RNA-Seq: Full-length transcriptome assembly from RNA-Seq data without a reference gen

Newsbot! · 05-17-2011, 03:10 AM

Syndicated from PubMed RSS Feeds

Full-length transcriptome assembly from RNA-Seq data without a reference genome.

Nat Biotechnol. 2011 May 15;

Authors: Grabherr MG, Haas BJ, Yassour M, Levin JZ, Thompson DA, Amit I, Adiconis X, Fan L, Raychowdhury R, Zeng Q, Chen Z, Mauceli E, Hacohen N, Gnirke A, Rhind N, di Palma F, Birren BW, Nusbaum C, Lindblad-Toh K, Friedman N, Regev A

Massively parallel sequencing of cDNA has enabled deep and efficient probing of transcriptomes. Current approaches for transcript reconstruction from such data often rely on aligning reads to a reference genome, and are thus unsuitable for samples with a partial or missing reference genome. Here we present the Trinity method for de novo assembly of full-length transcripts and evaluate it on samples from fission yeast, mouse and whitefly, whose reference genome is not yet available. By efficiently constructing and analyzing sets of de Bruijn graphs, Trinity fully reconstructs a large fraction of transcripts, including alternatively spliced isoforms and transcripts from recently duplicated genes. Compared with other de novo transcriptome assemblers, Trinity recovers more full-length transcripts across a broad range of expression levels, with a sensitivity similar to methods that rely on genome alignments. Our approach provides a unified solution for transcriptome reconstruction in any sample, especially in the absence of a reference genome.

PMID: 21572440 [PubMed - as supplied by publisher]

More...

Gangcai · 05-26-2011, 01:51 PM

Has anybody tried this method/software?

samanta · 07-28-2011, 06:58 PM

Yes, we tried the software extensively and found it to take much less RAM than Velvet+Oases.

SLB · 08-03-2011, 01:09 PM

I am performing RNA-seq without a reference genome and just tried Trinity. A quick look at some of the stats on the assemblies is very promising, an increase in the number of contigs above 1000 relative to my merged ABYSS assemblies from multiple kmers.

Has anyone used their Trinity assemblies directly for gene expression analysis? If so I would be interested to hear the approach people are taking. If Trinity is preserving transcript diversity and the resulting assemblies include 'all' possible isoforms of a gene, then is it possible to directly use these contigs as a reference for mapping the raw reads against to perform expression analysis?

morebasesplease · 08-06-2011, 08:24 PM

I just did a comparison between Velvet/Oases, Abyss/Trans-Abyss and Trinity. While Trinity can't be used in on multi k-values, it seems to be more less short transcripts.

jdjax · 08-07-2011, 09:50 AM

Trinity has a default k-mer length of 25 but you can change it. I used the k-mer at default and got much better transcript length then Abyss. Longest contig in Abyss was 1157 bp the longest in trinity was 1455 bp.

The program does not take as much memory as Velvet/Oases. However if your reads are long (100bp) it takes a while for inchworm to complete. For my 41 million ~ 100 bp reads it took four days. I think this is because of the short k-mer length and my long reads. If your reads are shorter I am sure if will complete faster.

The authors of the trinity suggested that for expression analysis to use the trinity contigs as a reference and align the reads using Bowtie and then cufflinks for expression comparison.

I am currently trying that now and even though it should be possible to go directly from Bowtie to SAMtools to Cufflinks I am getting errors and no one has come with an answer why.

Also I should mention that if you have a zombie process problem that occurs during butterfly, like I did. Check out the FAQ the answer to the problem is there. http://trinityrnaseq.sourceforge.net/

Kennels · 10-18-2011, 11:08 PM

Hi jdjax

Quote:

Originally Posted by jdjax

The program does not take as much memory as Velvet/Oases. However if your reads are long (100bp) it takes a while for inchworm to complete. For my 41 million ~ 100 bp reads it took four days. I think this is because of the short k-mer length and my long reads. If your reads are shorter I am sure if will complete faster.
http://trinityrnaseq.sourceforge.net/

What were your computer specs when you ran this analysis, and were your reads paired end?
Thanks

jdjax · 10-26-2011, 06:37 AM

Quote:

Originally Posted by Kennels

Hi jdjax

What were your computer specs when you ran this analysis, and were your reads paired end?
Thanks

I would recommend looking over this website for your answer:
http://sourceforge.net/mailarchive/f...tyrnaseq-users

This website going give you the answer you are looking for. If it is not in the website join the mailing list and then you can ask the manager of the program.

Similar Threads
Thread	Thread Starter	Forum	Replies	Last Post
RNA-Seq: Genome Wide Full-Length Transcript Analysis Using 5' and 3' Paired-End-Tag N	Newsbot!	Literature Watch	1	01-20-2012 06:38 PM
RNA-Seq: De novo transcriptome assembly of RNA-Seq reads with different strategies.	Newsbot!	Literature Watch	0	01-10-2012 05:00 AM
RNA-Seq: Composite Transcriptome Assembly of RNA-seq data in a Sheep Model for Delaye	Newsbot!	Literature Watch	0	03-26-2011 03:02 AM
RNA-Seq: Length Bias Correction for RNA-seq Data in Gene Set Analyses.	Newsbot!	Literature Watch	0	01-22-2011 03:02 AM
RNA-Seq: Comparison and calibration of transcriptome data from RNA-Seq and tiling arr	Newsbot!	Literature Watch	0	06-23-2010 03:00 AM

05-17-2011, 03:10 AM	#1
Newsbot! RSS Posting Maniac Join Date: Feb 2008 Posts: 1,443	RNA-Seq: Full-length transcriptome assembly from RNA-Seq data without a reference gen Syndicated from PubMed RSS Feeds Full-length transcriptome assembly from RNA-Seq data without a reference genome. Nat Biotechnol. 2011 May 15; Authors: Grabherr MG, Haas BJ, Yassour M, Levin JZ, Thompson DA, Amit I, Adiconis X, Fan L, Raychowdhury R, Zeng Q, Chen Z, Mauceli E, Hacohen N, Gnirke A, Rhind N, di Palma F, Birren BW, Nusbaum C, Lindblad-Toh K, Friedman N, Regev A Massively parallel sequencing of cDNA has enabled deep and efficient probing of transcriptomes. Current approaches for transcript reconstruction from such data often rely on aligning reads to a reference genome, and are thus unsuitable for samples with a partial or missing reference genome. Here we present the Trinity method for de novo assembly of full-length transcripts and evaluate it on samples from fission yeast, mouse and whitefly, whose reference genome is not yet available. By efficiently constructing and analyzing sets of de Bruijn graphs, Trinity fully reconstructs a large fraction of transcripts, including alternatively spliced isoforms and transcripts from recently duplicated genes. Compared with other de novo transcriptome assemblers, Trinity recovers more full-length transcripts across a broad range of expression levels, with a sensitivity similar to methods that rely on genome alignments. Our approach provides a unified solution for transcriptome reconstruction in any sample, especially in the absence of a reference genome. PMID: 21572440 [PubMed - as supplied by publisher] More...

07-28-2011, 06:58 PM	#3
samanta Senior Member Location: Seattle Join Date: Feb 2010 Posts: 109	Yes, we tried the software extensively and found it to take much less RAM than Velvet+Oases. __________________ http://homolog.us

08-07-2011, 09:50 AM	#6
jdjax Member Location: Denmark Join Date: Dec 2010 Posts: 23	Trinity has a default k-mer length of 25 but you can change it. I used the k-mer at default and got much better transcript length then Abyss. Longest contig in Abyss was 1157 bp the longest in trinity was 1455 bp. The program does not take as much memory as Velvet/Oases. However if your reads are long (100bp) it takes a while for inchworm to complete. For my 41 million ~ 100 bp reads it took four days. I think this is because of the short k-mer length and my long reads. If your reads are shorter I am sure if will complete faster. The authors of the trinity suggested that for expression analysis to use the trinity contigs as a reference and align the reads using Bowtie and then cufflinks for expression comparison. I am currently trying that now and even though it should be possible to go directly from Bowtie to SAMtools to Cufflinks I am getting errors and no one has come with an answer why. Also I should mention that if you have a zombie process problem that occurs during butterfly, like I did. Check out the FAQ the answer to the problem is there. http://trinityrnaseq.sourceforge.net/ __________________ jdjax Ph.d. Student Åarhus University Last edited by jdjax; 08-07-2011 at 09:55 AM. Reason: more info

05-26-2011, 01:51 PM	#2
Gangcai Member Location: Shanghai, China Join Date: Nov 2009 Posts: 30	Has anybody tried this method/software?

08-03-2011, 01:09 PM	#4
SLB Member Location: Ireland Join Date: Sep 2010 Posts: 21	I am performing RNA-seq without a reference genome and just tried Trinity. A quick look at some of the stats on the assemblies is very promising, an increase in the number of contigs above 1000 relative to my merged ABYSS assemblies from multiple kmers. Has anyone used their Trinity assemblies directly for gene expression analysis? If so I would be interested to hear the approach people are taking. If Trinity is preserving transcript diversity and the resulting assemblies include 'all' possible isoforms of a gene, then is it possible to directly use these contigs as a reference for mapping the raw reads against to perform expression analysis?

08-06-2011, 08:24 PM	#5
morebasesplease Junior Member Location: United States Join Date: Aug 2011 Posts: 2	I just did a comparison between Velvet/Oases, Abyss/Trans-Abyss and Trinity. While Trinity can't be used in on multi k-values, it seems to be more less short transcripts.