仅基于RNA-Seq数据拼装可变剪接转录组的计算方法研究

Alternative splicing is a process by which the exons of the RNA produced by transcription of a gene are reconnected in multiple ways during RNA splicing. The resulting different mRNAs may be translated into different protein isoforms. Alternative splicing occurs as a normal phenomenon in eukaryotes, where it greatly increases the biodiversity of proteins that can be encoded by the genome. Thus, it is an important mechanism for gene regulation expression. In humans, about 95% of multiexonic genes are alternatively spliced. Mechanisms of alternative splicing are highly variable, and new examples are constantly being found, particularly through the use of high-throughput techniques. Researchers hope to fully elucidate the regulatory systems involved in splicing, so that alternative splicing products from a given gene under particular conditions could be predicted by a splicing code. Abnormal variations in splicing are also implicated in disease; a large proportion of human genetic disorders result from splicing variants. Abnormal splicing variants are also thought to contribute to the development of cancer. High throughput cDNA sequencing technologies make it possible to predict spliced transcripts computationally. A couple of articles regarding computational prediction of spliced transcripts have already been published in Nature series, implying that this topic has become one of the most challenging problems. Very recently, we made a discovery that exons and their linear order encoded in a gene can be precisely predicted solely by assembling RNA-Seq data. This discovery implies that spliced transcripts can be computationally predicted without a reference genome. In this project we are going to develop a new diagram for de novo prediction of spliced transcripts.

可变剪接是指从一个前体mRNA中通过不同的剪接方式产生不同的成熟mRNA的过程。可变剪接是调控基因表达和产生蛋白质组多样性的重要机制。在人类基因组中，大约95% 的多外显子基因中存在可变剪接。基因的异常剪接与疾病有着密切的关系；人类相当一部分疾病包括癌症被认为起因于基因的可变剪接。高通量cDNA 测序技术使得可变剪接转录组的计算预测成为可能。近两年，NATURE系列期刊上连续刊出数篇有关基于RNA-Seq数据计算预测可变剪接转录组的科技文章和软件，使得可变剪接转录组的计算预测成为国际生物信息学研究领域最具挑战的研究课题之一。最近我们发现：基因的外显子以及它们在基因中的线性顺序完全可以通过拼装RNA-Seq数据预测出来。这就意味着可变剪接转录组的计算预测不需要参考基因组序列，我们将由此设计一个高效可靠的计算预测可变剪接转录组的算法和软件，使该问题的计算预测推向一个全新的高度。

随着生物测序数据潮水般的涌现，转录组重构的拼接算法研究是当今生物信息学领域最具挑战研究课题之一。项目团队在该领域开展了深入的研究，在转录组重构的拼接算法方面取得了实质性的进展，相继设计开发了三套转录组重构的算法和软件，其性能均比目前国际上最流行的同类算法提高了大约 20%，被数家学术媒体跟踪报道。其中两个软件Bridger和TransComb的文章发在Genome Biology（IF：11.313；生物类一区），另一个软件BinPacker的文章发在Plos Computational Biology（IF：4.587；生物计算类一区）。