mitohondrial genome assembly and annotation.

[chhavi.dawar]

I am about to start working with mitohondrial genome of animal systems. I am keen to know as to what an be used to form a draft genome for annotation and how should one go about it. I already have the assembled 454 contigs but how to form scaffolds or draft genome as most mitochondrial genome annotation programs take one large saffold or draft genome. Also I am going about a de novo assembly and annotation.
Thanks

[JackieBadger]

Ive used DOGMA where you can put in a multi-data file

http://dogma.ccbb.utexas.edu

[chhavi.dawar]

Ok Thanks Jackie, actualy I was trying MITOS and it takes a single fasta file with one entry. In any case how did u assemble the contigs in a draft genome?

[JackieBadger]

MITOs works yeah...
I aligned my contigs against a reference of a closely related species if available, otherwise we used MITObim...http://nar.oxfordjournals.org/conten...kt371.abstract

[AdrianP]

After assembling your 454 reads into contigs (i recommend mira), blast all mitochndrial genes against a custom blast database of your assembly. This will tell you which contigs are mitochondrial. Use a close species as a query, or just an animal model, like human. Animal mitochondria are small in size, up to 20kb, and have your typical set of genes, cob cox1-3, atp6 8 9, nadh1-6, etc, so use all when blasting vs your assembly.

Your genome like tgat of most animals likely maps as a circle. So you need to be able to circularize your genome at one point.

Is 454 all you got? Whats tge coverage?

[chhavi.dawar]

Hi Adrian
I have quite a good coverage of sequence data (100x) and I have done assembly using newbler as i needed to remove the MIDs. I am not very sure if MIRA can do that. My basic concern is if you could suggest some tool or script wherein I can form a scaffold or draft genome and then go for annotation.
Thanks

[maubp]

I would use 'sfffile' from the Roche 'off instrument applications' (the tool set including 'Newbler' gsAssember) to split and SFF file by MID and remove the MID markers, and then use the data with MIRA.

See also http://seqanswers.com/forums/showthread.php?t=8642

[JackieBadger]

MITObim is a wrapper for MIRA that produces a full length genome from a small seed fragment e.g. a COI reference sequence

[AdrianP]

Hi Adrian
I have quite a good coverage of sequence data (100x) and I have done assembly using newbler as i needed to remove the MIDs. I am not very sure if MIRA can do that. My basic concern is if you could suggest some tool or script wherein I can form a scaffold or draft genome and then go for annotation.
Thanks

You seem to want a script that will simply generate a scaffold from contigs and make it ready for annotation. Let me point out:
1) Scaffolds are orientations of contigs separated by NNN. No peer reviewer will accept a mitochondrial genome in a form of a scaffold, you will need to close all gaps and have one big circular contig that will represent the plasmid of your mitochondrion.

2) You are at a point where you have contigs. The best way to link these contigs together into a circular sequence, is by read mapping. You take your mitochondrial contigs, and you map reads to it. Than, you look at the edges of the contigs, are the contigs extended? If yes, generate a new consensus, with slightly bigger contigs, and repeat the step. Repeat until the contigs overlap and you can link them into supercontigs. You can read my paper (Pelin et. al. 2012) for the approach used and tools used. This isn't easy, took me a month to go from 7 contigs to 1 circular contig. I had to use PCR as well.

Adrian