Hi Greg, Thank you for your reply. It is very helpful. I am trying to build the chainFilter and netFilter from source. Could you please let me know which directory these two tools are under? I checked kent/src/utils but can't find them there. Thank you.
- Jia ________________________________________ From: Greg Roe [[email protected]] Sent: Wednesday, April 25, 2012 4:32 PM To: Zeng, Jia Cc: [email protected] Subject: Re: [Genome] How to interpret chain & net files in light of orthology Hi Jia, I would start by reading the help docs on chains and nets: http://genome.ucsc.edu/goldenPath/help/chain.html http://genome.ucsc.edu/goldenPath/help/net.html Those explain chains and nets and how the files are constructed. I'm not totally sure what you are expecting to see but are not. For example, if I click on the net showing at the position you listed (chr7:25681793-25788913), it has an ID of 5. If I look in the rbest.net file, I find the line corresponding to that, here: fill 1730704 157395378 chr7 + 370509 158704532 id 5 score 11940049674 ali 135239485 qDup 1095596 type top If you are comfortable building some of our utilities from source (http://hgdownload.cse.ucsc.edu/downloads.html#source_downloads), here are command lines that extract the rbest chain that overlaps that region and the complete structure of the enclosing rbest net: chainFilter hg19.panTro3.rbest.chain.gz -t=chr7 -tOverlapStart=25681792 -tOverlapEnd=25788913 > HOXA_region.chain netFilter hg19.panTro3.rbest.net.gz -t=chr7 -tOverlapStart=25681792 -tOverlapEnd=25788913 > HOXA_region.net chainFilter and netFilter both have many other options described in their usage messages. For understanding the positive and negative strand entries, you may find this previous mailing list answer helpful: https://lists.soe.ucsc.edu/pipermail/genome/2011-November/027526.html I hope that helps! Please let us know if you have any additional questions: [email protected]<mailto:[email protected]> - Greg Roe UCSC Genome Bioinformatics Group On 4/23/12 2:35 PM, Zeng, Jia wrote: Hi, I'm trying to infer orthologous regions between two different species using the chain and/or net file provided by UCSC. To understand how to do this, I decided to use a well conserved region (the HOXA cluster from human genome) as inquiry. This region's coordinate (panning from HOXA1 to HOXA13) is chr7:25681793-25788913 in hg19. Due to HOX genes' high conservation level, I expect there to be a corresponding region in the Chimp genome. To check whether this is indicated by the chain/net files, I downloaded the hg19.panTro3.rbest.chain and hg19.panTro3.rbest.net files. For some reason, I didn't locate any chain/net that overlap with the aforementioned region although if I input the said region on genome browser and turned on the comparative genomics track. it seems that the Chimp Alignment Net does exist for this region. I suspect that I may have some fundamental misunderstanding about the files of interest and I also am not very clear about the practical difference between the rbest.chain and rbest.net files. Your input will be greatly appreciated. Best, - Jia _______________________________________________ Genome maillist - [email protected]<mailto:[email protected]> https://lists.soe.ucsc.edu/mailman/listinfo/genome _______________________________________________ Genome maillist - [email protected] https://lists.soe.ucsc.edu/mailman/listinfo/genome
