Long Read Mapper

[mitochy]

Does anyone know program that map long read to reference genome (not de novo)?

For example: BWA has bwasw which basically map long read which is the only one I know.

Long read as in 1KBp - 10KBp.

I will have millions - billions of reads

[ulz_peter]

what about BLAT or BLAST (in case you haven't got millions of them)

[koadman]

You could try one of the following:
STELLAR - http://www.seqan.de/projects/stellar.html
bwt-sw - http://i.cs.hku.hk/~ckwong3/bwtsw/
SHRIMP, though the default settings might need some extensive tuning - http://compbio.cs.toronto.edu/shrimp/

Also, the STELLAR paper has a nice comparison of speed and how the methods handle different amounts of sequencing error:

http://www.biomedcentral.com/1471-2105/12/S9/S15

Update: the pacbio smrtanalysis pipeline has a program called blasr that can also map long reads, at least from their instrument and possibly others.

[arvid]

GMAP should be able to splice-align long reads.

[gaffa]

Other alternatives include SSAHA2 (http://www.sanger.ac.uk/resources/software/ssaha2/) and SMALT (http://www.sanger.ac.uk/resources/software/smalt/).

[steven]

Right, GMAP should do that. MapSplice too.

GMAP:
-code
-ref1
-ref2

MapSplice:
-code
-Ref
-user guide

These two should handle proper "spliced-alignment" (allowing long gaps to span over introns), which is what you would need if you are doing RNA-seq.

[darked89]

You may try LAST: http://last.cbrc.jp/
It accepts fastq format, outputs MAF, which then you can convert to SAM/BAM.

[mitochy]

Thank you! Trying them out now!

[mchaisso]

blasr is on git at

https://github.com/PacificBiosciences/blasr

cheers,
-mark

[twaddlac]

I'm a big fan of the mummer package http://mummer.sourceforge.net/