Hello,
BLAST, sometimes, does not propose direct links to Genbank pages as reports, and this leads to unaccepted entries in the UCSC Genome Browser's search term box. How could I avoid those results? Is there a way to interpret the LOCUS field as unsuitable for UCSC Genome Browser? Hope this is a clear question.
I assume, since you refer to links, that you are running BLAST online and viewing the results in a web browser. In this case, you cannot "avoid" results: your options are restricted to those supplied by whoever set up the web interface. If you need more flexibility, you need to run BLAST locally and write a parser for the output.
Similarly, the way to figure out which types of sequence IDs are recognised at UCSC genome browser is to try out different types and make a note of which do and do not work.
If your definition of "valid" is an accession that can be found in the UCSC browser, then one cannot know the answer to this a priori. UCSC maps the sequences associated with accessions to the genome. For those sequences that do not map, there will be no record of them and they will appear to be "invalid". There are some obvious ones that could be avoided (and I think XM records are an example), but perfectly valid genbank identifiers may not be found at UCSC since some of them do not map to the genome. So, knowing the "class" of identifier will only get you part way.
An exaple is this: http://blast.ncbi.nlm.nih.gov/Blast.cgi
The 5th transcript's locus (XM_001714378) could not be located in UCSC Genome Browser. That is OK, but I need to know how to avoid those results when running BLAST online using a script. I thought I could select the loci understanding which type of ID the UCSC Genome Browser accepts in its search term box.
As neilfws suggests, one way is to try every type of ID. I think I will solve the problem this way. Thank you all.
@ Sean Davis: Yes, I am trying to make an application to locate sequences on USCS Genome Browser starting from ESTs: every time BLAST finds somewhat "strange", obviously, UCSC Genome Browser cannot locate the sequence. Hence, I would like to teach the BLAST's script how to avoid result unrecognised by UCSC Genome Browser. I think I will try every type of ID, solving only part of my problems.
Use of this site constitutes acceptance of our User Agreement and Privacy Policy.
version 2.3.6
can you give an example of such BLAST result please ?
I understand the question (I think), but I'm not sure what relevance it has to your real problem. Could you elaborate on what you are really trying to do? Are you writing a web application or something else?
@wiee, you are not using BioStar correctly. You seem to be creating a new account every time you post, so you now have at least 3 different accounts on BioStar - can you please use a single account for all your posts in future. Also, if you want to provide additional information, please either (a) update your original question, or (b) post the updates as comments.