Hi Perdeep, Yes, setting minMatch to 01 (100%) would definitely be an issue.
- Greg On 4/15/11 7:05 AM, Mehta, Perdeep wrote: > Hi Greg, > > Thank you for your response. By looking at your suggestion, I think I have > found the reason though I have not tested it yet. I had already used the > -minMatch switch, but for some reason instead of 0.1 it was set to 01. I had > a typo in the file where this command was stored for later use and I had > copy/paste it. I apologize also for not explicitly writing the full command > that I had used. > > I will test it soon and if there are still any problem with let you know. > > Thanks again for helping with this. > perdeep > > Perdeep K. Mehta, PhD > Research Informatics, Information Sciences Division > St. Jude Children's Research Hospital > 262 Danny Thomas Place > Memphis, TN 38105 > > -----Original Message----- > From: Greg Roe [mailto:[email protected]] > Sent: Monday, April 11, 2011 4:17 PM > To: Mehta, Perdeep > Cc: '[email protected]' > Subject: Re: [Genome] lifting over Dog genomic coordinates to Human genome > > Hi Perdeep, > > Sorry for the delayed reply. > > "Partially deleted in new" means that the sequence intersects with only > part of a single alignment chain in the region. I suspect you have the > "Minimum ratio of bases that must remap/Minmatch" setting too high. The > online version of liftOver defaults minmatch to .95 for within species > liftovers and 0.1 for liftOvers between different species. The command > line liftOver application always defaults to 0.95. Thus you may get > different results between the two if you don't make sure nimMatch is set > the same. > > Try running the liftOver again at the command line and setting -minMatch=0.1 > > You can see all liftOver options at the command line by just typing > "liftOver". > > Let us know if you have any additional questions: [email protected] > > - > Greg Roe > UCSC Genome Bioinformatics Group > > > On 4/4/11 2:58 PM, Mehta, Perdeep wrote: >> Hi, >> >> I have a couple of files containing several Dog genomic coordinates that I >> want to lift over to the Human genome. I have down loaded the >> canFam2ToHg19.over.chain files from UCSC website and using it with >> previously installed UCSC "liftOver" program to do this. >> >> For a test, I used just two Dog genome coordinates, as listed below (in file) >> chr20 58241270 58241311 contig_1 42 + >> chr6 54123651 54123725 contig_2 75 + >> >> When I use the liftOver program with canFam2ToHg19.over.chain, I get one >> mapped to Human genome and second one goes to unmapped output. For example, >> This one gets mapped to Human genome (out file 1) >> chr1 99175258 99175341 contig_2 1 - >> >> And the other one remains unmapped (out file 2) >> #Partially deleted in new >> chr20 58241270 58241311 contig_1 42 + >> >> When I tested this by using their original sequences the both Dog contigs >> actually show corresponding interonic regions of Human genome. First one >> mapping to SH3GL2 gene and second to SNX7 gene (this is correctly identified >> by liftOver program). >> >> I do not also understand what "#Partially deleted in new" means? Why is it >> not able to lift over the coordinates of the first contig? >> >> Here are the two sequences used; >>> contig_1 >> GGCCAGCTCCTGAGTCTGCTGCCCTCTTAGAGCGTAGCCACA >>> contig_2 >> GGCAGGTCTGGATCTGAATTGAAGCTCTTAATAGCCTAGGAACTAATGTAAAAAGTGGGATGGGATCAGGTATAA >> >> I will appreciate any help. >> >> Thanks, >> perdeep >> >> Perdeep K. Mehta, PhD >> Research Scientist, Bioinformatics >> Research Informatics, Information Sciences Division >> St. Jude Children's Research Hospital >> >> >> >> ________________________________ >> Email Disclaimer: www.stjude.org/emaildisclaimer >> _______________________________________________ >> Genome maillist - [email protected] >> https://lists.soe.ucsc.edu/mailman/listinfo/genome > _______________________________________________ Genome maillist - [email protected] https://lists.soe.ucsc.edu/mailman/listinfo/genome
