LIMMA decideTests result zero from contrast matrix
1
0
Entering edit mode
Ekta Jain ▴ 370
@ekta-jain-5131
Last seen 10.2 years ago
Dear All, I am using the LIMMA package to create 2 contrasts for my data and then calculating the vennCounts of the decideTests from the contrast.fit to be able to create venn Diagrams. The code works fine but the summary(results) shows zeros for all i.e. no gene were up regulated or downregulated. This is not true for my data since toptable output shows Log fold change greater than > 2. I am certain it is a small glitch somewhere at my end that i get zero counts for my summary(decideTests). Please find below my code and i would really appreciate any help here at all. Thanks, Ekta ## R Script ### > numGenes <- nrow(eset) > library(limma) > samples <- c("Un","Un","DMSO10","DMSO10","DMSO5","DMSO5"); > fl <- as.factor(samples) > design <- model.matrix(~ 0+ fl) > colnames(design) <- levels(fl) > fit <- lmFit(eset, design) > cont.matrix <- makeContrasts(DMSO10-Un, DMSO5-Un, levels=design) > fit2 <- contrasts.fit(fit, cont.matrix) > fit2 <- eBayes(fit2) > tTUni<- topTable(fitUni, adjust="fdr", sort.by="B", number=numGenes) > results <- decideTests(fit2) > vennDiagram(results,include=c("up","down"),counts.col=c("red","green ")) ## ### I get zero genes for upregulation and downregulation ##### Senior Research Associate Bioinformatics Department Jubilant Biosys Pvt Ltd, #96, Industrial Suburb, 2nd Stage Yeshwantpur, Bangalore 560 022 Ph No : +91-80-66628346 The information contained in this electronic message and in any attachments to this message is confidential, legally privileged and intended only for use by the person or entity to which this electronic message is addressed. If you are not the intended recipient, and have received this message in error, please notify the sender and system manager by return email and delete the message and its attachments and also you are hereby notified that any distribution, copying, review, retransmission, dissemination or other use of this electronic transmission or the information contained in it is strictly prohibited. Please note that any views or opinions presented in this email are solely those of the author and may not represent those of the Company or bind the Company. Any commitments made over e-mail are not financially binding on the company unless accompanied or followed by a valid purchase order. This message has been scanned for viruses and dangerous content by Mail Scanner, and is believed to be clean. The Company accepts no liability for any damage caused by any virus transmitted by this email. www.jubl.com [[alternative HTML version deleted]]
limma limma • 3.0k views
ADD COMMENT •
0
Entering edit mode
Ekta Jain ▴ 370
@ekta-jain-5131
Last seen 10.2 years ago
An embedded and charset-unspecified text was scrubbed... Name: not available URL: <https: stat.ethz.ch="" pipermail="" bioconductor="" attachments="" 20120504="" a7471bc4="" attachment-0001.pl="">
ADD COMMENT •
0
Entering edit mode
Hi, First order of business: please don't cross post between mailing list -- it's generally considered bad etiquette. Second: On Fri, May 4, 2012 at 2:44 AM, Ekta Jain <ekta_jain at="" jubilantbiosys.com=""> wrote: > Dear All, > I am using the LIMMA package to create 2 contrasts for my data and then calculating the vennCounts of the decideTests from the contrast.fit to be able to create venn Diagrams. > > The code works fine but the summary(results) shows zeros for all i.e. no gene were up regulated or downregulated. This is not true for my data since toptable output shows Log fold change greater than > 2. What are the adjusted p-values for these genes? -steve -- Steve Lianoglou Graduate Student: Computational Systems Biology ?| Memorial Sloan-Kettering Cancer Center ?| Weill Medical College of Cornell University Contact Info: http://cbio.mskcc.org/~lianos/contact
ADD REPLY •
0
Entering edit mode
Hi Steve, What do you mean by "cross-post between mailing list" ? I usually mail to both Bioconductor and R to avoid the suggestion to email the 'correct' place. More so since i was not sure who would be better able to answer my question. Adjusted p-values are pretty bad - 0.999540865122392 -Ekta -----Original Message----- From: Steve Lianoglou [mailto:mailinglist.honeypot@gmail.com] Sent: 04 May 2012 19:18 To: Ekta Jain Cc: bioconductor at r-project.org Subject: Re: [BioC] LIMMA decideTests result zero from contrast matrix Hi, First order of business: please don't cross post between mailing list -- it's generally considered bad etiquette. Second: On Fri, May 4, 2012 at 2:44 AM, Ekta Jain <ekta_jain at="" jubilantbiosys.com=""> wrote: > Dear All, > I am using the LIMMA package to create 2 contrasts for my data and then calculating the vennCounts of the decideTests from the contrast.fit to be able to create venn Diagrams. > > The code works fine but the summary(results) shows zeros for all i.e. no gene were up regulated or downregulated. This is not true for my data since toptable output shows Log fold change greater than > 2. What are the adjusted p-values for these genes? -steve -- Steve Lianoglou Graduate Student: Computational Systems Biology ?| Memorial Sloan-Kettering Cancer Center ?| Weill Medical College of Cornell University Contact Info: http://cbio.mskcc.org/~lianos/contact The information contained in this electronic message and in any attachments to this message is confidential, legally privileged and intended only for use by the person or entity to which this electronic message is addressed. If you are not the intended recipient, and have received this message in error, please notify the sender and system manager by return email and delete the message and its attachments and also you are hereby notified that any distribution, copying, review, retransmission, dissemination or other use of this electronic transmission or the information contained in it is strictly prohibited. Please note that any views or opinions presented in this email are solely those of the author and may not represent those of the Company or bind the Company. Any commitments made over e-mail are not financially binding on the company unless accompanied or followed by a valid purchase order. This message has been scanned for viruses and dangerous content by Mail Scanner, and is believed to be clean. The Company accepts no liability for any damage caused by any virus transmitted by this email. www.jubl.com
ADD REPLY •
0
Entering edit mode
Hi Ekta, On Sun, May 6, 2012 at 11:38 PM, Ekta Jain <ekta_jain at="" jubilantbiosys.com=""> wrote: > Hi Steve, > What do you mean by "cross-post between mailing list" ? I usually mail to both Bioconductor and R to avoid the suggestion to email the 'correct' place. More so since i was not sure who would be better able to answer my question. I mean putting multiple mailing lists in the to and cc fields of the same email -- it's frowned upon. A quick googling of "cross posting mailing lists" will get you more info. Heck, there's even a wikipedia page :-) http://en.wikipedia.org/wiki/Crossposting So -- what you "usually" do: bad :-) If your question has to do w/ some bioinformatics topic (like this question), then post here. If it has nothing to do w/ bioinformatics, but more about some base R thing, eg: "how come I can't read this table" or whatever, then go to R-help. > Adjusted p-values are pretty bad - 0.999540865122392 As Jim also pointed out, I reckon this is why nothing's popping up in your venn diagram. -steve -- Steve Lianoglou Graduate Student: Computational Systems Biology ?| Memorial Sloan-Kettering Cancer Center ?| Weill Medical College of Cornell University Contact Info: http://cbio.mskcc.org/~lianos/contact
ADD REPLY •
0
Entering edit mode
Hi Steve, I have been using the lists for a long time now and was never told off before for emailing to both R (r-help at r-project.org) and Bioconductor(bioconductor at r-project.org) at the same time. I think, it saves time and is the quickest way to get your query across people since not everyone can be on both the lists. I can most certainly try and post to the relevant list but 95% of the time I am then told to post to the other list since the topic is not relevant here. I am afraid this process at times takes up an extra day before some good soul out there reads my post and offers help. I am sorry but I did not mean to stress anyone. I could have googled 'cross-posting' but I meant to ask in this context and not as a general rule. Ideally, I wouldn't rely on Wikipedia for all information :-). I've tried changing things and no results. I reckon, it's probably down to the data I have. It works fine on another dataset with decent p-values. I am clueless why it wouldn't work. You asked me for adjusted p-values, why? Thanks a lot ! -Ekta -----Original Message----- From: Steve Lianoglou [mailto:mailinglist.honeypot@gmail.com] Sent: 07 May 2012 19:00 To: Ekta Jain Cc: bioconductor at r-project.org Subject: Re: [BioC] LIMMA decideTests result zero from contrast matrix Hi Ekta, On Sun, May 6, 2012 at 11:38 PM, Ekta Jain <ekta_jain at="" jubilantbiosys.com=""> wrote: > Hi Steve, > What do you mean by "cross-post between mailing list" ? I usually mail to both Bioconductor and R to avoid the suggestion to email the 'correct' place. More so since i was not sure who would be better able to answer my question. I mean putting multiple mailing lists in the to and cc fields of the same email -- it's frowned upon. A quick googling of "cross posting mailing lists" will get you more info. Heck, there's even a wikipedia page :-) http://en.wikipedia.org/wiki/Crossposting So -- what you "usually" do: bad :-) If your question has to do w/ some bioinformatics topic (like this question), then post here. If it has nothing to do w/ bioinformatics, but more about some base R thing, eg: "how come I can't read this table" or whatever, then go to R-help. > Adjusted p-values are pretty bad - 0.999540865122392 As Jim also pointed out, I reckon this is why nothing's popping up in your venn diagram. -steve -- Steve Lianoglou Graduate Student: Computational Systems Biology ?| Memorial Sloan-Kettering Cancer Center ?| Weill Medical College of Cornell University Contact Info: http://cbio.mskcc.org/~lianos/contact The information contained in this electronic message and in any attachments to this message is confidential, legally privileged and intended only for use by the person or entity to which this electronic message is addressed. If you are not the intended recipient, and have received this message in error, please notify the sender and system manager by return email and delete the message and its attachments and also you are hereby notified that any distribution, copying, review, retransmission, dissemination or other use of this electronic transmission or the information contained in it is strictly prohibited. Please note that any views or opinions presented in this email are solely those of the author and may not represent those of the Company or bind the Company. Any commitments made over e-mail are not financially binding on the company unless accompanied or followed by a valid purchase order. This message has been scanned for viruses and dangerous content by Mail Scanner, and is believed to be clean. The Company accepts no liability for any damage caused by any virus transmitted by this email. www.jubl.com
ADD REPLY •
0
Entering edit mode
Hi, On Tue, May 8, 2012 at 12:42 AM, Ekta Jain <ekta_jain at="" jubilantbiosys.com=""> wrote: > Hi Steve, > I have been using the lists for a long time now and was never told off before for emailing to both R (r-help at r-project.org) and Bioconductor(bioconductor at r-project.org) at the same time. I think, it saves time and is the quickest way to get your query across people since not everyone can be on both the lists. It's also a quick way to annoy people ... I'm not trying to reprimand you or anything, I'm just pointing out what "the rules of the road" are on most mailing lists so that you can learn and stick to them. > I can most certainly try and post to the relevant list but 95% of the time I am then told to post to the other list since the topic is not relevant here. It seems as if you simply have to flip the sign of your rhelp vs. bioc classifier and all will be well ;-) > I am afraid this process at times takes up an extra day before some good soul out there reads my post and offers help. I am sorry but I did not mean to stress anyone. > > I could have googled 'cross-posting' but I meant to ask in this context and not as a general rule. Ideally, I wouldn't rely on Wikipedia for all information :-). I feel like wikipedia gets a worse wrap than it deserves -- but all the same, if you search the bioconductor or r-help mailing list for the relevant keywords, you'll get many results, eg: http://search.gmane.org/?query=please+don%27t+cross+post&author=&group =gmane.science.biology.informatics.conductor&sort=relevance&DEFAULTOP= and&xP=cross%09post&xFILTERS=Gscience.biology.informatics.conductor--- A > I've tried changing things and no results. I reckon, it's probably down to the data I have. It works fine on another dataset with decent p-values. I am clueless why it wouldn't work. You're getting warmer, here: > You asked me for adjusted p-values, why? Because it seems as if the default way you are calling vennDiagram results in reporting counts for genes that have *adjusted* pvalues below a certain threshold. You've said that your analysis results in 0 such genes, so this would explain why your vennDiagram is turning up empty. If you read the help in ?vennDiagram, the fact that "x is .. Usually created by `decideTests`". Now, looking back at your code, the `results` you are passing into your vennDiagram call are coming from a call to decideTests which will, by default (which you're doing), only pick the genes that pass a certain threshold for their *adjusted* pvalues. To convince yourself of this fact, it wouldn't be a bad idea to minimally read the help pages of the functions you're calling. I'd also recommend downloading the source code for limma and poke around these functions to see what's going on under the hood ... you can, of course, accomplish this w/o d/ling the code, but I find that browsing through the source code of packages is a good way to learn, as well. -steve -- Steve Lianoglou Graduate Student: Computational Systems Biology ?| Memorial Sloan-Kettering Cancer Center ?| Weill Medical College of Cornell University Contact Info: http://cbio.mskcc.org/~lianos/contact
ADD REPLY •
0
Entering edit mode
Hi Steve, Definitely not annoy people here on. I am being given help, which is nice. Apologies for the same. I wouldn't completely rely on Wikipedia, ever :) I am a warm person. Yes, (x,...) is generated from decideTests(fit). The summary(decideTests, lfc=0, method="separate",adjust.method="BH",p.value=0.05) gives 0 genes for upregulation and downregulation. I can only set the threshold for lfc, p-value and not for *adjusted p-values* - http://127.0.0.1:21780/library/limma/html/decideTests.html Working on the LIMMA source code is next on the list. Thank you much for your help and time. Best, Ekta -----Original Message----- From: Steve Lianoglou [mailto:mailinglist.honeypot@gmail.com] Sent: 08 May 2012 10:44 To: Ekta Jain Cc: bioconductor at r-project.org Subject: Re: [BioC] LIMMA decideTests result zero from contrast matrix Hi, On Tue, May 8, 2012 at 12:42 AM, Ekta Jain <ekta_jain at="" jubilantbiosys.com=""> wrote: > Hi Steve, > I have been using the lists for a long time now and was never told off before for emailing to both R (r-help at r-project.org) and Bioconductor(bioconductor at r-project.org) at the same time. I think, it saves time and is the quickest way to get your query across people since not everyone can be on both the lists. It's also a quick way to annoy people ... I'm not trying to reprimand you or anything, I'm just pointing out what "the rules of the road" are on most mailing lists so that you can learn and stick to them. > I can most certainly try and post to the relevant list but 95% of the time I am then told to post to the other list since the topic is not relevant here. It seems as if you simply have to flip the sign of your rhelp vs. bioc classifier and all will be well ;-) > I am afraid this process at times takes up an extra day before some good soul out there reads my post and offers help. I am sorry but I did not mean to stress anyone. > > I could have googled 'cross-posting' but I meant to ask in this context and not as a general rule. Ideally, I wouldn't rely on Wikipedia for all information :-). I feel like wikipedia gets a worse wrap than it deserves -- but all the same, if you search the bioconductor or r-help mailing list for the relevant keywords, you'll get many results, eg: http://search.gmane.org/?query=please+don%27t+cross+post&author=&group =gmane.science.biology.informatics.conductor&sort=relevance&DEFAULTOP= and&xP=cross%09post&xFILTERS=Gscience.biology.informatics.conductor--- A > I've tried changing things and no results. I reckon, it's probably down to the data I have. It works fine on another dataset with decent p-values. I am clueless why it wouldn't work. You're getting warmer, here: - > You asked me for adjusted p-values, why? Because it seems as if the default way you are calling vennDiagram results in reporting counts for genes that have *adjusted* pvalues below a certain threshold. You've said that your analysis results in 0 such genes, so this would explain why your vennDiagram is turning up empty. - If you read the help in ?vennDiagram, the fact that "x is .. Usually created by `decideTests`". Now, looking back at your code, the `results` you are passing into your vennDiagram call are coming from a call to decideTests which will, by default (which you're doing), only pick the genes that pass a certain threshold for their *adjusted* pvalues. To convince yourself of this fact, it wouldn't be a bad idea to minimally read the help pages of the functions you're calling. I'd also recommend downloading the source code for limma and poke around these functions to see what's going on under the hood ... you can, of course, accomplish this w/o d/ling the code, but I find that browsing through the source code of packages is a good way to learn, as well. -steve -- Steve Lianoglou Graduate Student: Computational Systems Biology ?| Memorial Sloan-Kettering Cancer Center ?| Weill Medical College of Cornell University Contact Info: http://cbio.mskcc.org/~lianos/contact The information contained in this electronic message and in any attachments to this message is confidential, legally privileged and intended only for use by the person or entity to which this electronic message is addressed. If you are not the intended recipient, and have received this message in error, please notify the sender and system manager by return email and delete the message and its attachments and also you are hereby notified that any distribution, copying, review, retransmission, dissemination or other use of this electronic transmission or the information contained in it is strictly prohibited. Please note that any views or opinions presented in this email are solely those of the author and may not represent those of the Company or bind the Company. Any commitments made over e-mail are not financially binding on the company unless accompanied or followed by a valid purchase order. This message has been scanned for viruses and dangerous content by Mail Scanner, and is believed to be clean. The Company accepts no liability for any damage caused by any virus transmitted by this email. www.jubl.com
ADD REPLY •
0
Entering edit mode
Hi Ekta, On 5/8/2012 1:49 AM, Ekta Jain wrote: > Hi Steve, > Definitely not annoy people here on. I am being given help, which is nice. Apologies for the same. > > I wouldn't completely rely on Wikipedia, ever :) > > I am a warm person. > > Yes, (x,...) is generated from decideTests(fit). The summary(decideTests, lfc=0, method="separate",adjust.method="BH",p.value=0.05) gives 0 genes for upregulation and downregulation. I can only set the threshold for lfc, p-value and not for *adjusted p-values* - http://127.0.0.1:21780/library/limma/html/decideTests.html It's ironic that you post a (busted) link to a help page that you have evidently not read yourself. From that man page: Usage: decideTests(object,method="separate",adjust.method="BH",p.value=0.05,l fc=0) and adjust.method: character string specifying p-value adjustment method. Possible values are "none" , "BH" , "fdr" (equivalent to "BH" ), "BY" and "holm" . See p.adjust for details. So by _default_ decideTests() uses a BH adjusted p-value to filter genes. Both Steve and I have pointed out numerous times that you are doing two different things and expecting the same result. Yet you persist in ignoring what we have been telling you. You may have got the same results before with this code, but that was a coincidence. The reason you are "clueless why it wouldn't work" is because you seem bent on ignoring what people tell you. So let me try once more. Please read the following very closely... When you are using topTable you are not filtering on p-values. In fact you aren't filtering at all and stating that there are lots of genes with a |fold| > 2. You then use decideTests(), which filters on adjusted p-values. Since your smallest adjusted p-value is something like 0.999, everything gets filtered out. So you don't filter using topTable(), but you do filter using decideTests(). That is why you get different results! Best, Jim > > Working on the LIMMA source code is next on the list. > > Thank you much for your help and time. > > Best, > Ekta > > > > > -----Original Message----- > From: Steve Lianoglou [mailto:mailinglist.honeypot at gmail.com] > Sent: 08 May 2012 10:44 > To: Ekta Jain > Cc: bioconductor at r-project.org > Subject: Re: [BioC] LIMMA decideTests result zero from contrast matrix > > Hi, > > On Tue, May 8, 2012 at 12:42 AM, Ekta Jain<ekta_jain at="" jubilantbiosys.com=""> wrote: >> Hi Steve, >> I have been using the lists for a long time now and was never told off before for emailing to both R (r-help at r-project.org) and Bioconductor(bioconductor at r-project.org) at the same time. I think, it saves time and is the quickest way to get your query across people since not everyone can be on both the lists. > It's also a quick way to annoy people ... I'm not trying to reprimand > you or anything, I'm just pointing out what "the rules of the road" > are on most mailing lists so that you can learn and stick to them. > >> I can most certainly try and post to the relevant list but 95% of the time I am then told to post to the other list since the topic is not relevant here. > It seems as if you simply have to flip the sign of your rhelp vs. bioc > classifier and all will be well ;-) > >> I am afraid this process at times takes up an extra day before some good soul out there reads my post and offers help. I am sorry but I did not mean to stress anyone. >> >> I could have googled 'cross-posting' but I meant to ask in this context and not as a general rule. Ideally, I wouldn't rely on Wikipedia for all information :-). > I feel like wikipedia gets a worse wrap than it deserves -- but all > the same, if you search the bioconductor or r-help mailing list for > the relevant keywords, you'll get many results, eg: > > http://search.gmane.org/?query=please+don%27t+cross+post&author=&gro up=gmane.science.biology.informatics.conductor&sort=relevance&DEFAULTO P=and&xP=cross%09post&xFILTERS=Gscience.biology.informatics.conductor ---A > >> I've tried changing things and no results. I reckon, it's probably down to the data I have. It works fine on another dataset with decent p-values. I am clueless why it wouldn't work. > You're getting warmer, here: - > >> You asked me for adjusted p-values, why? > Because it seems as if the default way you are calling vennDiagram > results in reporting counts for genes that have *adjusted* pvalues > below a certain threshold. You've said that your analysis results in 0 > such genes, so this would explain why your vennDiagram is turning up > empty. - > > If you read the help in ?vennDiagram, the fact that "x is .. Usually > created by `decideTests`". Now, looking back at your code, the > `results` you are passing into your vennDiagram call are coming from a > call to decideTests which will, by default (which you're doing), only > pick the genes that pass a certain threshold for their *adjusted* > pvalues. > > To convince yourself of this fact, it wouldn't be a bad idea to > minimally read the help pages of the functions you're calling. I'd > also recommend downloading the source code for limma and poke around > these functions to see what's going on under the hood ... you can, of > course, accomplish this w/o d/ling the code, but I find that browsing > through the source code of packages is a good way to learn, as well. > > -steve > -- James W. MacDonald, M.S. Biostatistician University of Washington Environmental and Occupational Health Sciences 4225 Roosevelt Way NE, # 100 Seattle WA 98105-6099
ADD REPLY •
0
Entering edit mode
Dear Jim, I did change things around when you pointed out for the first time. All I have been doing is > numGenes <- rownames(eset) > topTable(fit2, coef=1, adjust="BH", sort.by="B", number=numGenes) And > results<-decideTests(fit2, method ="global", lfc =0) As you mention in your email that "by default decideTests() uses a BH adjusted p-value to filter genes" so am i not applying the same adjustment for both the toptable() and decideTests() here? This is what I am clueless about since I still get zero genes. I cannot seem to figure out how to not let decideTests use a BH adjust for p-value. For the sake of detail, the code worked fine for all my cell lines treated with CPI since the p-values were not as bad as the ones for treatment with DMSO. Thank you, Ekta ---Original Message----- From: James W. MacDonald [mailto:jmacdon@uw.edu] Sent: 08 May 2012 18:53 To: Ekta Jain Cc: Steve Lianoglou; bioconductor at r-project.org Subject: Re: [BioC] LIMMA decideTests result zero from contrast matrix Hi Ekta, On 5/8/2012 1:49 AM, Ekta Jain wrote: > Hi Steve, > Definitely not annoy people here on. I am being given help, which is nice. Apologies for the same. > > I wouldn't completely rely on Wikipedia, ever :) > > I am a warm person. > > Yes, (x,...) is generated from decideTests(fit). The summary(decideTests, lfc=0, method="separate",adjust.method="BH",p.value=0.05) gives 0 genes for upregulation and downregulation. I can only set the threshold for lfc, p-value and not for *adjusted p-values* - http://127.0.0.1:21780/library/limma/html/decideTests.html It's ironic that you post a (busted) link to a help page that you have evidently not read yourself. From that man page: Usage: decideTests(object,method="separate",adjust.method="BH",p.value=0.05,l fc=0) and adjust.method: character string specifying p-value adjustment method. Possible values are "none" , "BH" , "fdr" (equivalent to "BH" ), "BY" and "holm" . See p.adjust for details. So by _default_ decideTests() uses a BH adjusted p-value to filter genes. Both Steve and I have pointed out numerous times that you are doing two different things and expecting the same result. Yet you persist in ignoring what we have been telling you. You may have got the same results before with this code, but that was a coincidence. The reason you are "clueless why it wouldn't work" is because you seem bent on ignoring what people tell you. So let me try once more. Please read the following very closely... When you are using topTable you are not filtering on p-values. In fact you aren't filtering at all and stating that there are lots of genes with a |fold| > 2. You then use decideTests(), which filters on adjusted p-values. Since your smallest adjusted p-value is something like 0.999, everything gets filtered out. So you don't filter using topTable(), but you do filter using decideTests(). That is why you get different results! Best, Jim > > Working on the LIMMA source code is next on the list. > > Thank you much for your help and time. > > Best, > Ekta > > > > > -----Original Message----- > From: Steve Lianoglou [mailto:mailinglist.honeypot at gmail.com] > Sent: 08 May 2012 10:44 > To: Ekta Jain > Cc: bioconductor at r-project.org > Subject: Re: [BioC] LIMMA decideTests result zero from contrast matrix > > Hi, > > On Tue, May 8, 2012 at 12:42 AM, Ekta Jain<ekta_jain at="" jubilantbiosys.com=""> wrote: >> Hi Steve, >> I have been using the lists for a long time now and was never told off before for emailing to both R (r-help at r-project.org) and Bioconductor(bioconductor at r-project.org) at the same time. I think, it saves time and is the quickest way to get your query across people since not everyone can be on both the lists. > It's also a quick way to annoy people ... I'm not trying to reprimand > you or anything, I'm just pointing out what "the rules of the road" > are on most mailing lists so that you can learn and stick to them. > >> I can most certainly try and post to the relevant list but 95% of the time I am then told to post to the other list since the topic is not relevant here. > It seems as if you simply have to flip the sign of your rhelp vs. bioc > classifier and all will be well ;-) > >> I am afraid this process at times takes up an extra day before some good soul out there reads my post and offers help. I am sorry but I did not mean to stress anyone. >> >> I could have googled 'cross-posting' but I meant to ask in this context and not as a general rule. Ideally, I wouldn't rely on Wikipedia for all information :-). > I feel like wikipedia gets a worse wrap than it deserves -- but all > the same, if you search the bioconductor or r-help mailing list for > the relevant keywords, you'll get many results, eg: > > http://search.gmane.org/?query=please+don%27t+cross+post&author=&gro up=gmane.science.biology.informatics.conductor&sort=relevance&DEFAULTO P=and&xP=cross%09post&xFILTERS=Gscience.biology.informatics.conductor ---A > >> I've tried changing things and no results. I reckon, it's probably down to the data I have. It works fine on another dataset with decent p-values. I am clueless why it wouldn't work. > You're getting warmer, here: - > >> You asked me for adjusted p-values, why? > Because it seems as if the default way you are calling vennDiagram > results in reporting counts for genes that have *adjusted* pvalues > below a certain threshold. You've said that your analysis results in 0 > such genes, so this would explain why your vennDiagram is turning up > empty. - > > If you read the help in ?vennDiagram, the fact that "x is .. Usually > created by `decideTests`". Now, looking back at your code, the > `results` you are passing into your vennDiagram call are coming from a > call to decideTests which will, by default (which you're doing), only > pick the genes that pass a certain threshold for their *adjusted* > pvalues. > > To convince yourself of this fact, it wouldn't be a bad idea to > minimally read the help pages of the functions you're calling. I'd > also recommend downloading the source code for limma and poke around > these functions to see what's going on under the hood ... you can, of > course, accomplish this w/o d/ling the code, but I find that browsing > through the source code of packages is a good way to learn, as well. > > -steve > -- James W. MacDonald, M.S. Biostatistician University of Washington Environmental and Occupational Health Sciences 4225 Roosevelt Way NE, # 100 Seattle WA 98105-6099 The information contained in this electronic message and in any attachments to this message is confidential, legally privileged and intended only for use by the person or entity to which this electronic message is addressed. If you are not the intended recipient, and have received this message in error, please notify the sender and system manager by return email and delete the message and its attachments and also you are hereby notified that any distribution, copying, review, retransmission, dissemination or other use of this electronic transmission or the information contained in it is strictly prohibited. Please note that any views or opinions presented in this email are solely those of the author and may not represent those of the Company or bind the Company. Any commitments made over e-mail are not financially binding on the company unless accompanied or followed by a valid purchase order. This message has been scanned for viruses and dangerous content by Mail Scanner, and is believed to be clean. The Company accepts no liability for any damage caused by any virus transmitted by this email. www.jubl.com
ADD REPLY •
0
Entering edit mode
On 09.05.2012 04:21, Ekta Jain wrote: > Dear Jim, > I did change things around when you pointed out for the first time. > All I have been doing is >> numGenes <- rownames(eset) >> topTable(fit2, coef=1, adjust="BH", sort.by="B", number=numGenes) > And >> results<-decideTests(fit2, method ="global", lfc =0) > As you mention in your email that "by default decideTests() uses a BH > adjusted p-value to filter genes" so am i not applying the same > adjustment for both the toptable() and decideTests() here? > > This is what I am clueless about since I still get zero genes. I > cannot seem to figure out how to not let decideTests use a BH adjust > for p-value. For the sake of detail, the code worked fine for all my > cell lines treated with CPI since the p-values were not as bad as the > ones for treatment with DMSO. > > Thank you, > Ekta I think it is the difference in the default P value cutoffs for topTable and decideTests that is confusing the issue. From the decideTests docs: Usage: decideTests(object,method="separate",adjust.method="BH",p.value=0.05,l fc=0) Look at the default p.value cutoff - 0.05. Earlier in the thread you said your adjusted P values were ~0.9, hence *nothing* will come through the filter. The default for topTable is 1: Usage: topTable(fit, coef=NULL, number=10, genelist=fit$genes, adjust.method="BH", sort.by="B", resort.by=NULL, p.value=1, lfc=0, confint=FALSE) If you want to apply the filter to unadjusted P values the docs say this: adjust.method: character string specifying p-value adjustment method. Possible values are ?"none"?, ?"BH"?, ?"fdr"? (equivalent to ?"BH"?), ?"BY"? and ?"holm"?. See ?p.adjust? for details. Though I'm not sure why you would want to do this. -- Alex Gutteridge
ADD REPLY •
0
Entering edit mode
Dear Alex, Thank you very much. It appears much clear now. I agree, I wouldn't want to apply the filter to unadjusted P values since it just shows that the results obtained are very insignificant. I did not understand in detail how the decideTests() works and was looking for some info regarding the same in case someone else had a similar issue. The venn Diagrams are not important if the data is bad, it is not going to lead anywhere. It was very intriguing when summary(decideTests) gave zeros for all contrasts and now I know exactly why. Thank you kindly, Ekta -----Original Message----- From: bioconductor-bounces@r-project.org [mailto:bioconductor- bounces@r-project.org] On Behalf Of Alex Gutteridge Sent: 09 May 2012 13:50 To: bioconductor at r-project.org Subject: Re: [BioC] LIMMA decideTests result zero from contrast matrix On 09.05.2012 04:21, Ekta Jain wrote: > Dear Jim, > I did change things around when you pointed out for the first time. > All I have been doing is >> numGenes <- rownames(eset) >> topTable(fit2, coef=1, adjust="BH", sort.by="B", number=numGenes) > And >> results<-decideTests(fit2, method ="global", lfc =0) > As you mention in your email that "by default decideTests() uses a BH > adjusted p-value to filter genes" so am i not applying the same > adjustment for both the toptable() and decideTests() here? > > This is what I am clueless about since I still get zero genes. I > cannot seem to figure out how to not let decideTests use a BH adjust > for p-value. For the sake of detail, the code worked fine for all my > cell lines treated with CPI since the p-values were not as bad as the > ones for treatment with DMSO. > > Thank you, > Ekta I think it is the difference in the default P value cutoffs for topTable and decideTests that is confusing the issue. From the decideTests docs: Usage: decideTests(object,method="separate",adjust.method="BH",p.value=0.05,l fc=0) Look at the default p.value cutoff - 0.05. Earlier in the thread you said your adjusted P values were ~0.9, hence *nothing* will come through the filter. The default for topTable is 1: Usage: topTable(fit, coef=NULL, number=10, genelist=fit$genes, adjust.method="BH", sort.by="B", resort.by=NULL, p.value=1, lfc=0, confint=FALSE) If you want to apply the filter to unadjusted P values the docs say this: adjust.method: character string specifying p-value adjustment method. Possible values are ?"none"?, ?"BH"?, ?"fdr"? (equivalent to ?"BH"?), ?"BY"? and ?"holm"?. See ?p.adjust? for details. Though I'm not sure why you would want to do this. -- Alex Gutteridge _______________________________________________ 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 The information contained in this electronic message and in any attachments to this message is confidential, legally privileged and intended only for use by the person or entity to which this electronic message is addressed. If you are not the intended recipient, and have received this message in error, please notify the sender and system manager by return email and delete the message and its attachments and also you are hereby notified that any distribution, copying, review, retransmission, dissemination or other use of this electronic transmission or the information contained in it is strictly prohibited. Please note that any views or opinions presented in this email are solely those of the author and may not represent those of the Company or bind the Company. Any commitments made over e-mail are not financially binding on the company unless accompanied or followed by a valid purchase order. This message has been scanned for viruses and dangerous content by Mail Scanner, and is believed to be clean. The Company accepts no liability for any damage caused by any virus transmitted by this email. www.jubl.com
ADD REPLY •
0
Entering edit mode
Hi Ekta, On 5/4/2012 2:44 AM, Ekta Jain wrote: > Dear All, > I am using the LIMMA package to create 2 contrasts for my data and then calculating the vennCounts of the decideTests from the contrast.fit to be able to create venn Diagrams. > > The code works fine but the summary(results) shows zeros for all i.e. no gene were up regulated or downregulated. This is not true for my data since toptable output shows Log fold change greater than> 2. > > I am certain it is a small glitch somewhere at my end that i get zero counts for my summary(decideTests). Please find below my code and i would really appreciate any help here at all. > > Thanks, > Ekta > ## R Script ### >> numGenes<- nrow(eset) >> library(limma) >> samples<- c("Un","Un","DMSO10","DMSO10","DMSO5","DMSO5"); >> fl<- as.factor(samples) >> design<- model.matrix(~ 0+ fl) >> colnames(design)<- levels(fl) >> fit<- lmFit(eset, design) >> cont.matrix<- makeContrasts(DMSO10-Un, DMSO5-Un, levels=design) >> fit2<- contrasts.fit(fit, cont.matrix) >> fit2<- eBayes(fit2) >> tTUni<- topTable(fitUni, adjust="fdr", sort.by="B", number=numGenes) Here you are outputting all genes, regardless of adjusted p-value. And you claim some have |log fold| > 2. >> results<- decideTests(fit2) >> vennDiagram(results,include=c("up","down"),counts.col=c("red","gree n")) Here you are requiring that the genes have an FDR adjusted p-value < 0.05, and not putting any constraints on fold change. Since these are different criteria, is it surprising you get different results? Best, Jim > ## ### I get zero genes for upregulation and downregulation ##### > > > Senior Research Associate > Bioinformatics Department > Jubilant Biosys Pvt Ltd, > #96, Industrial Suburb, 2nd Stage > Yeshwantpur, Bangalore 560 022 > Ph No : +91-80-66628346 > > The information contained in this electronic message and in any attachments to this message is confidential, legally privileged and intended only for use by the person or entity to which this electronic message is addressed. If you are not the intended recipient, and have received this message in error, please notify the sender and system manager by return email and delete the message and its attachments and also you are hereby notified that any distribution, copying, review, retransmission, dissemination or other use of this electronic transmission or the information contained in it is strictly prohibited. Please note that any views or opinions presented in this email are solely those of the author and may not represent those of the Company or bind the Company. Any commitments made over e-mail are not financially binding on the company unless accompanied or followed by a valid purchase order. This message has been scanned for viruses and dangerous content by Mail Scanner, a! > nd is believed to be clean. The Company accepts no liability for any damage caused by any virus transmitted by this email. > www.jubl.com > > [[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 -- James W. MacDonald, M.S. Biostatistician University of Washington Environmental and Occupational Health Sciences 4225 Roosevelt Way NE, # 100 Seattle WA 98105-6099
ADD REPLY •
0
Entering edit mode
Dear Jim, I have tried the same code on a different dataset and works alright. Doesn't the decideTests(fit2) take lfc as zero by default? Thank you, Ekta -----Original Message----- From: James W. MacDonald [mailto:jmacdon@uw.edu] Sent: 04 May 2012 19:36 To: Ekta Jain Cc: bioconductor at r-project.org Subject: Re: [BioC] LIMMA decideTests result zero from contrast matrix Hi Ekta, On 5/4/2012 2:44 AM, Ekta Jain wrote: > Dear All, > I am using the LIMMA package to create 2 contrasts for my data and then calculating the vennCounts of the decideTests from the contrast.fit to be able to create venn Diagrams. > > The code works fine but the summary(results) shows zeros for all i.e. no gene were up regulated or downregulated. This is not true for my data since toptable output shows Log fold change greater than> 2. > > I am certain it is a small glitch somewhere at my end that i get zero counts for my summary(decideTests). Please find below my code and i would really appreciate any help here at all. > > Thanks, > Ekta > ## R Script ### >> numGenes<- nrow(eset) >> library(limma) >> samples<- c("Un","Un","DMSO10","DMSO10","DMSO5","DMSO5"); >> fl<- as.factor(samples) >> design<- model.matrix(~ 0+ fl) >> colnames(design)<- levels(fl) >> fit<- lmFit(eset, design) >> cont.matrix<- makeContrasts(DMSO10-Un, DMSO5-Un, levels=design) >> fit2<- contrasts.fit(fit, cont.matrix) >> fit2<- eBayes(fit2) >> tTUni<- topTable(fitUni, adjust="fdr", sort.by="B", number=numGenes) Here you are outputting all genes, regardless of adjusted p-value. And you claim some have |log fold| > 2. >> results<- decideTests(fit2) >> vennDiagram(results,include=c("up","down"),counts.col=c("red","gree n")) Here you are requiring that the genes have an FDR adjusted p-value < 0.05, and not putting any constraints on fold change. Since these are different criteria, is it surprising you get different results? Best, Jim > ## ### I get zero genes for upregulation and downregulation ##### > > > Senior Research Associate > Bioinformatics Department > Jubilant Biosys Pvt Ltd, > #96, Industrial Suburb, 2nd Stage > Yeshwantpur, Bangalore 560 022 > Ph No : +91-80-66628346 > > The information contained in this electronic message and in any attachments to this message is confidential, legally privileged and intended only for use by the person or entity to which this electronic message is addressed. If you are not the intended recipient, and have received this message in error, please notify the sender and system manager by return email and delete the message and its attachments and also you are hereby notified that any distribution, copying, review, retransmission, dissemination or other use of this electronic transmission or the information contained in it is strictly prohibited. Please note that any views or opinions presented in this email are solely those of the author and may not represent those of the Company or bind the Company. Any commitments made over e-mail are not financially binding on the company unless accompanied or followed by a valid purchase order. This message has been scanned for viruses and dangerous content by Mail Scanner, a! > nd is believed to be clean. The Company accepts no liability for any damage caused by any virus transmitted by this email. > www.jubl.com > > [[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 -- James W. MacDonald, M.S. Biostatistician University of Washington Environmental and Occupational Health Sciences 4225 Roosevelt Way NE, # 100 Seattle WA 98105-6099 The information contained in this electronic message and in any attachments to this message is confidential, legally privileged and intended only for use by the person or entity to which this electronic message is addressed. If you are not the intended recipient, and have received this message in error, please notify the sender and system manager by return email and delete the message and its attachments and also you are hereby notified that any distribution, copying, review, retransmission, dissemination or other use of this electronic transmission or the information contained in it is strictly prohibited. Please note that any views or opinions presented in this email are solely those of the author and may not represent those of the Company or bind the Company. Any commitments made over e-mail are not financially binding on the company unless accompanied or followed by a valid purchase order. This message has been scanned for viruses and dangerous content by Mail Scanner, and is believed to be clean. The Company accepts no liability for any damage caused by any virus transmitted by this email. www.jubl.com
ADD REPLY •

Login before adding your answer.

Similar Posts
Loading Similar Posts
Traffic: 961 users visited in the last hour
Content Search
Users
Tags
Badges
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