Cluster analysis of genes in codon space

Organisms can essentially be classified according to two codon patterns

We recently classified 23 bacteria into two types based on their complete genomes; "S-type" as represented by Staphylococcus aureus and "E-type" as represented by Escherichia coli. Classification was characterized by concentrations of Arg, Ala or Lys in the amino acid composition calculated from the complete genome. Based on these previous classifications, not only prokaryotic but also eukaryotic genome structures were investigated by amino acid compositions and nucleotide contents. Organisms consisting of 112 bacteria, 15 archaea and 18 eukaryotes were classified into two major groups by cluster analysis using GC contents at the three codon positions calculated from complete genomes. The 145 organisms were classified into "AT-type" and "GC-type" represented by high A or T (low G or C) and high G or C (low A or T) contents, respectively, at every third codon position. Reciprocal changes between G or C and A or T contents at the third codon position occurred almost synchronously in every codon among the organisms. Correlations between amino acid concentrations (Ala, Ile and Lys) and the nucleotide contents at the codon position were obtained in both "AT-type" and "GC-type" organisms, but with different regression coefficients. In certain correlations of amino acid concentrations with GC contents, eukaryotes, archaea and bacteria showed different behaviors; thus these kingdoms evolved differently. All organisms are basically classifiable into two groups having characteristic codon patterns; organisms with low GC and high AT contents at the third codon position and their derivatives, and organisms with an inverse relationship.

Statistics of trinucleotides in coding sequences and evolution

The aim of this paper is to give measurements indicative of evolutional stages of the species. Two types of statistics of trinucleotides in coding regions are analysed for 27 species. The first one is the codon space, the nucleotide ratio for each of the three codon positions. We apply principal component analysis on this space and extract two principal components faithfully describing the original distribution of the codon space. The first principal component corresponds to the GC content. The second principal component classifies the species into three evolutional groups, Archaea, Bacteria and Eukaryota. The second statistics is the real and theoretical frequency of amino acids. The real frequency of an amino acid in a coding sequence is its frequency in the translated protein. The theoretical frequency is the expected frequency calculated from the ratio of nucleotides. We introduce the discrepancy between these two frequencies as an index of non-randomness of nucleotides in the sequence. This index of non-randomness divides the species into two groups: eukaryotes having smaller non-randomness (i.e. being more random) and prokaryotes having higher non-randomness.

Studies on the specificity of HIV protease: an application of Markov chain theory

A sequence-coupled (Markov chain) model is proposed to predict the cleavage sites in proteins by proteases with extended specificity subsites. In addition to the probability of an amino acid occurring at each of these subsites as observed from a training set of oligopeptides known cleavable by HIV protease, the conditional probabilities as reflected by the neighbor-coupled effect along the subsite sequence are also taken into account. These conditional probabilities are derived from an expanded training set consisting of sufficiently large peptide sequences generated by the Monte Carlo sampling process. Very high accuracy was obtained in predicting protein cleavage sites by both HIV-1 and HIV-2 proteases. The new method provides a rapid and accurate means for analyzing the specificity of HIV protease, and hence can be used to help find effective inhibitors of HIV protease as potential drugs against AIDS. The principle of this method can also be used to study the specificity of any multisubsite enzyme.

Diagrammatization of codon usage in 339 human immunodeficiency virus proteins and its biological implication

The occurrence frequencies of bases A (adenine), C (cytosine, G (guanine), and T (thymine) occurring in the 1st, 2nd, and 3rd codon positions in the codon usage table of viral genes for the 339 human immunodeficiency virus (HIV) proteins compiled recently have been calculated and diagrammatized. For comparison, the corresponding diagrammatic representations for the 2681 human proteins from the codon usage table for primate genes are also presented. The analyzed results based on these characteristic diagrams indicate that considerably similar features have been found between HIV and human proteins for the 1st and 2nd codon positions; i.e., they are all occupied predominantly by purine, especially base A. However, a significant difference in the 3rd codon position between HIV and human proteins has been observed; i.e., human proteins are of high C + G content and low A + G content in the 3rd codon position, whereas the case is just the opposite for HIV proteins. The biological implication of such a duality on the codon bias of HIV against human proteins is discussed. It is suggested that the 1st and 2nd codon positions can be termed as the structure-determining position, and the 3rd codon position termed as the species-determining position. The diagrammatic representation and analysis method described here possess a great potential for the study of molecular evolution from the viewpoint of the genetic code for which data have been accumulated rapidly and will continue to grow at a much faster pace.

A multivariate study of the relationship between the genetic code and the physical-chemical properties of amino acids

The 20 naturally occurring amino acids are characterized by 20 variables: pKNH2, pKCOOH, pI, molecular weight, substituent van der Waals volume, seven 1H and 13C nuclear magnetic resonance shift variables, and eight hydrophobicity-hydrophilicity scales. The 20-dimensional data set is reduced to a few new dimensions by principal components analysis. The three first principal components reveal relationships between the properties of the amino acids and the genetic code. Thus the amino acids coded for by adenosine (A), uracil (U), or cytosine (C) in their second codon position (corresponding to U, A, or G in the second anticodon position) are grouped in these components. No grouping was detected for the amino acids coded for by guanine (G) in the second codon position (corresponding to C in the second anticodon position). The results show that a relationship exists between the physical-chemical properties of the amino acids and which of the A (U), U (A), or C (G) nucleotide is used in the second codon (anticodon) position. The amino acids coded for by G (C) in the second codon (anticodon) position do not participate in this relationship.

A three-dimensional representation for base composition of protein-coding DNA sequences

Multi-dimensional scaling is applied to our codon space data on the protein coding sequences of DNA from a wide variety of organisms in an attempt to find the smallest number of parameters which will accurately represent these sequences. I find that a three-dimensional representation is satisfactory. One of the three resulting co-ordinates separates eukaryotes and their associated viruses from prokaryotes and their associated phages, while an orthogonal co-ordinate separates those organisms capable of synthesizing proteins (eukaryotes and prokaryotes) from those not so capable (viruses and phages). Mitochondria show no relation in our plots to any of these groups.