De novo assembled transcript mapping to genome scaffold

Question

[Bioinfonext]

Hi,

I did mapping of illumina raw reads to CDS transcripts and I extracted unmapped raw reads.

after that I did assembly of these unmapped raw reads. Now I want to map these assembled transcript to genome scaffold to find how many of them are able to map to genome and whether they map to different position to each others.

FINALLY will extract only those DE NOVO ASSEMBLED TRANSCRIPT which are able to map to genome but not annotated as CDS.

So my finally reference for differential gene expression will be CDS plus de novo assembled transcripts.

[Brian Bushnell]

It's helpful that you have explained what you have tried, but it would also help if you would explain your goal, and what kind of data you are working with. Also, it's worth noting that you did not technically ask a question, so perhaps it would be worthwhile rephrasing your text.

[Bioinfonext]

Actually I notice that some of the gene sequences were present in the genome scafold but not present in the CDS trasncripts.

when I map raw read to genome than around 95 % raw read able to map to genome but when I map to CDS than only 60% were able to map.

So now I want to confirm how many of the de novo assembled trascript from (unmapped raw read to CDS), can able to map on the genome scaffold. this is Transcriptome data from Plant roots at different development stages.

[Bioinfonext]

it will also confirm the quality of de novo assembled transcript

[Brian Bushnell]

You will get a much better response rate if you try to use proper English with correct capitalization. Sloppy linguistics is generally frowned upon in this forum.

Answer 1

[wrf]

For mapping de novo assembled transcripts to a genome, you may want to try GMAP, here. The usage is sort of like bowtie or blast, where you make the database from the scaffolds, then align in a second step.

[apa@stowers]

GMAP is an excellent first step for mapping -- way, way faster than blat -- but it achieves speed by playing fast and loose, often not achieving results as good as blat -- e.g. failing to align terminal exons, returning multiple tiny exons, etc where blat does not.

You can use GMAP to find the gene coords, but then I would strongly recommend using Exonerate to re-align the transcript into the GMAP region, and output GFF results.