Dear Professor, Many thanks for your reply and explanation. I just download again the desired output in BED format through UCSC. However, I still feel a bit with the following case: Case 1: Gene Name Chromosome Start Position End Position Remarks PCDH15 10 55568451 55569239 3'UTR PCDH15 10 55568451 55569306 coding exons PCDH15 10 55568502 55568819 intron
It seems like PCDH15 is overlap based on UCSC output? I can't really distinguish which regions is 3'UTR, coding exons or intron of PCDH15. Case 2: Gene Name Chromosome Start Position End Position Remarks ATP5L 11 118279813 118280562 3'UTR ATP5L 11 118279714 118280562 5'UTR It will be good if you can provide which some real case to distinguish promoter region, 3'UTR, coding exons, 5'UTR and intron from the UCSC output result. Thanks for your advice regarding miRNA, rRNA, tRNA, etc. I able to solve it out now. Looking forward to hear from you. best regards Edge ________________________________ From: Brooke Rhead <[email protected]> To: Edge Edge <[email protected]> Cc: ucsc <[email protected]> Sent: Saturday, June 30, 2012 11:31 AM Subject: Re: [Genome] Help with extract rRNA, tRNA, 3'-UTR and 5'-UTR from UCSC Genome Browser Hi Edge, > perfectly fine." In biology behaviour of human genome, is it normal > that 3'-UTR overlap with 5'-UTR? There is a bug in the Table Browser that I think is causing a lot of confusion here. When you choose to retrieve 3' UTR or 5' UTR output from a region where there is a non-coding gene, the Table Browser should not return any result for that non-coding gene. We have recorded this bug and we hope to fix it eventually. In the meantime, you will need another way to distinguish UTRs from genes that are entirely non-coding. > Can I know that whether UCSC got provide the coordinates of human > miRNA? If yes, can I know how to extract it out? There are several tracks that include miRNA annotations. To decide which track you want to use, you can click on the track name on the main Genome Browser page (http://genome.ucsc.edu/cgi-bin/hgTracks) and read about where the data comes from and how recent it is. Some tracks you might want to consider on hg19 are: sno/miRNA, tRNA Genes, RefSeq Genes, UCSC Genes (which contains annotations based on data from RefSeq, Genbank, CCDS, UniProt, Rfam, and the tRNA Genes track), and GENCODE. To get coordinate positions, there is no need to retrieve sequence from the Table Browser. There are two different ways to get coordinate positions. The first is to choose either "all fields from selected table" or "selected fields from primary and related tables" as the output format. The start and stop coordinates will be included in the output. Also, if the table includes UTRs (as in the refGene and knownGene tables, for instance), you can calculate the positions of the UTRs by subtracting txStart from cdsStart and cdsEnd from txEnd. (To see a description of the fields of any table, select it in the Table Browser and hit the "describe table schema" button.) To read more about the way genome coordinates are stored in our tables, see: http://genomewiki.ucsc.edu/index.php/Coordinate_Transforms. The other way to get coordinate positions is to select the "BED - browser extensible data" as the output format. BED format is described here: http://genome.ucsc.edu/FAQ/FAQformat.html#format1. If you use this method, you will be able to limit the output to just the UTRs. However, you will run into the bug described above of completely non-coding genes showing up in the output. I hope this helps explain what you are seeing in the Genome Browser and helps you get a start on using the Table Browser. If you haven't seen the Open Helix tutorials on the Browser, especially the Table Browser tutorial, you might find these helpful: http://www.openhelix.com/ucsc. If you have further questions, please contact us again at [email protected]. -- Brooke Rhead UCSC Genome Bioinformatics Group On 6/28/12 6:08 PM, Edge Edge wrote: > Dear Professor, > > > According your previous reply, "This is why you see overlap between > the 5’ UTR of uc010nxq.1 and the 3’ UTR of uc001aaa.3 and this is > perfectly fine." In biology behaviour of human genome, is it normal > that 3'-UTR overlap with 5'-UTR? > > Can I know that whether UCSC got provide the coordinates of human > miRNA? If yes, can I know how to extract it out? > > I got try to extract out the human exon coordinates by using the > following method: 1. Set the following options: Group: Genes and Gene > Prediction Tracks Track: RefSeq Genes Table: knownGene Output format: > sequence > > 2. Click the "get output" button > > 3. Select "genomic" and click the "submit" button > > 4. Check the " CDS Exons " > > 5. Click the "get sequence" button > > From the result, it shown that >> hg19_refGene_NM_032291 range=chr1:67000042-67208778 5'pad=0 3'pad=0 >> strand=+ repeatMasking=none > However, when I check the coordinates through UCSC Genome Browser, it > shown that chr1:67,000,042-67,208,778 is fall in Gene region but not > exons of that particular gene. > > Can I know how to just extract the coordinate of each exon in every > gene? My main purpose just hope can extract the coordinates of > promoter, 3'-UTR, 5'-UTR, miRNA, rRNA, tRNA, exon position of human > by using UCSC. > > Really thanks for your advice. Many thanks for your explanation in > detail previously. Very helpful and learn a lot from your > explanation. I'm very fresh at bioinformatics. > > Thanks. > > best regards Edge Research Student > _______________________________________________ Genome maillist - > [email protected] > https://lists.soe.ucsc.edu/mailman/listinfo/genome _______________________________________________ Genome maillist - [email protected] https://lists.soe.ucsc.edu/mailman/listinfo/genome
