[BioC] Conversion of pfm to pwm using PWM()

[Hervé Pagès]

Hi Ravi, Steve,

Yes the "PWM" method for matrix objects is new in BioC 2.8 (R-2.13).
It computes a Position Weight Matrix from a Position Frequency Matrix
using the algorithm described in the Wasserman & Sandelin paper.
See ?PWM for the details.

Cheers,
H.


On 11-05-26 08:33 PM, Steve Lianoglou wrote:
> Hi,
>
> On Thu, May 26, 2011 at 6:51 PM, Ravi Karra<ravi.karra at gmail.com>  wrote:
>> Hi,
>> I am trying to find TF binding sites in a set of promoters using position frequency matrices (pfm's) from jaspar and transfac.  I have gotten this to work before on my machine at home, but cannot seem to get this to run on campus.  I am not sure why I get the following error as I thought PWM can handle integer pfm's.  Any thoughts?
>>
>> Thanks in advance,
>> Ravi
>>
>> Code:
>>> library (Biostrings)
>>> pfm
>>   [,1] [,2] [,3] [,4] [,5] [,6]
>> A   13    0   50    1   13    6
>> C   14    0    2    0    6   16
>> G   15   53    0    1   25   22
>> T   11    0    1   51    9    9
>>> typeof (pfm)
>> [1] "integer"
>>> pwm = pfm (PWM)
>> Error: could not find function "pfm"
>
> Were you just warming up your keyboard for the next step, or .. ? :-)
>
>>> library (Biostrings)
>>> pfm
>>   [,1] [,2] [,3] [,4] [,5] [,6]
>> A   13    0   50    1   13    6
>> C   14    0    2    0    6   16
>> G   15   53    0    1   25   22
>> T   11    0    1   51    9    9
>>> typeof (pfm)
>> [1] "integer"
>>> PWM (pfm)
>> Error in function (classes, fdef, mtable)  :
>>   unable to find an inherited method for function "PWM", for signature "matrix"
>>> sessionInfo ()
>> R version 2.12.0 (2010-10-15)
>> Platform: x86_64-unknown-linux-gnu (64-bit)
>
> I'm not sure what to say, but this works for me.
>
> You can look at the output of `showMethods('PWM'), R 2.13/bioc-2.8 has
> a PWM method defined for matrix objects:
>
> R>  showMethods('PWM')
> Function: PWM (package Biostrings)
> x="character"
> x="DNAStringSet"
> x="matrix"
>
> Maybe you have to upgrade to the latest version of R/bioc? Sorry, I
> don't have a version of R 2.12 lying around to test further ... you
> say it worked for you at home, maybe your cpu at home is running R
> 2.13?
>
> HTH,
>
> -steve
>


-- 
Hervé Pagès

Program in Computational Biology
Division of Public Health Sciences
Fred Hutchinson Cancer Research Center
1100 Fairview Ave. N, M1-B514
P.O. Box 19024
Seattle, WA 98109-1024

E-mail: hpages at fhcrc.org
Phone:  (206) 667-5791
Fax:    (206) 667-1319