Correlations between the compositional properties of human genes, codon usage, and amino acid composition of proteins

Abstract

We have analyzed the correlation that exists between the GC levels of third and first or second codon position for about 1400 human coding sequences. The linear relationship that was found indicates that the large differences in GC level of third codon positions of human genes are paralleled by smaller differences in GC levels of first and second codon positions. Whereas third codon position differences correspond to very large differences in codon usage within the human genome, the first and second codon position differences correspond to smaller, yet very remarkable, differences in the amino acid composition of encoded proteins. Because GC levels of codon positions are linearly correlated with the GC levels of the isochores harboring the corresponding genes, both codon usage and amino acid composition are different for proteins encoded by genes located in isochores of different GC levels. Furthermore, we have also shown that a linear relationship with a unit slope and a correlation coefficient of 0.77 exists between GC levels of introns and exons from the 238 human genes currently available for this analysis. Introns are, however, about 5% lower in GC, on average, than exons from the same genes.

Codon Usage Patterns in Cytochrome Oxidase I Across Multiple Insect Orders

“…The pattern was first described by Grantham et al (1981) and has been ascribed to several potential causes. First, codon usage can be influenced by genome compositional constraints and mutational biases, as seen in mammalian (D'Onofrio et al 1991;Karlin and Mrazek 1996; but see Smith and Eyre-Walker 2001), protozoan (Musto et al 1999), and endosymbiotic bacterial (Wernegreen and Moran 1999) genomes. Second, selection among synonymous codons for either translational efficiency or translational accuracy can result in bias, as seen in bacterial, fungal, and insect genes (Ikemura 1982;Sharpe and Cowe 1991;Powell and Moriyama 1997).…”

Nucleotide Composition Bias Affects Amino Acid Content in Proteins Coded by Animal Mitochondria

“…A relationship between the nucleotide composition and amino acid content is known from several genomic sources, including bacterial DNA (Sueoka 1961;Andersson and Sharp 1996), viral RNA (Berkhout and van Hemert 1994), animal mitochondrial DNA (Jukes and Bhushan 1986;Jermiin et al 1994Jermiin et al , 1997, and eukaryotic nuclear DNA (D'Onofrio et al 1991;Collins and Jukes 1993;Porter 1995;Yoshida et al 1997). However, this relationship is not universal (Hashimoto et al 1994(Hashimoto et al , 1995.…”

Quantile distributions of amino acid usage in protein classes

“…McCaldon and Argos (1988) organized peptides ranging from 2 to 11 residues and projected certain preferences in protein sequences. Dcemura et al (1990) and D'Onofrio et al (1991) analyzed the aa composition of individual mammalian proteins under the isochore hypothesis (see also Aissani et al, 1991). All these comparative studies have centered on average residue usages of different protein collections.…”