which denovo assemble to use for assembly of meta-transcriptome data

[bambus]

Hi,

I have couple of questions.

Firstly, I am looking for a denovo assembly tool that can assemble a meta-transcriptome data(paired-end sequence(insert size=300) **consists of mixed sequence reads of multiple species in a microbial community,sequenced using Illumina Genome analyser.I have few number of libraries each holds from ~700,00 to 17million reads.

And in search of that I came across a METAVELVET,a de novo metagenome assembly and I am not sure whether this works well with my data?

Secondly,I used SORTMERNA tool,to detect and remove rRNA's from the same data(whole library) mentioned above and the output was,

file1:- non-rrna's

file2:- rrna's

i.e total percent of rrna it could find is 54.65% and nonrrna 45.35%,but when I checked file1(interested for my work)many rrna's were detected which means the tool misclassified rRNA's as nonrRna's(got to know by doing blastx)then I repeated the process again but this time not on the whole data library instead only on non-rRNA output (file1) generated after appying SORTMERNA on complete data library.This time it again classified data as 51.3% of rRNA and 48.7% nonrrna only from the non-rrna file.

Why the tool couldn't classify all rrna's present in the data with the first run,but in the second run?(still few misclassifications)

It would be very helpful if someone could help me out.

Thanks in advance!!

[westerman]

... I came across a METAVELVET,a de novo metagenome assembly and I am not sure whether this works well with my data?

Probably not. Velvet is meant for genomes. You have a transcriptome. Different types of information.

I can't answer your second question (about SortMerna) since I do not use that tool. I do wish to point out that rRNA filtering without know the species-specific rRNAs up-front is prone to false negative and false positives. Unless I can run a species-specific rRNA filtering then I am just happy to have most, but not all, of the rRNA removed.

[lostale]

try SortMeRNA v1.99 beta

The latest version (1.99 beta) uses sequence alignment and the E-value to classify reads as rRNA. This should give you more precision over the previous versions.

[gringer]

There's also Ray, which includes a metagenome analysis plugin.