Hi, Daniel (and James)

Thanks for your suggestion. It's what I am trying :)

I understand that this current behavior might be intentional. While you would embed this option into your future version, I'd like to suggest you could add some comments/help on the Intersection page to indicate the current behavior. Otherwise, it might be misleading.

Thanks

Xianjun

Daniel Blankenberg wrote:

Hi Xianjun,

In the mean time, you can filter the datasets by each strand (Filter by column tool), perform the operation on the datasets that are on the same strands, then concatenate (under text manipulation) the resultant datasets back together. 

Thanks for using Galaxy,

Dan

On May 5, 2010, at 12:45 PM, James Taylor wrote:

The behavior of this tool is intentional. Regardless of which direction a feature is encoded in, we consider to features having any overlapping DNA to intersect.

However, it would be good for this tool to have an option to handle your use case (only consider intervals overlapping if on the same strand).

On May 5, 2010, at 12:36 PM, Xianjun Dong wrote:

the two forward (+) regions in first query should be onlyintersected with the forward (+) regions in second query, NOT regions on reverse strand. That's my understanding to intersection. Isn't right? From the result, it seems that each region in first query is intersected with each region in second query (NO MATTER its strand). I think if my understanding is correct, this should be correct soon, since intersection is such a widely used tool in Galaxy.

_______________________________________________
galaxy-user mailing list
galaxy-user@lists.bx.psu.edu
http://lists.bx.psu.edu/listinfo/galaxy-user

-- 
==========================================
Xianjun Dong
PhD student, Lenhard group
Computational Biology Unit
Bergen Center for Computational Science
University of Bergen
Hoyteknologisenteret, Thormohlensgate 55
N-5008 Bergen, Norway
E-mail: xianjun.dong@bccs.uib.no
Tel.: +47 555 84022
Fax : +47 555 84295
==========================================