PREDICTION OF THE TERTIARY STRUCTURE OF THE PROTEOME OF THE ROCIO AND ILHÉUS FLAVIVIRUSES

The Rocio (ROCV) and Ilhéus (ILHV) viruses are classified as emerging flaviviruses with potential epidemiological impact in Neotropical regions; however, their proteomic structures remain poorly characterized. This study aimed to perform a comprehensive structural characterization of the viral proteomes of these species using amino acid sequences in FASTA format, leveraging AlphaFoldDB and ColabFold predictions with GPU acceleration. The approach sought to provide support for the development of antivirals and a better understanding of structural evolution within the Flavivirus genus. The generated models exhibited high structural homology with well-characterized flaviviruses such as Dengue and Zika viruses. The Predicted Local Distance Difference Test (pLDDT) scores were predominantly above the 70–80 range, with standard deviations below 10, indicating reliable prediction quality for most residues within the proteins. For both ROCV and ILHV, the E, NS1, NS3, and NS5 proteins stood out, with pLDDT average scores above 90, highlighting strong confidence in the model. The Predicted Aligned Error (PAE) further supported the reliability of these models, revealing high confidence in the folding and relative positioning of structural domains. Specifically, ROCV’s E, C, NS1, NS3, NS4B, and NS5 proteins, and ILHV’s M, NS1, NS3, NS4B, and NS5 proteins presented average PAE values below 10 Å. Further work will include homology searches using Foldseek against flavivirus-specific databases, analysis of conformational variability and structural domains with tools such as FoldMason and FoldTree, and the construction of phylogenetic trees integrating both sequence and structural alignment data, allowing a more comprehensive view of the structural evolution of these emerging viruses.

This work was supported by the Coordination for the Improvement of Higher Education Personnel (CAPES 88887.173677/2025-00) and FAPESP (2022/00347-0).