Influence of expression correlation on false positive ratio
1
0
Entering edit mode
@january-weiner-3999
Last seen 10.3 years ago
Hello, statistical methods for assessing significance of differences in expression assume, correct me if I'm wrong, independence of the tests. Does anyone have at hand any papers on the performance -- in terms of type I error -- of methods such as limma / eBayes? I'm sure this issue has been investigated in depth. Kind regards, January -- -------- Dr. January Weiner 3 -------------------------------------- Max Planck Institute for Infection Biology Charit?platz 1 D-10117 Berlin, Germany Web : www.mpiib-berlin.mpg.de Tel : +49-30-28460514 Fax : +49-30-28450505
limma limma • 1.3k views
ADD COMMENT
0
Entering edit mode
Jeff Leek ▴ 650
@jeff-leek-5015
Last seen 3.8 years ago
United States
Hi January, If the tests are only dependent in small groups, say because genes are grouped into small modules, then most FDR methods in the p.adjust() function or the methods in the qvalue package will work. The Bonferroni correction controls a more conservative error rate, but also holds under dependence. If the sources of dependence are more pervasive, like if there are batch effects: http://www.nature.com/nrg/journal/v11/n10/full/nrg2825.html Then you can either use the batch correction methods in Limma if, say, you know the date the samples were processed. Or, if you don't know the sources of large scale dependence, you can use the sva package: http://www.bioconductor.org/packages/devel/bioc/html/sva.html which implements the methods described here: http://www.pnas.org/content/early/2008/11/24/0808709105.abstract Best, Jeff On Jul 9, 2012 7:08 AM, "January Weiner" <january.weiner@mpiib- berlin.mpg.de=""> wrote: > Hello, > > statistical methods for assessing significance of differences in > expression assume, correct me if I'm wrong, independence of the tests. > Does anyone have at hand any papers on the performance -- in terms of > type I error -- of methods such as limma / eBayes? I'm sure this issue > has been investigated in depth. > > Kind regards, > > January > > -- > -------- Dr. January Weiner 3 -------------------------------------- > Max Planck Institute for Infection Biology > Charitéplatz 1 > D-10117 Berlin, Germany > Web : www.mpiib-berlin.mpg.de > Tel : +49-30-28460514 > Fax : +49-30-28450505 > > _______________________________________________ > 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 COMMENT
0
Entering edit mode
Hi Jeff, many thanks. > If the tests are only dependent in small groups, say because genes are > grouped into small modules, then most FDR methods in the p.adjust() > function or the methods in the qvalue package will work. Yes, but I wonder how it behaves in light of a more extensive co-expression network (see for example the 2003 Stuart paper in Science, http://www.sciencemag.org/content/302/5643/249.short). Has anyone tried to simulate this? The co-expression modules, as have been found in many papers, are sometimes anything but small. > The Bonferroni > correction controls a more conservative error rate, but also holds under > dependence. Sure, Bonferroni does not assume independence of the test, but it's meager power means that many are not even considering this as an option (I definitely use it when the test is strictly in hypothesis testing mode and not further experiments are planned). Best regards, j. > > If the sources of dependence are more pervasive, like if there are batch > effects: > > http://www.nature.com/nrg/journal/v11/n10/full/nrg2825.html > > Then you can either use the batch correction methods in Limma if, say, you > know the date the samples were processed. Or, if you don't know the sources > of large scale dependence, you can use the sva package: > > http://www.bioconductor.org/packages/devel/bioc/html/sva.html > > which implements the methods described here: > > http://www.pnas.org/content/early/2008/11/24/0808709105.abstract > > > Best, > > > Jeff > > > > On Jul 9, 2012 7:08 AM, "January Weiner" > <january.weiner at="" mpiib-berlin.mpg.de=""> wrote: >> >> Hello, >> >> statistical methods for assessing significance of differences in >> expression assume, correct me if I'm wrong, independence of the tests. >> Does anyone have at hand any papers on the performance -- in terms of >> type I error -- of methods such as limma / eBayes? I'm sure this issue >> has been investigated in depth. >> >> Kind regards, >> >> January >> >> -- >> -------- Dr. January Weiner 3 -------------------------------------- >> Max Planck Institute for Infection Biology >> Charit?platz 1 >> D-10117 Berlin, Germany >> Web : www.mpiib-berlin.mpg.de >> Tel : +49-30-28460514 >> Fax : +49-30-28450505 >> >> _______________________________________________ >> 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 -- -------- Dr. January Weiner 3 -------------------------------------- Max Planck Institute for Infection Biology Charit?platz 1 D-10117 Berlin, Germany Web : www.mpiib-berlin.mpg.de Tel : +49-30-28460514 Fax : +49-30-28450505
ADD REPLY
0
Entering edit mode
hi, On 07/09/2012 01:53 PM, January Weiner wrote: > Hi Jeff, many thanks. > >> If the tests are only dependent in small groups, say because genes are >> grouped into small modules, then most FDR methods in the p.adjust() >> function or the methods in the qvalue package will work. > > Yes, but I wonder how it behaves in light of a more extensive > co-expression network (see for example the 2003 Stuart paper in > Science, http://www.sciencemag.org/content/302/5643/249.short). Has > anyone tried to simulate this? The co-expression modules, as have been > found in many papers, are sometimes anything but small. i guess a simple way to simulate this kind of situation would be sampling multivariate normal observations from a covariance matrix with a specific pattern of conditional independences reflected on its inverse, this can be accomplished with the function qpG2Sigma() from the Bioconductor package 'qpgraph'. i'll work out a very simple example with 1000 genes out of which 100 are differentially expressed (DE): library(limma) ## for DE analysis: lmFit(), etc. library(mvtnorm) ## for rmvnorm() p <- 1000 ## number of genes pDE <- 100 ## number of DE genes ## let's say all genes are truly mutually independent and simulate ## observations for two groups of samples with differences in means ## (4 vs 5) in the first 100 genes set.seed(123) y_c1 <- rmvnorm(n=20, mean=rep(4, p), sigma=diag(p)) y_c2 <- rmvnorm(n=20, mean=c(rep(5, pDE), rep(4, p-pDE)), sigma=diag(p)) y <- t(rbind(y_c1, y_c2)) rownames(y) <- paste0("g", 1:p) dim(y) [1] 1000 40 ## find DE genes using limma design <- cbind(grp1=1, grp2vs1=c(rep(0, 20), rep(1, 20))) fit <- lmFit(y, design) fit <- eBayes(fit) summary(decideTests(fit)) ## number DE genes at 5% FDR grp1 grp2vs1 -1 0 0 0 0 953 1 1000 47 tt <- topTable(fit, coef=2, p.value=0.05, n=Inf) ## how many of the previous 47 genes are truly DE? sum(as.integer(gsub("g", "", tt$ID)) <= 100) [1] 47 ## now let's simulate the setting with dependence among genes library(graph) ## for randomEgraph() library(qpgraph) ## for qpG2Sigma() set.seed(123) ## simulate an E-R random graph with 5% density G <- randomEGraph(V=as.character(1:p), p=0.05) ## simulate a covariance matrix whose inverse contains zeroes ## on the row and column indexed by missing edges in G and ## edges have a mean marginal Pearson correlation rho=0.5 set.seed(123) Sigma <- qpG2Sigma(G, rho=0.5, verbose=TRUE) ## some 2 or 3 minutes ## just check that we get those 0s in the inverse covariance ## for the missing edges of G ... invSigma <- solve(Sigma) mean(invSigma[upper.tri(invSigma) & as(G, "matrix") == 0]) [1] -4.836399e-07 ## ... and the mean marginal correlation is rho=0.5 as specified ## in the call to qpG2Sigma() mean(cov2cor(as.matrix(Sigma))[upper.tri(Sigma) & as(G, "matrix")]) [1] 0.501911 ## simulate data again under this pattern of dependences and the ## previous grouping of samples and DE gene distribution set.seed(123) y_c1 <- rmvnorm(n=20, mean=rep(4, p), sigma=as.matrix(Sigma)) y_c2 <- rmvnorm(n=20, mean=c(rep(5, pDE), rep(4, p-pDE)), sigma=as.matrix(Sigma)) y <- t(rbind(y_c1, y_c2)) rownames(y) <- paste0("g", 1:p) dim(y) [1] 1000 40 ## find DE genes using limma design <- cbind(grp1=1, grp2vs1=c(rep(0, 20), rep(1, 20))) fit <- lmFit(y, design) fit <- eBayes(fit) summary(decideTests(fit)) grp1 grp2vs1 -1 0 0 0 0 995 1 1000 2 i.e., under dependence, a classical DE analysis assuming independence among genes can become overly conservative, at least with this very straightforward simulated data and this very specific dependence pattern. robert. >> The Bonferroni >> correction controls a more conservative error rate, but also holds under >> dependence. > > Sure, Bonferroni does not assume independence of the test, but it's > meager power means that many are not even considering this as an > option (I definitely use it when the test is strictly in hypothesis > testing mode and not further experiments are planned). > > Best regards, > j. > > > >> >> If the sources of dependence are more pervasive, like if there are batch >> effects: >> >> http://www.nature.com/nrg/journal/v11/n10/full/nrg2825.html >> >> Then you can either use the batch correction methods in Limma if, say, you >> know the date the samples were processed. Or, if you don't know the sources >> of large scale dependence, you can use the sva package: >> >> http://www.bioconductor.org/packages/devel/bioc/html/sva.html >> >> which implements the methods described here: >> >> http://www.pnas.org/content/early/2008/11/24/0808709105.abstract >> >> >> Best, >> >> >> Jeff >> >> >> >> On Jul 9, 2012 7:08 AM, "January Weiner" >> <january.weiner at="" mpiib-berlin.mpg.de=""> wrote: >>> >>> Hello, >>> >>> statistical methods for assessing significance of differences in >>> expression assume, correct me if I'm wrong, independence of the tests. >>> Does anyone have at hand any papers on the performance -- in terms of >>> type I error -- of methods such as limma / eBayes? I'm sure this issue >>> has been investigated in depth. >>> >>> Kind regards, >>> >>> January >>> >>> -- >>> -------- Dr. January Weiner 3 -------------------------------------- >>> Max Planck Institute for Infection Biology >>> Charit?platz 1 >>> D-10117 Berlin, Germany >>> Web : www.mpiib-berlin.mpg.de >>> Tel : +49-30-28460514 >>> Fax : +49-30-28450505 >>> >>> _______________________________________________ >>> 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 > > > -- Robert Castelo, PhD Associate Professor Dept. of Experimental and Health Sciences Universitat Pompeu Fabra (UPF) Barcelona Biomedical Research Park (PRBB) Dr Aiguader 88 E-08003 Barcelona, Spain telf: +34.933.160.514 fax: +34.933.160.550
ADD REPLY
0
Entering edit mode
January, if you only require per-gene p-values and no multiple testing adjustment, then the dependency is never a problem. The validity of unadjusted per-gene p-values is unaffected by whether there is dependency between the genes. For multiple testing, if you do FWER by the Westfall-Young method, any dependence is also no problem. If you do FDR by the Benjamini-Hochberg method, problems can in principle occur if there is pervasive dependence. Often this is caused by technical artifacts, which would be addressed (and removed) by the methods mentioned by Jeff. If it is biological, then a serial univariate analysis (gene-by-gene testing) does not seem the cleverest choice of approach, and a truly multivariate approach seems more advisable. Best wishes Wolfgang Jeff Leek scripsit 07/09/2012 01:17 PM: > Hi January, > > If the tests are only dependent in small groups, say because genes are > grouped into small modules, then most FDR methods in the p.adjust() > function or the methods in the qvalue package will work. The Bonferroni > correction controls a more conservative error rate, but also holds under > dependence. > > If the sources of dependence are more pervasive, like if there are batch > effects: > > http://www.nature.com/nrg/journal/v11/n10/full/nrg2825.html > > Then you can either use the batch correction methods in Limma if, say, you > know the date the samples were processed. Or, if you don't know the sources > of large scale dependence, you can use the sva package: > > http://www.bioconductor.org/packages/devel/bioc/html/sva.html > > which implements the methods described here: > > http://www.pnas.org/content/early/2008/11/24/0808709105.abstract > > > Best, > > > Jeff > > > > On Jul 9, 2012 7:08 AM, "January Weiner" <january.weiner at="" mpiib-="" berlin.mpg.de=""> > wrote: > >> Hello, >> >> statistical methods for assessing significance of differences in >> expression assume, correct me if I'm wrong, independence of the tests. >> Does anyone have at hand any papers on the performance -- in terms of >> type I error -- of methods such as limma / eBayes? I'm sure this issue >> has been investigated in depth. >> >> Kind regards, >> >> January >> >> -- >> -------- Dr. January Weiner 3 -------------------------------------- >> Max Planck Institute for Infection Biology >> Charit?platz 1 >> D-10117 Berlin, Germany >> Web : www.mpiib-berlin.mpg.de >> Tel : +49-30-28460514 >> Fax : +49-30-28450505 >> >> _______________________________________________ >> 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 >> > > [[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 > -- Best wishes Wolfgang Wolfgang Huber EMBL http://www.embl.de/research/units/genome_biology/huber
ADD REPLY
0
Entering edit mode
Hi Wolfgang, It's not just technical artifacts. Everyone believes (probably correctly) that gene expression in biological samples is in fact correlated, a fact that is exploited all the time when people run algorithms to try to (re)construct networks or pathways based on coexpression. And while I agree that a truly multivariate approach would be more advisable, (a) there is no consensus on how best to do this and (b) it is not the current standard practice. There are already gazillions of papers (and more are being written and published as I write this email) that compute p-values from univariate gene-by-gene tests and follow with a method to estimate the FDR. The operative word here is "estimate", which should make you think that there might be some uncertainty in the estimates. We recently did some simulations to get an idea of how much the precision of the FDR estimates is affected by correlation. We also point out a couple of examples from real data that suggest that the effect of correlation could be large. The paper has been accepted at BMC Bioinformatics, so I can supply the advance URL for people who want more information: http://www.biomedcentral.com/1471-2105/13/S13/S1/abstract Best, Kevin On 7/11/2012 7:22 AM, Wolfgang Huber wrote: > January, > > if you only require per-gene p-values and no multiple testing > adjustment, then the dependency is never a problem. The validity of > unadjusted per-gene p-values is unaffected by whether there is > dependency between the genes. > > For multiple testing, if you do FWER by the Westfall-Young method, any > dependence is also no problem. If you do FDR by the Benjamini- Hochberg > method, problems can in principle occur if there is pervasive > dependence. Often this is caused by technical artifacts, which would > be addressed (and removed) by the methods mentioned by Jeff. If it is > biological, then a serial univariate analysis (gene-by-gene testing) > does not seem the cleverest choice of approach, and a truly > multivariate approach seems more advisable. > > Best wishes > Wolfgang > > > Jeff Leek scripsit 07/09/2012 01:17 PM: >> Hi January, >> >> If the tests are only dependent in small groups, say because genes are >> grouped into small modules, then most FDR methods in the p.adjust() >> function or the methods in the qvalue package will work. The Bonferroni >> correction controls a more conservative error rate, but also holds under >> dependence. >> >> If the sources of dependence are more pervasive, like if there are batch >> effects: >> >> http://www.nature.com/nrg/journal/v11/n10/full/nrg2825.html >> >> Then you can either use the batch correction methods in Limma if, >> say, you >> know the date the samples were processed. Or, if you don't know the >> sources >> of large scale dependence, you can use the sva package: >> >> http://www.bioconductor.org/packages/devel/bioc/html/sva.html >> >> which implements the methods described here: >> >> http://www.pnas.org/content/early/2008/11/24/0808709105.abstract >> >> >> Best, >> >> >> Jeff >> >> >> >> On Jul 9, 2012 7:08 AM, "January Weiner" >> <january.weiner at="" mpiib-berlin.mpg.de=""> >> wrote: >> >>> Hello, >>> >>> statistical methods for assessing significance of differences in >>> expression assume, correct me if I'm wrong, independence of the tests. >>> Does anyone have at hand any papers on the performance -- in terms of >>> type I error -- of methods such as limma / eBayes? I'm sure this issue >>> has been investigated in depth. >>> >>> Kind regards, >>> >>> January >>> >>> -- >>> -------- Dr. January Weiner 3 -------------------------------------- >>> Max Planck Institute for Infection Biology >>> Charit?platz 1 >>> D-10117 Berlin, Germany >>> Web : www.mpiib-berlin.mpg.de >>> Tel : +49-30-28460514 >>> Fax : +49-30-28450505 >>> >>> _______________________________________________ >>> 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 >>> >> >> [[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 REPLY
0
Entering edit mode
Dear Kevin I agree with your points. Two comments: - You very appropriately point out below and in your paper that point estimates of the FDR can be all over the place, and in particular so when there is a lot of correlation. I have often seen that, too, and it should probably be much more emphasized to statistics-naive users. But isn't avoiding that exactly the point of methods like that of Benjamini-Hochberg or Benjamini-Yekutieli that control FDR under various types of dependence [1]? - It is not quite as bad with multivariate methods, if you include (as I do) clustering, heatmaps and classification. In microarray analysis, these even predate the gene-by-gene testing. In particular heatmaps can be surprisingly useful for detecting the major correlation structures. Things get a bit more ambiguous and less automatable than with gene-by-gene testing, but I don't think that's a reason not to do it. Best wishes Wolfgang [1] The control of the false discovery rate in multiple testing under dependency, Yoav Benjamini and Daniel Yekutieli, Ann. Statist. Volume 29, Number 4 (2001), 1165-1188. Kevin R. Coombes scripsit 07/11/2012 06:22 PM: > Hi Wolfgang, > > It's not just technical artifacts. Everyone believes (probably > correctly) that gene expression in biological samples is in fact > correlated, a fact that is exploited all the time when people run > algorithms to try to (re)construct networks or pathways based on > coexpression. And while I agree that a truly multivariate approach > would be more advisable, (a) there is no consensus on how best to do > this and (b) it is not the current standard practice. There are already > gazillions of papers (and more are being written and published as I > write this email) that compute p-values from univariate gene-by- gene > tests and follow with a method to estimate the FDR. > > The operative word here is "estimate", which should make you think that > there might be some uncertainty in the estimates. We recently did some > simulations to get an idea of how much the precision of the FDR > estimates is affected by correlation. We also point out a couple of > examples from real data that suggest that the effect of correlation > could be large. The paper has been accepted at BMC Bioinformatics, so I > can supply the advance URL for people who want more information: > http://www.biomedcentral.com/1471-2105/13/S13/S1/abstract > > Best, > Kevin > > On 7/11/2012 7:22 AM, Wolfgang Huber wrote: >> January, >> >> if you only require per-gene p-values and no multiple testing >> adjustment, then the dependency is never a problem. The validity of >> unadjusted per-gene p-values is unaffected by whether there is >> dependency between the genes. >> >> For multiple testing, if you do FWER by the Westfall-Young method, any >> dependence is also no problem. If you do FDR by the Benjamini- Hochberg >> method, problems can in principle occur if there is pervasive >> dependence. Often this is caused by technical artifacts, which would >> be addressed (and removed) by the methods mentioned by Jeff. If it is >> biological, then a serial univariate analysis (gene-by-gene testing) >> does not seem the cleverest choice of approach, and a truly >> multivariate approach seems more advisable. >> >> Best wishes >> Wolfgang >> >> >> Jeff Leek scripsit 07/09/2012 01:17 PM: >>> Hi January, >>> >>> If the tests are only dependent in small groups, say because genes are >>> grouped into small modules, then most FDR methods in the p.adjust() >>> function or the methods in the qvalue package will work. The Bonferroni >>> correction controls a more conservative error rate, but also holds under >>> dependence. >>> >>> If the sources of dependence are more pervasive, like if there are batch >>> effects: >>> >>> http://www.nature.com/nrg/journal/v11/n10/full/nrg2825.html >>> >>> Then you can either use the batch correction methods in Limma if, >>> say, you >>> know the date the samples were processed. Or, if you don't know the >>> sources >>> of large scale dependence, you can use the sva package: >>> >>> http://www.bioconductor.org/packages/devel/bioc/html/sva.html >>> >>> which implements the methods described here: >>> >>> http://www.pnas.org/content/early/2008/11/24/0808709105.abstract >>> >>> >>> Best, >>> >>> >>> Jeff >>> >>> >>> >>> On Jul 9, 2012 7:08 AM, "January Weiner" >>> <january.weiner at="" mpiib-berlin.mpg.de=""> >>> wrote: >>> >>>> Hello, >>>> >>>> statistical methods for assessing significance of differences in >>>> expression assume, correct me if I'm wrong, independence of the tests. >>>> Does anyone have at hand any papers on the performance -- in terms of >>>> type I error -- of methods such as limma / eBayes? I'm sure this issue >>>> has been investigated in depth. >>>> >>>> Kind regards, >>>> >>>> January >>>> >>>> -- >>>> -------- Dr. January Weiner 3 -------------------------------------- >>>> Max Planck Institute for Infection Biology >>>> Charit?platz 1 >>>> D-10117 Berlin, Germany >>>> Web : www.mpiib-berlin.mpg.de >>>> Tel : +49-30-28460514 >>>> Fax : +49-30-28450505 >>>> >>>> _______________________________________________ >>>> 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 >>>> >>> >>> [[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 >>> >> >> -- Best wishes Wolfgang Wolfgang Huber EMBL http://www.embl.de/research/units/genome_biology/huber
ADD REPLY

Login before adding your answer.

Traffic: 445 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