query builder - from genes to deficiency lines?

[Geo]

Hi!

I'm struggling with the Query Builder. I can easily get a list of genes linked to e.g. a compartiment or function, but I seem to be unable to get a list of all e.g. Exelixis stocks that carry a deletion in either of these genes. Shouldn't this be possible using Query Builder and if so, how would that work?

Thanks for any suggestion,

Geo

[strelets]

Hi Geo,

Problem is that genes are not directly crossreferenced
to aberrations, therefore query like that should be
performed in two independent steps.

Let's assume you are interested in some gene (dpp).

Create QueryBuilder (QB) leg, select dataset Alleles,
select field Gene and enter context dpp.
Create another QB leg, select dataset Aberrations,
select field class_of_aberration' and from dictionary
select contect chromosomal_deletion.
Run this query asking for aberrations, you get 5 matches.
So you searched for aberrations which are deletions and
which are crossreferenced to some alleles of gene dpp.

Now in HitList click on HitList Conversion Tools
button. Select 'export to new QB session'. Your
deletions will be uploaded as a leg in new QB query.
Run this query asking for 'related stocks', that's all.

Regards,

Victor

[Geo]

Hi, Victor!

Thank you for your reply. It finally worked, but not without using field 'type' instead of field 'class_of_aberration'. That's kind of strange. Any idea why?

Best,

Geo

[Geo]

Although it works for the given example, I don't get it run for others.

Another example: The Flybase report page of cdc42 says it is disrupted in Df(1)Exel6253. How can I retrieve this information using QB?

Best,

Geo

[Boris]

Hi Geo,

Do you insist in using QB? For queries of a certain complexity, I usually prefer FlyMine (www.flymine.org) for its convenience. Seem to remember they do have the Exelixis lines in their data collection.

Best,
Boris

[Geo]

Hi, Boris!

Not really insisting - but that's the first thing that came into my mind.

So, coming back to my original question. I found a solution to get from a set of genes to a sublist containing only those which have an Exelixis Df entry (using the Ryder cross-reference FBrf0184335 as second leg). That's already something (kind of positive control). I'm still struggeling on how to get from this list of genes towards a list of the according Exelixis stocks. I can also get the Exelixis stock list (chromosomal deletions + bloomington stock + Ryder PubmedID). With "show me aberrations" it will list all the FBabXXXXXX, and each of these entries contains the list of deleted genes. The intersection of both lists is what I want. But there is no way to access the last piece of information (genes deleted in FBabXXXX entries) from within Query Builder.

Any ideas how to proceed?

Geo