Redundancy in the Genetic Code: Selection Analysis and its Implications for Reconstruction of Ancestral Protein Sequences

Abstract

Ancestral Sequence Reconstruction is a technique used to statistically infer the most likely ancestor of a set of evolutionarily related sequences, but research which relies solely on protein data has the disadvantage of sequence information being lost upon translation of a protein from its gene transcript, due to the redundancy inherent in the genetic code. In this project, the amino acid sequences, and separately the corresponding codon sequences, of 184 homologous Acetylcholine receptor protein sequences were aligned, and phylogenetic analysis and ancestral sequence reconstruction was performed based on both alignments to infer several ancestral sequences representing important milestones in the evolutionary history of the homologous protein family. To further extract meaningful information from the nucleotide sequences, positive selection analysis was performed on the codon alignment using the Mixed Effects Model of Evolution method, which estimates and compares between the rates of synonymous and non-synonymous mutations across the alignment to detect the occurrence of positive selection events throughout their evolution. The Mixed Effects Model of Evolution can infer positive selection across both sites and evolutionary branches in a sequence alignment, thus highlighting residues along the evolutionary trajectory of the proteins which may have been functionally important in their evolution. Positive selection analysis detected positive selection at a multitude of sites and branches, and by mapping signatures at which selection is strongest with changes in the trajectory of ancestral states, several important sites were chosen as likely to be most valuable for future experimental testing. The implications of this study on the benefits of conducting ancestral sequence reconstruction with protein and codon sequences are discussed.