Hello Richard, One of our developers had this to say about your question:
You can pick up this tar image: http://genome-test.cse.ucsc.edu/~hiram/sacCer2/sacCer1ToSacCer2.liftOver.procedure.tgz It is a copy of the directory where all procedures were performed to construct the sacCer1 to sacCer2 liftOver file. You can read the scripts there to see the procedure. The procedure is as follows: 1. partition each genome into small chunks, creates liftUp files to translate from chunk coordinates to parent chromosome coordinate 2. blat each chunk from one genome to all chunks of the other genome (cluster operation) the blat command is: blat $targetList reSplitQuery.lst tmpUnlifted.psl \ -tileSize=11 -ooc=/hive/data/genomes/sacCer1/11.ooc -minScore=100 \ -minIdentity=98 -fastMap -noHead the results are "lifted" back to their chromosome coordinates 3. run axtChain on the chunks to get the corresponding chunks lined up. The axtChain command is: axtChain -verbose=0 -linearGap=medium -psl stdin sacCer1.2bit sacCer2.2bit $tmpOut You can find the scripts used in these procedures in the kent source tree. For more information about installing the kent source tree please see this FAQ about installing Genome Browser source code: http://genome.ucsc.edu/FAQ/FAQdownloads.html#download27 Best regards, Pauline Fujita UCSC Genome Bioinformatics Group http://genome.ucsc.edu On 06/08/11 14:57, Richard Badge wrote: > Hello Richard, > > Thank you for your question. In order to receive a more timely answer > and for the benefit of all of our users, for whom we try to have the > most comprehensive archives, I would appreciate it if you resent your > email to [email protected]. Thank you for your understanding in this > matter. > > Best regards, > > Pauline Fujita > > UCSC Genome Bioinformatics Group > http://genome.ucsc.edu > > > > > > On 6/7/11 2:09 PM, Richard Badge wrote: >> Hi Pauline, >> >> I wonder you could clarify what "somewhat involved" is. >> >> I can see how using blat to align the sequences would be feasible, but we >> would still need to parse the alignment to achieve equivalencing between >> base positions in the two strains. >> >> I was assuming that the over.chain file, once generated, could be used as a >> substrate for the liftover utility, along with our list of SNPs to convert >> coordinates between the two. As we have 24 complete genomes worth of SNPs, >> manually or even scripted individual alignments would seem an inefficient >> approach... >> >> Any light that you could shed would be greatly appreciated, >> >> Regards >> >> Richard >> On 3 Jun 2011, at 02:11, Pauline Fujita wrote: >> >> >>> Hello Richard, >>> >>> Unfortunately we cannot provide lifts between species we do not display >>> in the browser let alone strains. The process we use to generate the >>> liftover files is somewhat involved and an easier solution would be to >>> simply BLAT the regions you are looking to map. It sounds like you have >>> quite a few regions - in which case we recommend downloading and using >>> our standalone BLAT utility. You can read more about that in this FAQ: >>> >>> http://genome.ucsc.edu/FAQ/FAQblat.html#blat3 >>> >>> Best regards, >>> >>> Pauline Fujita >>> UCSC Genome Bioinformatics Group >>> http://genome.ucsc.edu >>> >>> >>> >>> On 06/01/11 05:03, Richard Badge wrote: >>> >>>> Dear genome browser mailing list, >>>> >>>> I came to send this email at the suggestion found on this page >>>> (http://genome.ucsc.edu/goldenPath/help/hgTracksHelp.html#Convert) that if >>>> an over.chain file does not exist for a given pair of species that you may >>>> be able to generate one for us... >>>> >>>> My colleagues and I are engaged in mapping SNPs between two yeast strains >>>> (Y55 and SK1) that have partially completed reference genome sequences. We >>>> would like to be able to ascertain the coordinates of corresponding SNP >>>> positions between the two genomes, as part of a recombination breakpoint >>>> mapping project involving genome re-sequencing of meiotic tetrads. >>>> >>>> We were aware of the utility of liftover in moving annotation from one >>>> assembly to another, and wondered if it could similarly be used for >>>> different strains (these are estimated to be about 0.5% diverged at the >>>> nucleotide level). Ultimately the over.chain file will give us a means to >>>> equate SNP positions, and to validate our sequence read mappings... >>>> >>>> If this is a possibility, or you could let us know how such a file can be >>>> generated, we would be very grateful, >>>> >>>> Regards >>>> >>>> Richard >>>> Dr Richard Badge, Lecturer >>>> Department of Genetics, >>>> Office 130, Adrian Building, >>>> University of Leicester, >>>> University Road, >>>> Leicester >>>> LE1 7RH >>>> Tel (Office): 0116 2525042 >>>> Tel (Lab): 0116 2523416 >>>> Fax: 0116 2523378 >>>> >>>> >>>> _______________________________________________ >>>> Genome maillist - [email protected] >>>> https://lists.soe.ucsc.edu/mailman/listinfo/genome >>>> >> Dr Richard Badge, Lecturer >> Department of Genetics, >> University of Leicester, >> Leicester >> LE1 7RH >> Tel: 0116 2525042 >> >> > Dr Richard Badge, Lecturer > Department of Genetics, > University of Leicester, > Leicester > LE1 7RH > Tel: 0116 2525042 > > _______________________________________________ > Genome maillist - [email protected] > https://lists.soe.ucsc.edu/mailman/listinfo/genome _______________________________________________ Genome maillist - [email protected] https://lists.soe.ucsc.edu/mailman/listinfo/genome
