Mitochondrial Genome Assembly

[chofman]

Hi all,

I'm new to the bioinformatics world and am looking for an assembly/alignment program that would work well for the mammalian mitochondrial genome. I've played around with Mira and Velvet for de Novo assembly but could use some advice. I have 454 reads and a divergent reference genome. I also will be mapping ancient DNA reads once I have a reference. Any suggestions would be great!

Thanks!

[maubp]

Given you have 454 reads, have you not tried Newbler (gsAssember) from Roche?

I've used Newbler to map 454 genome reads to animal mitochondria sequences from close strains, and it worked nicely. However, with more divergent sequences you may have issues.

One thing to be aware of is that the Newbler gsMapper (and every other tool I've explored) does not support a circular reference sequence. There are ways round this, such as doing the mapping twice (with a shifted origin the second time), or first extending the linear sequence (by at least the length or the longest read, or just double the reference). Both of these then require post processing of the alignment information, and presenting it cleanly in an assembly/mapping file format like ACE or SAM/BAM is a problem.