[BioC] create Affymetrix .chp file for viewing in affy Integrated Genome Browser

Shi, Tao shidaxia at yahoo.com
Fri May 9 20:33:03 CEST 2008


Hi Steve,

I basically did a brutal-force probe-level analysis, so I could get around the large memory requirement of 'tilingArray' packages.  

In our case, we used tiling array like other expression array for differentially expressed gene identification.  I parsed the PM signals from the .cel files and did scaling, quantile normalization, then assessed the significance of signal change between the treatments vs. controls for each probe.  Nothing fancy.

I'm fully aware of the problems of this approach, but it's a start point for me.  All your questions are relevant but I just don't have time to consider them yet.  I was still using the .bpmap file Affy provides.

best,

...Tao



----- Original Message ----
From: Steve Lianoglou <mailinglist.honeypot at gmail.com>
To: "Shi, Tao" <shidaxia at yahoo.com>
Cc: bioconductor at stat.math.ethz.ch
Sent: Friday, May 9, 2008 10:40:06 AM
Subject: Re: [BioC] create Affymetrix .chp file for viewing in affy Integrated Genome Browser

Hi Tao,

I was curious if you took an approach in your analysis that borrowed  
from the BioC `tilingArray` package, or is your take something  
completely new/different?

I'm playing around with analyzing some tiling array data myself for  
the first time, and was curious how you approached some issues in the  
analysis, such as:

1) normalization of signal with respect to probe sequence
--> The process outlined in the tilingArray package for dealing with  
this sounds great, but I don't have genomic dna to use as a control.

2) dealing with probes that hit several parts of the genome

3) did you reblast the probe sequences to get a better idea of where  
they map to the latest reference genome?

Just curious,
-steve


On May 8, 2008, at 6:01 PM, Shi, Tao wrote:

> Hi Kasper,
>
> Thanks for the reply!
>
> Actually, I just found out that .sgr format, which is just a simple  
> text with columns of coordinate and score, will do.
>
> Thanks a lot.
>
> ...Tao
>
>
>
>
> ----- Original Message ----
> From: Kasper Daniel Hansen <khansen at stat.Berkeley.EDU>
> To: "Shi, Tao" <shidaxia at yahoo.com>
> Cc: bioconductor at stat.math.ethz.ch
> Sent: Thursday, May 8, 2008 12:28:19 PM
> Subject: Re: [BioC] create Affymetrix .chp file for viewing in affy  
> Integrated Genome Browser
>
> There are two my mind no facilities for this. The natural place for
> such functions would be affxparser (which I maintain) and right now we
> have no plans for such an addition).
>
> But IGB supports many different formats, most of which are super easy
> to write using write.table. For examples BED files (I think they are
> called). Look in the IGB manual for further info. There might even be
> a package that does this for you.
>
> Kasper
>
> On May 8, 2008, at 10:47 AM, Shi, Tao wrote:
>
>> Hi list,
>>
>> I have my customized analysis
>> results of a tiling array data set (signal changes, p values, and
>> their
>> corresponding chormosomal coordinates). Are there ways to package  
>> them
>> into Affy .chp files so I can to view them in IGB?  Or if you know
>> any other applications that function like IGB, please also let me
>> know.
>>
>> Thanks,
>>
>> ...Tao
>>
>>
>>
>> ____________________________________________________________________________________
>>
>> [[elided Yahoo spam]]
>>    [[alternative HTML version deleted]]
>>
>> _______________________________________________
>> Bioconductor mailing list
>> Bioconductor at stat.math.ethz.ch
>> https://stat.ethz.ch/mailman/listinfo/bioconductor
>> Search the archives: http://news.gmane.org/gmane.science.biology.informatics.conductor
>
>
>      
> ____________________________________________________________________________________
>
> [[elided Yahoo spam]]
>
> _______________________________________________
> Bioconductor mailing list
> Bioconductor at stat.math.ethz.ch
> https://stat.ethz.ch/mailman/listinfo/bioconductor
> Search the archives: http://news.gmane.org/gmane.science.biology.informatics.conductor


      ____________________________________________________________________________________

[[elided Yahoo spam]]



More information about the Bioconductor mailing list