mRNA 上的多聚腺苷酸化在人類基因中是一項重要的機制，並且直接關係到轉錄的終結。而人類基因上的多聚腺苷酸化具有多重的作用點，選擇不同的作用點將會改變 3’UTR 的長度。mRNA 的 3’UTR 中包含了許多調控因子，因此多聚腺苷酸化作用點的選擇就非常重要。正確的預測多聚腺苷酸化作用點能夠幫助了解作用點選擇的機制。本研究的目的是要利用核苷酸分佈分析，設計一個預測人類基因中 mRNA 多聚腺苷酸化作用點的方法。此分析利用了新的定序技術PET(Paired-End diTag)，PET 能準確的對應到多聚腺苷酸尾前端，有助於多聚腺苷酸化作用點的預測。此外本研究還個別分析了包含基因中唯一作用點的序列與包含基因中有複數作用點的序列。經過分析發現兩種序列在核苷酸分佈機率上有一點差異。mRNA polyadenylation is an essential mechanism in human genes, it is direct linked to termination of transcription. And alternative polyadenylation changes the length of mature mRNAs 3’UTRs. Since 3’UTRs have been shown to contain regulatory elements controlling mRNA, alternative polyadenylation plays an important role in human genes. Prediction of polyadenylation sites can help identify genes and understand mechanism of alternative polyadenylation. In this study, we constructed a system of mRNA polyadenylation sites prediction in human genes using SVM, based on analysis of sequences align between pair-end diTags (PET) and genome sequences. The PET sequences can mapped to the reference genome more accurate compare with past methodology. We also analyzed single-site type and multiple-site type sequences PET sequences dataset. We found that the frequencies of each nucleotide between single-site type and multiple-site type were different.
