Question: Overlap of Gene Lists from 2 Different Platforms
gravatar for tucanj
4.2 years ago by
tucanj70 wrote:

I have a list of differentially expressed probesets on one microarray platform. Many of these probesets map to more than one transcript, and there are some transcripts measured by more than one probeset on the platform (many to many mapping).

I have performed differential expression analysis at the probeset level on another platform, with all probesets only mapping to one transcript, but some transcripts measured by multiple probesets (many probes to one transcript). These 2 platforms do not measure the same population of transcripts but have some overlap.

I am interested to see if the lists of differentially expressed transcripts (with concordant direction), as determined by mapping from differentially expressed probesets on each platform, significantly overlap between the two platforms. Is there any way to do this with the current experiment design (without a 1 to 1 mapping)?

R • 1.2k views
ADD COMMENTlink written 4.2 years ago by tucanj70

What are the two platforms? Show some data.

ADD REPLYlink written 4.2 years ago by komal.rathi3.4k

Platforms are Affymetrix U133 Plus 2.0 and a custom printed microarray by a different lab. There are 259 differentially expressed probesets on U133Plus2 mapping to 458 Refseq transcripts. The custom array measures 15539 refseq IDs with 20 035 probesets. There are 4790 probesets differentially expressed on the custom array representing 4201 Refseq transcripts. There are 112 probesets on the custom array representing a differentially expressed transcript from the U133Plus2 differential list that are also differentially expressed. 83 of these probesets show concordant direction of fold change as the transcript they are measuring from U133Plus2. Let me know what else is needed. Thanks!

ADD REPLYlink modified 4.2 years ago • written 4.2 years ago by tucanj70

If it is custom printed microarray, then how do you expect anyone to match the probe IDs in the two microarrays without knowing the platform. The only way you can match is by gene name or if you have Ensembl ID (or any other ID that is common between those two!!).

ADD REPLYlink written 4.2 years ago by komal.rathi3.4k

I am not asking to match the probes - I have done that with Refseq IDs. The problem is the many to many mapping and many to one mapping for computing overlap.

ADD REPLYlink written 4.2 years ago by tucanj70
Please log in to add an answer.


Use of this site constitutes acceptance of our User Agreement and Privacy Policy.
Powered by Biostar version 2.3.0
Traffic: 883 users visited in the last hour