Hi Jeff, On Tue, 2010-10-19 at 16:09 -0400, Rosenfeld, Jeffrey wrote: > Hi Aaron, > > I am trying to use Mauve for aligning human exome sequences. I > know this is not the conventional use of Mauve, but I think it shouldn't > be too hard to get done.
The unusual ways in which people use Mauve continue to surprise me. I would recommend reading at least the introduction to this paper: http://www.plosone.org/article/info%3Adoi%2F10.1371% 2Fjournal.pone.0011147 And thinking about whether the way duplications are treated will cause problems with gene families in the data you are analyzing. > Each of the exomes consists of ~30,000 > genes/contigs and the total size of the contigs is ~30MB. I have been > able to get Mauve to just do a simple alignment of the contigs, but that > is not very useful for me. My main purpose in doing the comparative > alignment of the two exomes is to identify rearrangements between them. > In order to effectively do this, I need to be able to do one of two > things, but preferably both of them. > > 1. Have Mauve maintain the names of my contigs so that I can look at the > alignment and see which contigs from one exome align to two contigs of > the other exome (indicating a rearrangment). The exomes are in fsta > format with the contig names in the tags. > Unfortunately I don't have an easy recommendation for this issue. The current XMFA alignment output records aligned blocks in a coordinate system that represents all contigs as a single concatenated sequence. I've been wanting to add contig representation to the XMFA for a while now so I'll see about moving it higher up on my (ever growing) todo list. In the meantime, you might try using the Mauve Contig Mover which is part of Mauve 2.3.1 and was published as Rissman et al 2009 in Bioinformatics. One of the files it creates is called _contigs.tab and contains a set of contigs with conflicting ordering information. That list should contain most of the cases where one of your 2nd exome contigs aligns to two contigs in the first exome, with the only exception being when the two contigs from the first exome are adjacent to each other. > 2. Align the exomes to the set of Human Genes from Genbank. I have > tried downloading the full set of human genes from Genbank, but Mauve > will not accept it. > What sequence file format were the genes in? The Mauve GUI uses BioJava to read sequences, and can parse GenBank, FastA, INSDseq XML, and embl. The aligner is a separate program that only made it through primary school and is literate only in GenBank and FastA. If they are GenBank format, I'd recommend verifying that there isn't any stray and unwanted HTML formatting in the files since that seems to be a common occurrence with downloads from NCBI's web site. Best, -Aaron ------------------------------------------------------------------------------ Nokia and AT&T present the 2010 Calling All Innovators-North America contest Create new apps & games for the Nokia N8 for consumers in U.S. and Canada $10 million total in prizes - $4M cash, 500 devices, nearly $6M in marketing Develop with Nokia Qt SDK, Web Runtime, or Java and Publish to Ovi Store http://p.sf.net/sfu/nokia-dev2dev _______________________________________________ Mauve-users mailing list [email protected] https://lists.sourceforge.net/lists/listinfo/mauve-users
