Scaffolds Aligment

Question

[Matteo]

Hi all,

I have the sequence of a scaffold in fasta format and i would like to get the chromosome coordinates (boundaries of the scaffold) on the target genome.

Are there any tools to perform this issue?

I have tried Blast but the scaffold spans over different contigs and i cannot retrieve the left and right end coordinates over the genome.

Greetings,

Matteo

[Ashwin]

probably you can use BLAT utility from UCSC [ URL : http://hgdownload.cse.ucsc.edu/downloads.html ]

[Matteo]

I'm working on Arabidopsis scaffolds. Plants are not covered in UCSC :(

[Neilfws]

Ashwin refers to software available for download from UCSC, not the genome browser.

[Michael]

local alignment is not required, you need semi-local (global local) alignment. If it's only one scaffold glsearch should do.

[lh3]

No, for contig alignment, it is crucial to do local alignment. There are structural variations and misassemblies. Glocal alignment will miss all these interesting events.

[Michael]

yes, you are right

Answer 1

[lh3]

For bacteria and arabidopsis, mummer could be a good candidate. You may also try bwa-sw, which is designed for accurate alignment. Blat also works, but it is slow for long contigs. It may miss some split contig in the presence of repeats, too.

[Philipp Bayer]

Another good candidate if BLAT is too slow is LASTZ: http://www.bx.psu.edu/miller_lab/dist/README.lastz-1.02.00/README.lastz-1.02.00a.html

[Matteo]

Thanks! Mummer is what i was looking for.

Answer 2

[biomed.lu]

http://contiguator.sourceforge.net/ This tool will help you to map your scaffolds/contigs back to ref genome and output the chromosome coordinates...

[Matteo]

Can contiguator theoretically works for non-bacterial genomes? I've take a glimpse to the software and seems that it needs overlapping contigs. Unfortunately i'm dealing with non-overlapping scaffolds.

[mgalactus]

I, i'm the creator of contiguator: i've developed the tool to cope with bacterial genomes, but it may work well also for non-bacterial genomes. If you can give it a try we could even try to improve it in order to work better with those genomes. Let me know :)