Hi Ao, When you try to lift the two biggest gaps within the coordinates you give, liftOver gives the following error: Partially deleted in new. This means too little of the sequence intersects with a single alignment chain in the region. One of our developers gave this explanation "Our general interpretation of alignment gaps is that sequences have accumulated too many mutations since the common ancestor for our alignment algorithms to identify a match, and/or there have been chromosomal rearrangements (insertions, deletions, inversions, duplications, translocations)." You can try by using blat or lastz to find regions that are similar to the DNA from those gaps, but there is no guarantee that you will find one closely aligned region that is homologous.
I hope this information is helpful. Please feel free to contact the mail list again if you require further assistance. Best, Mary ------------------ Mary Goldman UCSC Bioinformatics Group On 8/24/10 3:13 PM, Ao Zhou wrote: > ---------- Forwarded message ---------- > From: Ao Zhou<[email protected]> > Date: 2010/8/24 > Subject: Re: [Genome] About: Getting homolog regions in Rat, with Mouse > genomic coordinates > To: Mary Goldman<[email protected]> > > > Update: Liftover works, thank you very much!!! :) > I thought it could only be used in version transforms~ I does work between > mouse and rat! > > There is another question, in this page: > http://genome.ucsc.edu/cgi-bin/hgTracks?org=Mouse&db=mm8&position=chr1:175561150-175578575 > > I think the level 1 net alignment should have the most sequences mapped > right? > But in this page it seems like a big gap... what is the reason please? > > Thank you :) > > 2010/8/24 Ao Zhou<[email protected]> > > Dear Mary, > >> Thank you so much for answering my questions! >> >> For the first tool LiftOver, this is helpful in transforming the MM8 >> coordinates to MM9 coordinates, >> but cannot help in the homolog region search between Mouse and Rat. >> >> For the second tool, actually the regions I'm looking for are all >> inter-genic regions (we are looking for unknown genes), >> so it may not work either... T_T >> >> So is there any possible methods to get the results? >> What I wanted to do is: Given a mouse region, then search for the rat >> region with the highest alignment score to the mouse region. >> Thank you! :) >> >> >> >> >> 2010/8/24 Mary Goldman<[email protected]> >> >> Hi Ao, >> >>> There two other tools that may be able to help you. The first is the >>> liftOver tool (http://genome.ucsc.edu/cgi-bin/hgLiftOver), which allows >>> you to lift batch coordinates between assemblies. I am not sure if this will >>> work for you, given that it uses the same tables you were accessing in your >>> first method, but it is probably worth a try. >>> >>> The other tool that may be able to help you is the rnBlastTab table. This >>> table was built by taking all the genes in UCSC Genes, blasting them against >>> the rat genome and then filtering out the non-sytenic hits. However, this >>> table is only helpful if your all your genomic regions are genes in the UCSC >>> Genes track. If this is so, please write back and I will email further >>> instructions on how to use this table. >>> >>> I hope this information is helpful. Please feel free to contact the mail >>> list again if you require further assistance. >>> >>> Best, >>> Mary >>> ------------------ >>> Mary Goldman >>> UCSC Bioinformatics Group >>> >>> >>> On 8/22/10 10:47 AM, Ao Zhou wrote: >>> >>> >>>> Hi, there, >>>> >>>> I'm a beginner in UCSC genome browser, and I'm attempting to find out >>>> Rat homolog regions with Mouse genomic coordinates. >>>> The query composites of thousands tuples like this, the db version is >>>> *mm8*: >>>> chrX 138292125 138297400 >>>> >>>> What I want is like this, the db version is Rat* Rn4*: >>>> chrX - 35494226 35500120 >>>> >>>> I've already tried two methods, but both of them are not perfect: >>>> >>>> 1. In the table browser, Mammal -> Mouse -> Comparative Genomice -> >>>> Rat/Net >>>> -> input coordinate restrictions >>>> Pros: Batch download is available in this method >>>> Cons: The derived regions are actually highly homolog sub-regions >>>> within >>>> the query Mouse region, >>>> however, what I need is a single Rat region that can be >>>> considered as a homolog gene to the query Mouse region. >>>> 2. In the genome browser, Comparative Genomics -> Rat Full -> Click on >>>> the >>>> net alignment -> See the alignments >>>> Pros: The two regions are mapped, and a side by side alignment is >>>> also >>>> shown. >>>> Though the Rat coordinates are not clearly displayed, it can >>>> be >>>> infered from the alignments. >>>> Cons: Batch download is not available, I have thousands of Mouse >>>> regions >>>> to query... >>>> >>>> Other methods: >>>> I'm also thinking of do the mapping manually, or use PHP to search >>>> the >>>> result pages of the side by side alignment. >>>> But before doing that I have to make sure there's not an efficient >>>> way >>>> to do this task~ >>>> >>>> Could anyone tell me what tool I should use please...? >>>> Thank you! >>>> >>>> >>>> >>>> >>> >> >> -- >> Ao Zhou >> Ph.D. Student, >> Bioinformatics Program, >> School of Informatics, IUPUI >> Tel: 317-372-7004 >> >> > > > _______________________________________________ Genome maillist - [email protected] https://lists.soe.ucsc.edu/mailman/listinfo/genome
