[BioC] Gene-Metabolite correlation network

Sean Davis sdavis2 at mail.nih.gov
Sat Mar 3 13:23:03 CET 2012


On Fri, Mar 2, 2012 at 9:35 AM, B Usadel <b.usadel at bio1.rwth-aachen.de> wrote:
> This would be a crude solution but will do the job
> m<-matrix(rnorm(1000),nrow=100,ncol=10,dimnames=list(paste("met",c(1:100))))
> g<-matrix(rnorm(10000),nrow=1000,ncol=10,dimnames=list(paste("gen",c(1:1000))))

cm <- cor(t(m),t(g))

No need for the rbind, I don't think.

Sean


> mg<-rbind(m,g)
> cm<-cor(t(mg))
>
> res<-cm[1:100,101:1100]
>
> For networks, you can try the GeneNet package using the mixed matrix. Or
> just use res above with some sensible threshold and then make an
> adjacency matrix out of it and convert it into a graphNEL object.
>
> (This is all assuming Pearson correlation (or derivatives thereof) is
> the way to go between metabolites and transcripts)
>
> Cheers,
> -björn
>
>> Thank you Prof. Björn,
>>
>> Yes they are are  10 identical conditions. But two matrices  are of
>> not the same dimension. A has 1000 genes (rows) and B has 100
>> metabolites(rows).
>>
>> Is there any package available in Bio-conductor or R  for generating
>> gene -metabolite correlation network ?
>>
>> Regards,
>>
>> Pankaj2k3
>>
>> Pankaj Barah Department of Biology, Norwegian University of Science &
>> Technology (NTNU) Realfagbygget, N-7491 Trondheim, Norway Telephone:
>> (+47) 73 59 86 92 Mobile: (+47) 45063435 Fax: (+47) 73 59 61 00
>> E.mail: pankaj.barah at bio.ntnu.no
>> Homepage:http://www.ntnu.no/employees/pankaj.barah
>> ------------------------------------------------------------------------
>> *From:* B Usadel <b.usadel at bio1.rwth-aachen.de>
>> *To:* pankaj borah <pankajborah2k3 at yahoo.co.in>
>> *Cc:* "bioconductor at r-project.org" <bioconductor at r-project.org>
>> *Sent:* Friday, 2 March 2012 4:13 PM
>> *Subject:* Re: [BioC] Gene-Metabolite correlation network
>>
>> Hi Pankaj
>>
>> if it is the same 10 conditions, I would merge the matrices.
>> (Otherwise subset both matrices on the identical conditions)
>> And then calculate the whole correlation/covariance matrix and extract
>> the relevant regions, With this small number of genes, it will be done
>> almost instantaneously.
>>
>> Cheers,
>> björn
>>> Hi All,
>>>
>>> I have two matrices   A and B.
>>> A contains expression values for 1000 genes in 10 conditions  A[1:1000,1:10]
>>> B contains metabolic profiles of 100 metabolites in 10 conditions B[1:100,1:10]
>>>
>>> Is there a way that I can calculate gene-metabolite correlation OR co-variance  matrix using A and B ?
>>>
>>> Thanks
>>>
>>> Pankaj2k3
>>>      [[alternative HTML version deleted]]
>>>
>>>
>>> ------------------------------------------------------------------------
>>>
>>> _______________________________________________
>>> Bioconductor mailing list
>>> Bioconductor at r-project.org <mailto:Bioconductor at r-project.org>
>>> https://stat.ethz.ch/mailman/listinfo/bioconductor
>>> Search the archives: http://news.gmane.org/gmane.science.biology.informatics.conductor
>>
>>
>> --
>> Make the (Gabi) primary database better by just taking a short questionaire:
>> https://www.surveymonkey.com/s/RK3GMPC
>>
>> Prof. Dr. Björn Usadel
>> RWTH Aachen University
>> Institute for Biology I
>> Worringer Weg 1
>> 52056 Aachen
>>
>> phone: +49 (0)241 80 26634
>> fax:   +49 (0)241 80 22637
>> web:   http://www.usadellab.org
>> ---
>>
>>
>>
>
>
> --
> Make the (Gabi) primary database better by just taking a short questionaire:
> https://www.surveymonkey.com/s/RK3GMPC
>
> Prof. Dr. Björn Usadel
> RWTH Aachen University
> Institute for Biology I
> Worringer Weg 1
> 52056 Aachen
>
> Forschungszentrum Jülich
> IBG-2: Plant Sciences
> 52425 Jülich
> Germany
>
> phone: +49 (0)241 80 26634
> fax:   +49 (0)241 80 22637
> web:   http://www.usadellab.org
> ---
>
>
>        [[alternative HTML version deleted]]
>
>
> _______________________________________________
> Bioconductor mailing list
> Bioconductor at r-project.org
> https://stat.ethz.ch/mailman/listinfo/bioconductor
> Search the archives: http://news.gmane.org/gmane.science.biology.informatics.conductor



More information about the Bioconductor mailing list