How to extract sequences from multz sequence alignment on UCSC genome browser?

Question

[epigene]

I want to do some realignment of a segment of the genome that show conservation between different species (human, zebrafish, mouse, rat,etc). I know the genomic coordinate of the human region and if I just view the region on human in UCSC genome browser, I can see the Multiz sequence alignment track. What I need is a way to extract the sequences that contribute to the sequence alignment in the view range. I looked around but couldn't find a function to do this. I'm wondering what's the traditional way of doing this?

I can think of using liftover to find corresponding syntenic regions of the human sequence and do this for each species. Is this the right way to go? Are there some tools out there that can do it more efficiently?

Thanks!

Answer 1

[Denise CS]

In Ensembl we have got several sets of multiple whole genome alignments (e.g. for primates, birds and reptiles, vertebrates, and eutherian mammals). For the latter we have sequences from the species you mentioned plus 13 other genomes. You can view them on the browser and download/export regions from there. You can export them as FASTA, MEGA, CLUSTALW, etc...They are also available from our FTP, REST and Perl APIs. Be aware that some of the species may have no alignment in the region you are interested in.

[epigene]

thanks! I'm not familiar with Ensembl browser, where can I access and download the alignments on the browser?

[Denise CS]

Go to ensembl.org, select the species e.g. human and enter the region you are interested in (note the format should be like this 17:63973115-64437414) in one of the search boxes. Hit 'Go'. You are now in the Location tab. In the left hand side menu there is the link called 'Alignments (text)'. Click on it, select the alignment (e.g. 17 eutherian mammals EPO) and then click Go. Once you select which block alignment to view, you then click on 'Download the alignment' button and select the format you wish to export it (e.g. FAST, etc...). More details on our 'Alignments (text)' can be found in our help page. More details on the comparative genomics aspect of Ensembl can be found in this tutorial video. More details on the Download feature are depicted in our blog post.

[siddharth.sethi5]

Hi Denise, Can you please tell me how do this using perl API. I am trying to use the code from compara tutorials, but I do not want pairwise alignments , as extracted by this fetch_by_method_link_type_GenomeDBs('LASTZ_NET'). I want the alignments in multiple species , for example for 17 eutherian mammals EPO. How to do this using API. I cannot find the method name to use for this instead of lastz_net. Thanks Sid

Answer 2

[Ram]

I don't see a way to do it just for the view region (not without multiple clicks per view regions at least). Would it be feasible to download the multiz for the entire chromosome and restrict it by co-ordinate, then extract individual sequences using sequence analysis tools?

[epigene]

not sure if this is feasible. I will check it out.