Divergence Of Duplicatecd Yeast Genes In The Molecular Ineraction Network And Construction Of Interaction Network

Gene duplication has long been thought to be a powerful source of functional innovation and genome complexity. Gene duplication especially whole-genome duplication has been found as a pivotal factor in the evolution. The main contribution of gene duplication is providing materials for mutation, transition and selection, and furthermore, a probability for genetic evolution. Recent years, with the advent of high-throughput technology, gene duplication was found to take place generally in the evolution of many species. The generation mechanism and evolution of gene duplication are more and more concerned.Researchers postulated some models to delineate the process of gene duplication. Firstly, a few neutral or nearly neutral mutations occur after gene duplication. Then these mutations create new but weakly active functions in one daughter gene. Subsequently, positive selection tends to fix the advantageous mutations to enhance the newly emerged functions. Lots of researches support the model regarding single gene duplication, however, the divergence pattern of duplicates on the whole genome scale remains elusive in current perspectives. After gene duplication, the evolution of duplicate genes can be influenced by many factors, such as the position of protein in the interaction network, the fitness of gene, expression level, and"essentialness". Furthermore, the function of gene can only be brought into effect by interacting with other partners. Does the structure of network influence the divergence pattern of duplicate genes on the genome scale?To investigate the regulatory mechanism, we analyzed the divergence of duplicate genes in the interaction network after the whole-genome duplication. All the duplicate genes used in this study arisen from one whole-genome duplication, indicating that these pairs are of the same age. We also used the most recently updated dataset of the protein and genetic interaction and increased its credibility. Incorporating highly accurate interaction dataset and duplicates data with rigidly validated time of duplication occurrence, we analyzed the influence of network structure on the divergence of duplicate genes. Our main conclusions are as follows:1. Our results showed that the difference of connectivity level between two duplicate copies followed the power law, and the difference had significant correlation with the connectivity level of ancestral gene.2. Having analyzed the evolution process, we found that there was a significantly negative correlation between the connectivity level of ancestral genes and evolutionary distance of duplicated genes. This suggests a relatively stable interaction network core whereas an increasingly changing edge.3. We also found that the divergence pattern of duplicate genes follows the power law, however, cannot be simply reflected on their evolutionary distance.4. Moreover, according to the classification of the duplicate genes with their divergence rate, we found that the evolutionary distance of completely divergent genes was significantly longer than that of incompletely divergent pairs. This suggests that duplicates tend to provide functional redundancy during early stage of interactions turnover; whereas duplicates become beneficial to the emergence of functional novelty by increasing the number of mutations when interactions vary greatly.5. Through functional analysis of essential genes on the scale of duplicated genes, we found that the majority of duplicated genes corresponding to the essential gene are non-essential genes, and the connectivity level of essential genes in the network is significantly higher than that of other genes. Therefor, the genome denies functional redundancy and the network preserves key functional genes via maintaining high connectivity level of essential genes.The mechanisms mentioned above ensure yeast genome evolving rapidly under the premise of hereditary stability.By using the method of homologous alignment, we constructed the interaction network of chicken which contains 294 proteins and 1210 interactions. This work provided the reference for experimental analysis.