Hi Sean, just a few comments to add to what Guy and Pat already said... On Thu, 2009-10-29 at 09:40 -0500, Pat Burke wrote: > The web site is //psb.stanford.edu/psb-online/proceedings/psb02/
> > On 10/28/09, Sean Conlan wrote: > >> > >> I have used Mauve to align a set of syntenic regions in a group of 7 > >> mammals for a colleague's manuscript. One of the reviewers requested > >> additional information about the use of the HOXD scoring model. Is > >> this documented anywhere (I can't find it in the manual)? From what I > >> can tell, it is used for assigning penalties for mismatches, gaps, > >> etc... Is there any additional information on how the scoring matrix > >> was calculated and its appropriateness for various tpyes of analysis? > >> Indeed the derivation of the HOXD70 matrix which Mauve uses by default is described in that paper, but I couldn't get it from the PSB site so I found other PDF links in google scholar: http://scholar.google.com/scholar?cluster=14579713005665430682&hl=en And now to give some background regarding why we use HOXD70: from the paper abstract, "One particular scoring matrix, called HOXD70, has proven to be generally effective for human-mouse comparisons." The HOXD70 matrix was derived on sequences with approximately 50% GC content and indeed others have found that it works very well for aligning mammalian sequence. Work by Gerton Lunter suggests that mammalian alignments using HOXD achieve nearly maximal accuracy with standard alignment approaches (but more computationally expensive Bayesian statistical alignment techniques can do better). Progressive Mauve uses that matrix by default, although the scoring matrix can be changed via the GUI. While 50%GC might be a good middle ground for a default substitution matrix, many organisms have a GC content nowhere near 50% and progressiveMauve takes some measures to correct for alignment artifacts that might be introduced by specifying the wrong substitution matrix. After an initial alignment has been constructed, the alignment is subjected to a quality filter (we call it a Homology HMM) which is tuned for the actual GC content of the organisms being aligned. We described the Homology HMM and its tuning to arbitrary GC content in this paper: http://portal.acm.org/citation.cfm?id=1544181 Finally, the scoring matrix is used in various other parts of the Progressive Mauve algorithm. I have just today posted a draft manuscript describing the progressiveMauve algorithm to the arXiv preprint server: http://arxiv.org/abs/0910.5780 That link should become active sometime after 6pm EDT. Although the manuscript is dated June 19th 2008, it describes the progressiveMauve algorithm fairly accurately. Cheers, -Aaron ------------------------------------------------------------------------------ Come build with us! The BlackBerry(R) Developer Conference in SF, CA is the only developer event you need to attend this year. Jumpstart your developing skills, take BlackBerry mobile applications to market and stay ahead of the curve. Join us from November 9 - 12, 2009. Register now! http://p.sf.net/sfu/devconference _______________________________________________ Mauve-users mailing list [email protected] https://lists.sourceforge.net/lists/listinfo/mauve-users
