Thanks a lot Ann! That explains a lot of my confusions. I thought the intersection was going to return only portions of MAF blocks that intersect with the custom track. Actually, it returns complete blocks that have ANY intersection with the custom tracks.
In fact, there was an option when you click "intersect", checking "Base-pair-wise intersection" would give me the right thing I wanted! My apology for not reading it carefully... Thanks very much. Have a good day, Wen On Jan 6, 2009, at 11:19 AM, Ann Zweig wrote: > Hello Wen, > > First of all, let me thank you for sending such a clear example. > > The MAF records are stored in blocks that are usually much larger > than the width of an exon. So, when you ask the Table Browser for > an intersection between your Custom Track and the multiz table, the > Table Browser returns any MAF block(s) that intersects with your > exon(s). The coordinates (and sequence) for your exon will be found > somewhere within the MAF block. You can read about how to decipher > MAF files here: http://genome.ucsc.edu/FAQ/FAQformat#format5 That > should explain the apparent coordinate shift. > > I have another solution to offer you; a possibly easier way to find > the MAF blocks for only the coding exons. We have a new tool > available on the website that provides MAF alignments for gene sets > in either DNA- or protein-space, for either full genes or exons > only. If you don't have too many genes to check, this tool will > work well for you. > > For example, in bosTau4, one of the RefSeq Genes in your example > location is: PIP5K1A (NM_001103109 chr3:21084405-21123537). Click > directly on that RefSeq Gene to go to the details page. On the > details page, click on the following link: CDS FASTA alignment from > multiple alignment: _NM_001103109_ > > Configure the page like so: > > Formatting options: > x Separate into exons > x Show nucleotides > > Then press 'submit' to see the per-exon MAF alignment for all 5 > organisms. > > I hope this information is helpful to you. Please don't hesitate to > contact the mail list again if you require further assistance. > > Regards, > > ---------- > Ann Zweig > UCSC Genome Bioinformatics Group > http://genome.ucsc.edu > > > > > > Wen wrote: >> Hello, >> I have a question on mutiz5way table in the conservation track of the >> Oct.2007 assembly of cow genome. I'm trying to fetch the 5-way MAF >> for, >> say, coding regions of some genes (or whole gene, 5'utr, 3'utr...) >> A little example here, query the region chr3:21083267-21232422 >> 1. make a custom track of RefGene coding exons in this region >> 2. intersect the multiz5way with custom track made in the previous >> step. >> 3. look at coordinates in the returned MAF and compare them to what >> are >> in the custom track, apparently they don't intersect... >> I am just wondering if I am doing something wrong and want to bring >> potential problem to your attention. >> Thanks a lot for your help, >> Wen >> _______________________________________________ >> Genome maillist - [email protected] >> http://www.soe.ucsc.edu/mailman/listinfo/genome _______________________________________________ Genome maillist - [email protected] http://www.soe.ucsc.edu/mailman/listinfo/genome
