Highly expressed proteins prefer less informative amino acids

Abstract

The transcriptional and translational systems are essentially information processing systems. However, how to quantify the amount of information decoded during expression remains a mystery. Here, we have proposed a simple method to evaluate the amount of information transcribed and translated during gene expression. We found that although proteins with a high copy number have more information translated, the average number of bits per amino acid is not high. The negative correlation between protein copy number and bits per amino acid indicates the selective pressure to reduce translational errors. Moreover, interacting proteins have similar bits per residue translated. All of these findings highlight the importance of understanding transcription and translation from an information processing perspective.