NSMAP: A method for spliced isoforms identification and quantification from RNA-Seq

BMC Bioinformatics

Methodology Article

Jianguo Wen¹  Chung-Che Chang¹  Xiaobo Zhou²  Zheng Xia² 

[1] Department of Pathology, The Methodist Hospital Research Institute, 77030, Houston, TX, USA;Weill Cornell Medical College, 10065, New York, NY, USA;Department of Radiology, The Methodist Hospital Research Institute, 77030, Houston, TX, USA;Weill Cornell Medical College, 10065, New York, NY, USA;

关键词: Splice Junction;    Putative Exon;    Expression Level Estimation;    Fragment Length Distribution;    Annotate Reference Genome;   

DOI  :  10.1186/1471-2105-12-162

 received in 2010-10-12, accepted in 2011-05-16,  发布年份 2011

来源: Springer

【 摘 要 】

BackgroundThe development of techniques for sequencing the messenger RNA (RNA-Seq) enables it to study the biological mechanisms such as alternative splicing and gene expression regulation more deeply and accurately. Most existing methods employ RNA-Seq to quantify the expression levels of already annotated isoforms from the reference genome. However, the current reference genome is very incomplete due to the complexity of the transcriptome which hiders the comprehensive investigation of transcriptome using RNA-Seq. Novel study on isoform inference and estimation purely from RNA-Seq without annotation information is desirable.ResultsA N onnegativity and S parsity constrained M aximum AP osteriori (NSMAP) model has been proposed to estimate the expression levels of isoforms from RNA-Seq data without the annotation information. In contrast to previous methods, NSMAP performs identification of the structures of expressed isoforms and estimation of the expression levels of those expressed isoforms simultaneously, which enables better identification of isoforms. In the simulations parameterized by two real RNA-Seq data sets, more than 77% expressed isoforms are correctly identified and quantified. Then, we apply NSMAP on two RNA-Seq data sets of myelodysplastic syndromes (MDS) samples and one normal sample in order to identify differentially expressed known and novel isoforms in MDS disease.ConclusionsNSMAP provides a good strategy to identify and quantify novel isoforms without the knowledge of annotated reference genome which can further realize the potential of RNA-Seq technique in transcriptome analysis. NSMAP package is freely available at https://sites.google.com/site/nsmapforrnaseq.

【 授权许可】

Unknown   
© Xia et al; licensee BioMed Central Ltd. 2011. This article is published under license to BioMed Central Ltd. This is an Open Access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/2.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.