how to combine p values?
1
0
Entering edit mode
Reema Singh ▴ 570
@reema-singh-4373
Last seen 10.1 years ago
Hi All I want to combine the two datasets using p-values. Can anybody tell me how to combine p values from two different datasets using R? It would be a great help Waiting for reply.... REGARDS~ Reema Singh [[alternative HTML version deleted]]
• 3.9k views
ADD COMMENT
0
Entering edit mode
@moshe-olshansky-2329
Last seen 10.1 years ago
Hi Reema, there are several ways to do so (you can search the Internet). I am not sure whether there exists a really good one. One possibility is as below: let P1 and P2 be the p-values from dataset 1 and dataset 2. Under the null hypothesis both p-values are distributed uniformly in [0,1]. If we assume independence then the probability that their product is less than x equals x - x*log(x) (where log is the natural - base e - logarithm). So your combined p-value is P1*P2 - P1*P2*log(P1*P2). Best regards, Moshe. --- On Thu, 9/12/10, Reema Singh <reema28sep at="" gmail.com=""> wrote: > From: Reema Singh <reema28sep at="" gmail.com=""> > Subject: [BioC] how to combine p values? > To: bioconductor at r-project.org > Received: Thursday, 9 December, 2010, 8:50 PM > Hi All > > I want to combine the two datasets using p-values. > Can anybody tell me how to combine p values from two > different datasets > using R? > > It would be a great help > > Waiting for reply.... > > REGARDS~ > Reema Singh > > ??? [[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 >
ADD COMMENT
0
Entering edit mode
There are several methods to combine p-values. Fisher method, Z-transform method (also called Stouffer method), etc.The paper below explains the methods. Whitlock, M. C. (2005) "Combining probability from independent tests: the weighted Z-method is superior to Fisher’s approach", J. Evol. Biol., 18, pages 1368–1373. You can download the package called "survcomp" ( http://cran.r-project.org/web/packages/survcomp/). This package contains Fisher method using "combine.test" routine. You can also refer the following paper for methods: http://statgen.ncsu.edu/zaykin/tpm/tpmfinal.pdf Good luck, ~Nikhil On Thu, Dec 9, 2010 at 6:37 PM, Moshe Olshansky <m_olshansky@yahoo.com>wrote: > Hi Reema, > > there are several ways to do so (you can search the Internet). I am not > sure whether there exists a really good one. > One possibility is as below: > let P1 and P2 be the p-values from dataset 1 and dataset 2. Under the null > hypothesis both p-values are distributed uniformly in [0,1]. If we assume > independence then the probability that their product is less than x equals x > - x*log(x) (where log is the natural - base e - logarithm). So your combined > p-value is P1*P2 - P1*P2*log(P1*P2). > > Best regards, > Moshe. > > --- On Thu, 9/12/10, Reema Singh <reema28sep@gmail.com> wrote: > > > From: Reema Singh <reema28sep@gmail.com> > > Subject: [BioC] how to combine p values? > > To: bioconductor@r-project.org > > Received: Thursday, 9 December, 2010, 8:50 PM > > Hi All > > > > I want to combine the two datasets using p-values. > > Can anybody tell me how to combine p values from two > > different datasets > > using R? > > > > It would be a great help > > > > Waiting for reply.... > > > > REGARDS~ > > Reema Singh > > > > [[alternative HTML version deleted]] > > > > _______________________________________________ > > Bioconductor mailing list > > Bioconductor@r-project.org > > https://stat.ethz.ch/mailman/listinfo/bioconductor > > Search the archives: > http://news.gmane.org/gmane.science.biology.informatics.conductor > > > > _______________________________________________ > Bioconductor mailing list > Bioconductor@r-project.org > https://stat.ethz.ch/mailman/listinfo/bioconductor > Search the archives: > http://news.gmane.org/gmane.science.biology.informatics.conductor > [[alternative HTML version deleted]]
ADD REPLY
0
Entering edit mode
Thank you for references, Paul and Nikhil! --- On Fri, 10/12/10, Nikhil Garge <nikhil.garge@gmail.com> wrote: From: Nikhil Garge <nikhil.garge@gmail.com> Subject: Re: [BioC] how to combine p values? To: "Moshe Olshansky" <m_olshansky@yahoo.com>, bioconductor@r-project.org, "Reema Singh" <reema28sep@gmail.com> Received: Friday, 10 December, 2010, 12:25 PM There are several methods to combine p-values. Fisher method, Z-transform method (also called Stouffer method), etc.The paper below explains the methods. Whitlock, M. C. (2005) "Combining probability from independent tests: the weighted Z-method is superior to Fisher’s approach", J. Evol. Biol., 18, pages 1368–1373. You can download the package called "survcomp" (http://cran.r-project.org/web/packages/survcomp/). This package contains Fisher method using "combine.test" routine. You can also refer the following paper for methods: http://statgen.ncsu.edu/zaykin/tpm/tpmfinal.pdf Good luck,~Nikhil On Thu, Dec 9, 2010 at 6:37 PM, Moshe Olshansky <m_olshansky@yahoo.com> wrote: Hi Reema, there are several ways to do so (you can search the Internet). I am not sure whether there exists a really good one. One possibility is as below: let P1 and P2 be the p-values from dataset 1 and dataset 2. Under the null hypothesis both p-values are distributed uniformly in [0,1]. If we assume independence then the probability that their product is less than x equals x - x*log(x) (where log is the natural - base e - logarithm). So your combined p-value is P1*P2 - P1*P2*log(P1*P2). Best regards, Moshe. --- On Thu, 9/12/10, Reema Singh <reema28sep@gmail.com> wrote: > From: Reema Singh <reema28sep@gmail.com> > Subject: [BioC] how to combine p values? > To: bioconductor@r-project.org > Received: Thursday, 9 December, 2010, 8:50 PM > Hi All > > I want to combine the two datasets using p-values. > Can anybody tell me how to combine p values from two > different datasets > using R? > > It would be a great help > > Waiting for reply.... > > REGARDS~ > Reema Singh > >     [[alternative HTML version deleted]] > > _______________________________________________ > Bioconductor mailing list > Bioconductor@r-project.org > https://stat.ethz.ch/mailman/listinfo/bioconductor > Search the archives: http://news.gmane.org/gmane.science.biology.informatics.conductor > _______________________________________________ Bioconductor mailing list Bioconductor@r-project.org https://stat.ethz.ch/mailman/listinfo/bioconductor Search the archives: http://news.gmane.org/gmane.science.biology.informatics.conductor [[alternative HTML version deleted]]
ADD REPLY
0
Entering edit mode
I found, A systematic comparison of methods for combining p-values from independent tests , Thomas M. Loughin, Computational Statistics & Data Analysis 47 (2004) 467 – 485 a useful paper. I have used some of the methods in there with exon arrays If you enquiring about Meta analysis there is an R program called GeneMeta, and a another called "metal" (perl program I think). Metal is used for GWAS typically. Cheers Paul -----Original Message----- From: Moshe Olshansky <m_olshansky@yahoo.com> To: bioconductor@r-project.org, Reema Singh <reema28sep@gmail.com> Subject: Re: [BioC] how to combine p values? Date: Thu, 9 Dec 2010 15:37:03 -0800 Hi Reema, there are several ways to do so (you can search the Internet). I am not sure whether there exists a really good one. One possibility is as below: let P1 and P2 be the p-values from dataset 1 and dataset 2. Under the null hypothesis both p-values are distributed uniformly in [0,1]. If we assume independence then the probability that their product is less than x equals x - x*log(x) (where log is the natural - base e - logarithm). So your combined p-value is P1*P2 - P1*P2*log(P1*P2). Best regards, Moshe. --- On Thu, 9/12/10, Reema Singh <reema28sep@gmail.com> wrote: > From: Reema Singh <reema28sep@gmail.com> > Subject: [BioC] how to combine p values? > To: bioconductor@r-project.org > Received: Thursday, 9 December, 2010, 8:50 PM > Hi All > > I want to combine the two datasets using p-values. > Can anybody tell me how to combine p values from two > different datasets > using R? > > It would be a great help > > Waiting for reply.... > > REGARDS~ > Reema Singh > > [[alternative HTML version deleted]] > > _______________________________________________ > Bioconductor mailing list > Bioconductor@r-project.org > https://stat.ethz.ch/mailman/listinfo/bioconductor > Search the archives: http://news.gmane.org/gmane.science.biology.informatics.conductor > _______________________________________________ Bioconductor mailing list Bioconductor@r-project.org https://stat.ethz.ch/mailman/listinfo/bioconductor Search the archives: http://news.gmane.org/gmane.science.biology.informatics.conductor [[alternative HTML version deleted]]
ADD REPLY

Login before adding your answer.

Traffic: 792 users visited in the last hour
Help About
FAQ
Access RSS
API
Stats

Use of this site constitutes acceptance of our User Agreement and Privacy Policy.

Powered by the version 2.3.6