Improving de novo metatranscriptome assembly via machine learning algorithms Online publication date: Tue, 25-Apr-2017
by Hussein Mohsen; Haixu Tang; Yuzhen Ye
International Journal of Computational Biology and Drug Design (IJCBDD), Vol. 10, No. 2, 2017
Abstract: In this paper, we present DNPipe, a Pipeline that processes and filters the transcript contigs reported by existing De Novo metatranscriptome assembly algorithms, aiming to improve the quality of de novo assembly of metatranscriptomic sequences. DNPipe consists of expectation-maximisation (EM) and sampling approaches that utilise abundance information of transcript contigs. We tested DNPipe on six metatranscriptomic datasets acquired from a mock microbial community dominated by three (among more than 15) known bacterial genomes. Results show that DNPipe can substantially improve the quality of metatranscriptome assembly, producing longer and more accurate transcripts. The N50 of the contigs increases by 19% (from around 1880 bps to 2250 bps), and the precision of the assembly improves by up to 8.7%, achieving about 81%. DNPipe assemblies are of higher quality than those assembled by Trinity as well. The DNPipe tool can be downloaded as open source software at https://sourceforge.net/projects/dnpipe/files/.
Existing subscribers:
Go to Inderscience Online Journals to access the Full Text of this article.
If you are not a subscriber and you just want to read the full contents of this article, buy online access here.Complimentary Subscribers, Editors or Members of the Editorial Board of the International Journal of Computational Biology and Drug Design (IJCBDD):
Login with your Inderscience username and password:
Want to subscribe?
A subscription gives you complete access to all articles in the current issue, as well as to all articles in the previous three years (where applicable). See our Orders page to subscribe.
If you still need assistance, please email subs@inderscience.com
Our Blog 
Follow us on Twitter 
Visit us on Facebook