first description in 2006, and understanding its genetic diversity is essential for disease surveillance and control.
Researchers at Michigan State University and Benha University set out to characterize two newly sequenced variants of a divergent amnoonvirus — AmnoonvirusEGY1F and AmnoonvirusEGY1H — detected in farmed Nile tilapia in Egypt. Using Illumina sequencing, the team assembled full-length segment 2 sequences from both variants, which encode the PB2 subunit of the viral RNA-dependent RNA polymerase (RdRp), an enzyme central to viral replication and a key driver of pathogenicity and host immune evasion.
A core question of the study was whether the amino acid substitutions distinguishing the Egyptian variants from the TiLV index strain altered the 3D structure, function, or antigenicity of the PB2 protein. Answering that required reliable computational models as a starting point.
ProteinIQ in action
Before any mutation analysis could begin, the team had to confirm that the predicted protein structures were physically sound. The researchers generated 3D models of all three PB2 proteins using SWISS-MODEL, then faced a problem common in structural bioinformatics: computationally predicted models can contain geometric errors that look fine in a viewer but invalidate downstream comparisons. Structural differences between strains could easily reflect modeling artifacts rather than true biological divergence. They ran all three models through MolProbity to rule that out.
MolProbity assessed each model across three dimensions: the clash score (steric overlaps between atoms), Ramachandran statistics (backbone dihedral angle distributions), and rotamer outlier counts (correctness of sidechain conformations). These three metrics feed into a single MolProbity score that gives an overall measure of stereochemical quality comparable to experimental crystallography benchmarks. As the authors describe in their methods:
"This valuable tool collates multifaceted computational analyses to check for the overall geometric correctness of the predicted 3D protein structure and combine them into one value: the MolProbity score."
Once the models passed validation, the team used MolProbity's per-residue bond angle output to locate exactly where the Egyptian variants deviated structurally from TiLV. This granular report turned a pass/fail validation step into a map of structurally distinct sites across the three strains.
Key findings
All three predicted structures received MolProbity scores in the 1–2 range, reflecting the quality of high-resolution crystal structures: AmnoonvirusEGY1F scored 1.11, the TiLV index strain 1.12, and AmnoonvirusEGY1H 1.24. The bond angle analysis revealed meaningful differences at specific residues: deviations at six positions (C55–C57, C102–C103, C141–C142, C317–C318, C324–C325, and C331–C332) were present in the TiLV index strain but absent in both Egyptian variants, while deviations at C144 and C147 appeared only in the Egyptian strains. Sequence-level analysis identified 28 amino acid substitutions between the two Egyptian contigs, with two substitutions unique to AmnoonvirusEGY1H — A230E and A231T — predicted by SIFT and ConSurf to be functionally significant due to their location in a highly conserved region. FoldX energy calculations added further evidence, with predicted structure energies of 292.3 kcal/mol for TiLV, 282.89 for EGY1F, and 277.86 for EGY1H. The Egyptian variants are genetically and structurally distinct from known TiLV strains, a finding with direct implications for diagnostic assay design and disease notification decisions by animal health authorities.
