Prediction Of Alternative Splice Site And Analysis Of Sequence Characteristic In Model Organism Genome

Alternative processing of mRNA is a basic distinction between enkaryotes and prokaryotes, which is a key mechanism enriching proteomic diversity and functional complexity of higher multicellular eukaryotes by producing several transcripts from single gene. Alternative splicing of pre-mRNA often is specific to different stages of development and particular tissues of organism. On the other hand, it plays an important role in development, differentiation and cancer of system. Some basic features of cassette exon and intron retention in human genome were analysed, and based on the conservation of nucleotides, di-nucleotides, and tri-nucleotides at splice sites, the method of increment of diversity with quadratic discriminant analysis was used to predict the donor and acceptor splice sites of cassette exon and intron retention in human genome. The test results show that prediction accuracy of alternative donor and acceptor splice sites are higher than 93% and 84% for cassette exon and higher than 89% and 81% for intron retention.More than 90% mouse and human genomes is homologous. The mouse genome is a crucial experimental data for understanding the contents of the human genome. In article, some conserved features of splice sites of cassette exon and intron retention were analysed in mouse genome. Based on these conserved features, the method of increment of diversity with SVM and quadratic discriminant analysis was used to predict the donor and acceptor splice sites of cassette exon and intron retention in mouse genome. The test-independent results indicate that the total accuracy and F-score of donor splice site are 96.17% and 0.7055 respectively for cassette exon. However, acceptor splice site's prediction accuracy is slightly lower than donor splice site. Total accuracy is higher 88%, and F-score is 0.3815. For intron retention, the total accuracy of donor splice site have attained 88%, in while F-score is 0.4700. For acceptor splice site, total accuracy is higher than 88% and F-score is 0.3588. The prediction accuracy of alternative donor and acceptor splice sites by IDQD are close to SVM. The results indicate that IDQD is one of methods predicting problem in bioinformatics.