Tophat for finding long ncRNA with short reads?

[KevinLam]

Hi,
has anyone used Tophat to look at long non coding RNAs or mRNA like ncRNAs using NGS?

I am assuming of course that things would be easier if I had 454 or even Sanger reads. But Solexa and SOLiD would be more bang for the buck.

aside from de novo assembly of the transcriptome. I am wondering if it is possible to use Tophat / Cufflinks to generate the transcripts

[natstreet]

I actually discovered an interesting ncRNA using the tophat-->cufflinks-->cuffcompare-->cudffdiff pipeline.

I map my reads to the reference using tophat, then run cufflinks without providing a reference gtf file. Following that I run cuffcompare where I do provide the reference gtf file and finally cuffdiff. I then screened through the significant cuffdiff results looking at the loci with no corresponding gene name entry from the reference gtf and tried to classify these. One is clearly a novel sRNA loci that has previously not been described and certainly not in the trait that my different samples come from.

So, in summary, it seems to work.

[yjhua2110]

grouping small rna dataset into long RNA clusters

You can try to group your small rna dataset into long RNA clusters using methods described in our deepBase ( http://deepbase.sysu.edu.cn/ ) (Yang et al. Nucleic Acids Res. 2010 Jan;38: D123-D130.).

we have identified ~1.2 million long RNA clusters by grouping all the mapped small RNA reads according to their distance in our deepBase platform ( http://deepbase.sysu.edu.cn/ ), which were developed to discover small and long ncRNAs from deep-sequencing data.

These ~1.2 million RNA clusters that include multiple classes of infrastructural ncRNAs, miRNAs precursor, piRNA precursors, repeat-associated siRNA precursors and evolutionarily conserved phastCons elements.

we have identified ~2000 microRNA candidates and ~1890 snoRNA candidates using improved miRDeep and our snoSeeker programs from these RNA clusters.

[sahusarika]

Hello everyone
I hv rnaseq data but reference genome is not available.I want to predict lncRNA from rnaseq data. Pls give ur valuable suggestions.