| The discovery of the central dogma has made people clearly understand the composition and transmission of genetic information,and a series of scientific researches around it explained the mysteries of life more clearly one after another at the molecular level.The codon-level bioinformatics analysis plays an important role in the study of genetic information transmission and studies of nucleotides usage in each phase of codon can facilitate our understanding of the origin and evolution of the genetic code,understand the mechanism of related diseases and guide protein design and directed mutagenesis.The most widely studied position is the third position of the codon,while the second position of the codon has been less studied.In this dissertation,focusing on the second nucleotide of codon,the clustering patterns of genes,protein function and translation process,the following researches are carried out.The first study is about clustering patterns of genes and functional preference distribution based on the second nucleotide of codon.By a systematic analysis of the nucleotide frequencies at the second codon position of 3799 genomes across the three domains of life,there was found that the protein coding genes of prokaryotes can be divided into two unequal clusters based on the nucleotide A and T relative frequencies at the second codon position.This phenomenon is almost universal and is especially remarkable for prokaryotes,97.47%of the prokaryotic genomes could be optimally divided into two clusters.Further analysis showed significant difference in the functional distributions of genes between the two unequal clusters by chi-squared test.Finally,at the species level,phylum level and prokaryote level,a consistent preference of functional category was observed,P-related genes were more prevalent in the smaller cluster,while J-,K-and L-related genes were more likely to be included in the larger cluster.Secondly,this dissertation revealed a complete link from codon identity（A2 versus T2）to amino acid property（hydrophilic versus hydrophobic）and then to protein functions（informational versus operational）.There was proposed a quantitative measurement of codon-amino acid association and used it to explore 13 physicochemical properties of amino acids.Consequently,seven properties have a higher correlation with the middle position than the other two sites,indicating its major role in determining proteins’functions.At the protein level,the correlation between the frequency of A2 relative to T2 and the hydrophobicity score becomes stronger than the single amino acid level.All 3799involved genomes of three domains have regression coefficient R higher than 0.73,indicating the universal appearance of such association.Finally,it was observed that informational function holding proteins have lower GRAVY values than operational proteins and this difference was elucidated from the macroscopic biological evolutionary level and the microscopic subcellular localization level.Then,this dissertation analyzed the connection between the codon phase nucleotides and 553 groups AAindex in the AAindex database,based on the previously proposed quantitative measurement of codon-amino acid association.It was observed that the second phase of codons generally had much significantly stronger correlation coefficients than the first and third phases.Further statistical analysis of the strongest correlation coefficient of each phase,it is concluded that the highest average correlation coefficient and the highest single combination are in the second phase of codons.Then,the strongest correlation of the second phase of the codon was analyzed,and it was concluded that there was a strong correlation between the hydrophobicity properties of the amino acid and the nucleotide in the second phase of the codon,which proved once again that the nucleotide at the second position of the codon had the strongest influence on the hydrophilicity or hydrophobicity of amino acids.Finally,analyzing the relationship between the six type properties of amino acid and the functional classes of COG in E.coli.The last part is about the research between the second nucleotide of codon and translation.From the biogenesis process there was established an association between the various omics characteristics（proteome abundance,mRNA abundance,gene fitness level,RPF,mRNA translation efficiency）and the second nucleotides A and T of the codon and proved that nucleotide A at the second codon does have a significant effect on the translation efficiency.Further analysis proved that the translation efficiency of the amino acids encoded by the second nucleotide A of the codon is faster than that the other amino acids encoded by the second nucleotide T,G,C of the codon at the three levels of prokaryotic genome,gene and codon.Finally,linear fitting was performed on the translation efficiency and amino acid composition of three model organisms:E.coli,S.cerevisiae and Homo sapiens. |
