Bioinformatics Centre, School of Biotechnology, Madurai Kamaraj University, Madurai 625 021. India
A search for the correlation between synonymous codon usage and secondary structure (alpha helix, beta sheet and beta turn) in proteins shows that there is secondary structural bias in the synonymous codon usage for amino acids. 46 protein structures whose protein coding regions were available in the nucleic acid sequence database were used. This has been quantified by calculating the conformational parameters P(alpha), P(beta) and P(turn) for codons following the relations used for calculating the secondary structure propensities of amino acids. The non uniform distribution of the codons has been taken into consideration. Of the synonymous codons for Cys,Phe, Ile,Leu,Ser and His the codons TGC(beta), TTT(beta), TTC(alpha), ATC(beta),TTA(alpha), AGC(turn) and CAC(turn) respectively show significant bias towards the secondary structures indicated in the parantheses.
Within the overall framework of the accepted amino acid secondary structure propensities, the conformational preferences calculated at the codon level go towards accounting for the conformational ambiguities seen at the amino acid level. For example, Phe residues in the alpha helical structure are coded by TTC [P(alpha)=1.12, P(beta)=0.83 ] and those in beta sheet conformations are coded by TTT [P(alpha)=0.78 and P(beta)=1.77 ] while at the amino acid level Phe is considered to have similar propensities for forming alpha helix and beta sheet.