Simulating protein and nucleic acid sequence co‐evolution

Protein structures and interactions can be predicted by identifying co‐evolution of residues, both within the same protein or across different proteins and nucleic acids. While conserved residues might represent invariant features of a protein's structure, co‐evolving residue pairs can suggest a sequence‐specific and consequential interaction. Different computational approaches have been used to detect co‐evolution in sequence alignments, but biological data is often used as a benchmark. Such evaluation is important due in part to the confounding presence of indirect co‐evolution. However, using biological data as a benchmark is problematic, because not all the interactions and evolutionary constraints can be known a priori . Instead, the ideal benchmark would be a set of sequences generated with constraints such that they simulate co‐evolution. This was achieved using new cross‐platform, user friendly software tools, which were then used to benchmark and evaluate existing detection methods. This new software can generate sequences in which intramolecular, intermolecular, and indirect co‐evolution can be simulated. Systematic tuning of these evolutionary constraints shed new light on how they drive co‐evolution between residues. Better understanding how to detect co‐evolution and the residue interactions they predict can lead to a wide range of insights important for synthetic biologists interested in engineering new, orthogonal interactions between two proteins or nucleic acids. This abstract is from the Experimental Biology 2019 Meeting. There is no full text article associated with this abstract published in The FASEB Journal .