著者

Ryo Harada Keitaro Kume Kazumasa Horie Takuro Nakayama Yuji Inagaki Toshiyuki Amagasa

出版者

Information Processing Society of Japan

雑誌

IPSJ Transactions on Bioinformatics (ISSN:18826679)

巻号頁・発行日

vol.16, pp.20-27, 2023 (Released:2023-07-25)

参考文献数

48

被引用文献数

1

---

Eukaryotic genomes contain exons and introns, and it is necessary to accurately identify exon-intron boundaries, i.e., splice sites, to annotate genomes. To address this problem, many previous works have proposed annotation methods/tools based on RNA-seq evidence. Many recent works exploit neural networks (NNs) as their prediction models, but only a few can be used to generate new genome annotation in practice. In this study, we propose AtLASS, a fully automated method for predicting splice sites from genomic and RNA-seq data using attention-based Bi-LSTM (Bidirectional Long Short-Term Memory). We exploit two-stage training on RNA-seq data to address the problem of biased label problem, thereby reducing the false positives. The experiments on the genomes of three species show that the performance of the proposed method itself is comparable to that of existing methods, but we can achieve better performance by combining the outputs of the proposed method and the existing method. The proposed method is the first program specialized in end-to-end splice site prediction using NNs.

著者

Takeshi OHAMA Yuji INAGAKI Yoshitaka BESSHO Syozo OSAWA

出版者

The Japan Academy

雑誌

Proceedings of the Japan Academy, Series B (ISSN:03862208)

巻号頁・発行日

vol.84, no.2, pp.58-74, 2008-02-28 (Released:2008-02-12)

参考文献数

47

被引用文献数

17 23

---

In 1985, we reported that a bacterium, Mycoplasma capricolum, used a deviant genetic code, namely UGA, a “universal” stop codon, was read as tryptophan. This finding, together with the deviant nuclear genetic codes in not a few organisms and a number of mitochondria, shows that the genetic code is not universal, and is in a state of evolution. To account for the changes in codon meanings, we proposed the codon capture theory stating that all the code changes are non-disruptive without accompanied changes of amino acid sequences of proteins. Supporting evidence for the theory is presented in this review. A possible evolutionary process from the ancient to the present-day genetic code is also discussed.(Communicated by Takao SEKIYA, M.J.A.)