Thanks to followup from Steve on the other thread, here's how to do what
I'm asking about here, in case somebody down the line finds this before
that:

Generate data from knownGene.

Add a filter.

In the filter page, activate the knownCanonical checkbox, and then click
the button at the very bottom of the page to reload the filter page with
the new set of fields.

Go back to the top and find the knownCanonical section and add a constraint
that the transcript _in that section_ must NOT match [n/a] (brackets not
included in actual form input).

Submit filter, and generate output.  Bob's your uncle.

Cheers,

-John

On Mon, Feb 6, 2012 at 3:18 PM, John Hawkins <[email protected]> wrote:

> Hi, all,
>
> A reframing of my earlier question*:
>
> (This question is in the context of the hg19 Human genome assembly.)
>
> How can I join the information in the knownGene and knownCanonical tables
> on transcript ID so that I get all the information in knownGene but
> restricted to the transcripts that co-occur?
>
> Is the best way to do this an "intersection"?  I've tried and failed thus
> far to make that meet my needs, but it sounds conceptually promising.
>
> Or perhaps I need to export a list of transcript IDs from knownCanonical
> and then feed it back in as a restriction?
>
> I know this is possible because somebody in my lab previously generated a
> .bed file of this data, but the details have been lost to the annals of
> history, and I'm failing to recreate the output either in bed format or in
> GTF format (my actual objective).
>
> Any guidance appreciated.
>
> Thanks!
>
> -John
>
> *Presumably too vague, given the lack of response.
>



-- 
"mFphrrrbm hoAmn rggh, hnllmmgh rnbnrrl."  --Demosthenes, "The Early Works"
_______________________________________________
Genome maillist  -  [email protected]
https://lists.soe.ucsc.edu/mailman/listinfo/genome

Reply via email to