Using limma for quantitative proteomics data
3
0
Entering edit mode
Yong Li ▴ 80
@yong-li-5277
Last seen 7.4 years ago
Hello, limma has been so valuable in microarray data analysis, but has anyone used limma for finding differentially expressed proteins from quantitative proteomics data? The data I got has tumor/normal ratios of thousands proteins, and both tumor and normal have a number of replicates. Could such data be analyzed with limma? If limma can not be used here, what statistics method is suitable so that we can get statistically significant proteins with p-values? Any suggestion is appreciated. Kind regards, Yong
Microarray Proteomics limma Microarray Proteomics limma • 8.7k views
ADD COMMENT
0
Entering edit mode
Aaron Mackey ▴ 170
@aaron-mackey-4358
Last seen 7.4 years ago
yes, it should be possible with a voom()-based analysis to get the variances "right". -Aaron On Tue, Jun 19, 2012 at 12:47 PM, Yong Li <mail.yong.li@googlemail.com>wrote: > Hello, > > limma has been so valuable in microarray data analysis, but has anyone > used limma for finding differentially expressed proteins from > quantitative proteomics data? The data I got has tumor/normal ratios > of thousands proteins, and both tumor and normal have a number of > replicates. Could such data be analyzed with limma? > > If limma can not be used here, what statistics method is suitable so > that we can get statistically significant proteins with p-values? Any > suggestion is appreciated. > > Kind regards, > Yong > > _______________________________________________ > 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
Dear Aaron, thank you for your quick answer! I have checked the help page of voom() but it seems to be used for count data. My data are just tumor/normal ratios. I am wondering if you could provide more details? Best regards, Yong On Tue, Jun 19, 2012 at 8:18 PM, Aaron Mackey <amackey at="" virginia.edu=""> wrote: > yes, it should be possible with a voom()-based analysis to get the variances > "right". > > -Aaron > > On Tue, Jun 19, 2012 at 12:47 PM, Yong Li <mail.yong.li at="" googlemail.com=""> > wrote: >> >> Hello, >> >> limma has been so valuable in microarray data analysis, but has anyone >> used limma for finding differentially expressed proteins from >> quantitative proteomics data? The data I got has tumor/normal ratios >> of thousands proteins, and both tumor and normal have a number of >> replicates. Could such data be analyzed with limma? >> >> If limma can not be used here, what statistics method is suitable so >> that we can get statistically significant proteins with p-values? Any >> suggestion is appreciated. >> >> Kind regards, >> Yong >> >> _______________________________________________ >> 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
1
Entering edit mode
There's a thread on the bioconductor mailing list about using voom for RSEM-based RNA-seq quantification, in which Gordon Smythe explained that while voom() was designed for count data, it doesn't require it. As Tim Triche has suggested, if you're raw data is really ratios (and not absolute values for normal and tumor), then you should take log2 of those ratios and use limma from there; you can then also hijack the arrayQualityMetrics package to check QC (MA plots, mean-variance relationships, etc.) -Aaron On Tue, Jun 19, 2012 at 3:39 PM, Yong Li <mail.yong.li@googlemail.com>wrote: > Dear Aaron, > > thank you for your quick answer! I have checked the help page of > voom() but it seems to be used for count data. My data are just > tumor/normal ratios. I am wondering if you could provide more details? > > Best regards, > Yong > > On Tue, Jun 19, 2012 at 8:18 PM, Aaron Mackey <amackey@virginia.edu> > wrote: > > yes, it should be possible with a voom()-based analysis to get the > variances > > "right". > > > > -Aaron > > > > On Tue, Jun 19, 2012 at 12:47 PM, Yong Li <mail.yong.li@googlemail.com> > > wrote: > >> > >> Hello, > >> > >> limma has been so valuable in microarray data analysis, but has anyone > >> used limma for finding differentially expressed proteins from > >> quantitative proteomics data? The data I got has tumor/normal ratios > >> of thousands proteins, and both tumor and normal have a number of > >> replicates. Could such data be analyzed with limma? > >> > >> If limma can not be used here, what statistics method is suitable so > >> that we can get statistically significant proteins with p-values? Any > >> suggestion is appreciated. > >> > >> Kind regards, > >> Yong > >> > >> _______________________________________________ > >> 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'd suggest taking logit(ratio) instead of log2(ratio) as the former is quasi-normally-distributed (OK, logistic-normal) and the latter isn't. voom() would then presumably come up with the appropriate weights to deal with the remaining inflation/deflation WARNING: I don't use voom() on ratios, myself, so YMMV On Tue, Jun 19, 2012 at 2:09 PM, Aaron Mackey <amackey@virginia.edu> wrote: > There's a thread on the bioconductor mailing list about using voom for > RSEM-based RNA-seq quantification, in which Gordon Smythe explained that > while voom() was designed for count data, it doesn't require it. As Tim > Triche has suggested, if you're raw data is really ratios (and not absolute > values for normal and tumor), then you should take log2 of those ratios and > use limma from there; you can then also hijack the arrayQualityMetrics > package to check QC (MA plots, mean-variance relationships, etc.) > > -Aaron > > On Tue, Jun 19, 2012 at 3:39 PM, Yong Li <mail.yong.li@googlemail.com> >wrote: > > > Dear Aaron, > > > > thank you for your quick answer! I have checked the help page of > > voom() but it seems to be used for count data. My data are just > > tumor/normal ratios. I am wondering if you could provide more details? > > > > Best regards, > > Yong > > > > On Tue, Jun 19, 2012 at 8:18 PM, Aaron Mackey <amackey@virginia.edu> > > wrote: > > > yes, it should be possible with a voom()-based analysis to get the > > variances > > > "right". > > > > > > -Aaron > > > > > > On Tue, Jun 19, 2012 at 12:47 PM, Yong Li <mail.yong.li@googlemail.com> > > > > wrote: > > >> > > >> Hello, > > >> > > >> limma has been so valuable in microarray data analysis, but has anyone > > >> used limma for finding differentially expressed proteins from > > >> quantitative proteomics data? The data I got has tumor/normal ratios > > >> of thousands proteins, and both tumor and normal have a number of > > >> replicates. Could such data be analyzed with limma? > > >> > > >> If limma can not be used here, what statistics method is suitable so > > >> that we can get statistically significant proteins with p-values? Any > > >> suggestion is appreciated. > > >> > > >> Kind regards, > > >> Yong > > >> > > >> _______________________________________________ > > >> 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]] > > _______________________________________________ > 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 > -- *A model is a lie that helps you see the truth.* * * Howard Skipper<http: cancerres.aacrjournals.org="" content="" 31="" 9="" 1173.full.pdf=""> [[alternative HTML version deleted]]
ADD REPLY
0
Entering edit mode
oops, Aaron was exactly right, log2(ratio) would make much more sense. I had been looking at beta values (thus logit(beta) was on my mind) and in that case you do want logit(proportion). But log2(ratio) is more sensible. Again, apologies for the gaffe --t On Tue, Jun 19, 2012 at 2:50 PM, Tim Triche, Jr. <tim.triche@gmail.com>wrote: > I'd suggest taking logit(ratio) instead of log2(ratio) as the former is > quasi-normally-distributed (OK, logistic-normal) and the latter isn't. > voom() would then presumably come up with the appropriate weights to deal > with the remaining inflation/deflation > > WARNING: I don't use voom() on ratios, myself, so YMMV > > > > On Tue, Jun 19, 2012 at 2:09 PM, Aaron Mackey <amackey@virginia.edu>wrote: > >> There's a thread on the bioconductor mailing list about using voom for >> RSEM-based RNA-seq quantification, in which Gordon Smythe explained that >> while voom() was designed for count data, it doesn't require it. As Tim >> Triche has suggested, if you're raw data is really ratios (and not >> absolute >> values for normal and tumor), then you should take log2 of those ratios >> and >> use limma from there; you can then also hijack the arrayQualityMetrics >> package to check QC (MA plots, mean-variance relationships, etc.) >> >> -Aaron >> >> On Tue, Jun 19, 2012 at 3:39 PM, Yong Li <mail.yong.li@googlemail.com>> >wrote: >> >> > Dear Aaron, >> > >> > thank you for your quick answer! I have checked the help page of >> > voom() but it seems to be used for count data. My data are just >> > tumor/normal ratios. I am wondering if you could provide more details? >> > >> > Best regards, >> > Yong >> > >> > On Tue, Jun 19, 2012 at 8:18 PM, Aaron Mackey <amackey@virginia.edu> >> > wrote: >> > > yes, it should be possible with a voom()-based analysis to get the >> > variances >> > > "right". >> > > >> > > -Aaron >> > > >> > > On Tue, Jun 19, 2012 at 12:47 PM, Yong Li < >> mail.yong.li@googlemail.com> >> > > wrote: >> > >> >> > >> Hello, >> > >> >> > >> limma has been so valuable in microarray data analysis, but has >> anyone >> > >> used limma for finding differentially expressed proteins from >> > >> quantitative proteomics data? The data I got has tumor/normal ratios >> > >> of thousands proteins, and both tumor and normal have a number of >> > >> replicates. Could such data be analyzed with limma? >> > >> >> > >> If limma can not be used here, what statistics method is suitable so >> > >> that we can get statistically significant proteins with p-values? Any >> > >> suggestion is appreciated. >> > >> >> > >> Kind regards, >> > >> Yong >> > >> >> > >> _______________________________________________ >> > >> 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]] >> >> _______________________________________________ >> 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 >> > > > > -- > *A model is a lie that helps you see the truth.* > * > * > Howard Skipper<http: cancerres.aacrjournals.org="" content="" 31="" 9="" 1173.full.pdf=""> > > -- *A model is a lie that helps you see the truth.* * * Howard Skipper<http: cancerres.aacrjournals.org="" content="" 31="" 9="" 1173.full.pdf=""> [[alternative HTML version deleted]]
ADD REPLY
0
Entering edit mode
Dear Aaron, thank you and others for suggestions. My data is really ratios and not absolute values for normal and tumor. Sorry that I am still not quite sure how to move forward with limma when I take log2 of the ratios. It looks like I then will have the M component of the MAList, but how can I construct the A to make an MAList? Or I am missing something here? Kind regards, Yong On Tue, Jun 19, 2012 at 11:09 PM, Aaron Mackey <amackey at="" virginia.edu=""> wrote: > There's a thread on the bioconductor mailing list about using voom for > RSEM-based RNA-seq quantification, in which ?Gordon Smythe explained that > while voom() was designed for count data, it doesn't require it. ?As Tim > Triche has suggested, if you're raw data is really ratios (and not absolute > values for normal and tumor), then you should take log2 of those ratios and > use limma from there; you can then also hijack the arrayQualityMetrics > package to check QC (MA plots, mean-variance relationships, etc.) > > -Aaron > > On Tue, Jun 19, 2012 at 3:39 PM, Yong Li <mail.yong.li at="" googlemail.com=""> > wrote: >> >> Dear Aaron, >> >> thank you for your quick answer! I have checked the help page of >> voom() but it seems to be used for count data. My data are just >> tumor/normal ratios. I am wondering if you could provide more details? >> >> Best regards, >> Yong >> >> On Tue, Jun 19, 2012 at 8:18 PM, Aaron Mackey <amackey at="" virginia.edu=""> >> wrote: >> > yes, it should be possible with a voom()-based analysis to get the >> > variances >> > "right". >> > >> > -Aaron >> > >> > On Tue, Jun 19, 2012 at 12:47 PM, Yong Li <mail.yong.li at="" googlemail.com=""> >> > wrote: >> >> >> >> Hello, >> >> >> >> limma has been so valuable in microarray data analysis, but has anyone >> >> used limma for finding differentially expressed proteins from >> >> quantitative proteomics data? The data I got has tumor/normal ratios >> >> of thousands proteins, and both tumor and normal have a number of >> >> replicates. Could such data be analyzed with limma? >> >> >> >> If limma can not be used here, what statistics method is suitable so >> >> that we can get statistically significant proteins with p-values? Any >> >> suggestion is appreciated. >> >> >> >> Kind regards, >> >> Yong >> >> >> >> _______________________________________________ >> >> 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 >> > >> > >> >> _______________________________________________ >> 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 Yong, I don't think you need an MAList -- all limma functions will accept a simple matrix of your log2 ratios... or at least, all limma functions I have ever used, will do that... :-) Cheers, - axel Axel Klenk Research Informatician Actelion Pharmaceuticals Ltd / Gewerbestrasse 16 / CH-4123 Allschwil / Switzerland From: Yong Li <mail.yong.li at="" googlemail.com=""> To: bioconductor at r-project.org Date: 26.06.2012 00:01 Subject: Re: [BioC] Using limma for quantitative proteomics data Sent by: bioconductor-bounces at r-project.org Dear Aaron, thank you and others for suggestions. My data is really ratios and not absolute values for normal and tumor. Sorry that I am still not quite sure how to move forward with limma when I take log2 of the ratios. It looks like I then will have the M component of the MAList, but how can I construct the A to make an MAList? Or I am missing something here? Kind regards, Yong On Tue, Jun 19, 2012 at 11:09 PM, Aaron Mackey <amackey at="" virginia.edu=""> wrote: > There's a thread on the bioconductor mailing list about using voom for > RSEM-based RNA-seq quantification, in which Gordon Smythe explained that > while voom() was designed for count data, it doesn't require it. As Tim > Triche has suggested, if you're raw data is really ratios (and not absolute > values for normal and tumor), then you should take log2 of those ratios and > use limma from there; you can then also hijack the arrayQualityMetrics > package to check QC (MA plots, mean-variance relationships, etc.) > > -Aaron > > On Tue, Jun 19, 2012 at 3:39 PM, Yong Li <mail.yong.li at="" googlemail.com=""> > wrote: >> >> Dear Aaron, >> >> thank you for your quick answer! I have checked the help page of >> voom() but it seems to be used for count data. My data are just >> tumor/normal ratios. I am wondering if you could provide more details? >> >> Best regards, >> Yong >> >> On Tue, Jun 19, 2012 at 8:18 PM, Aaron Mackey <amackey at="" virginia.edu=""> >> wrote: >> > yes, it should be possible with a voom()-based analysis to get the >> > variances >> > "right". >> > >> > -Aaron >> > >> > On Tue, Jun 19, 2012 at 12:47 PM, Yong Li <mail.yong.li at="" googlemail.com=""> >> > wrote: >> >> >> >> Hello, >> >> >> >> limma has been so valuable in microarray data analysis, but has anyone >> >> used limma for finding differentially expressed proteins from >> >> quantitative proteomics data? The data I got has tumor/normal ratios >> >> of thousands proteins, and both tumor and normal have a number of >> >> replicates. Could such data be analyzed with limma? >> >> >> >> If limma can not be used here, what statistics method is suitable so >> >> that we can get statistically significant proteins with p-values? Any >> >> suggestion is appreciated. >> >> >> >> Kind regards, >> >> Yong >> >> >> >> _______________________________________________ >> >> 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 >> > >> > >> >> _______________________________________________ >> 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 > > _______________________________________________ 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 of this email and in any file transmitted with it is strictly confidential and may be legally privileged. It is intended solely for the addressee. If you are not the intended recipient, any copying, distribution or any other use of this email is prohibited and may be unlawful. In such case, you should please notify the sender immediately and destroy this email. The content of this email is not legally binding unless confirmed by letter. Any views expressed in this message are those of the individual sender, except where the message states otherwise and the sender is authorised to state them to be the views of the sender's company. For further information about Actelion please see our website at http://www.actelion.com
ADD REPLY
0
Entering edit mode
take the logit transform of the ratios and fit the variance inflation/deflation factors to that On Tue, Jun 19, 2012 at 12:39 PM, Yong Li <mail.yong.li@googlemail.com>wrote: > Dear Aaron, > > thank you for your quick answer! I have checked the help page of > voom() but it seems to be used for count data. My data are just > tumor/normal ratios. I am wondering if you could provide more details? > > Best regards, > Yong > > On Tue, Jun 19, 2012 at 8:18 PM, Aaron Mackey <amackey@virginia.edu> > wrote: > > yes, it should be possible with a voom()-based analysis to get the > variances > > "right". > > > > -Aaron > > > > On Tue, Jun 19, 2012 at 12:47 PM, Yong Li <mail.yong.li@googlemail.com> > > wrote: > >> > >> Hello, > >> > >> limma has been so valuable in microarray data analysis, but has anyone > >> used limma for finding differentially expressed proteins from > >> quantitative proteomics data? The data I got has tumor/normal ratios > >> of thousands proteins, and both tumor and normal have a number of > >> replicates. Could such data be analyzed with limma? > >> > >> If limma can not be used here, what statistics method is suitable so > >> that we can get statistically significant proteins with p-values? Any > >> suggestion is appreciated. > >> > >> Kind regards, > >> Yong > >> > >> _______________________________________________ > >> 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 > -- *A model is a lie that helps you see the truth.* * * Howard Skipper<http: cancerres.aacrjournals.org="" content="" 31="" 9="" 1173.full.pdf=""> [[alternative HTML version deleted]]
ADD REPLY
0
Entering edit mode
Dear Yong, It would be helpful if you could say something about the method used to identify differentially expressed proteins from quantitative proteomics data. Is it a protein microarray - if so which platform. Is it mass spec? I would think, and somebody please correct me if I am wrong, that continuous protein data could be analyzed similarly to contnuous mRNA data as far as differential expression goes - although preprocessing might be signficantly different. For example. I am currently analyzing a JPT peptide array and I am doing the preprocessing with Rapmad and the differential expression with Limma. with hopes that this helps, Rich On Jun 19, 2012, at 3:39 PM, Yong Li wrote: > Dear Aaron, > > thank you for your quick answer! I have checked the help page of > voom() but it seems to be used for count data. My data are just > tumor/normal ratios. I am wondering if you could provide more details? > > Best regards, > Yong > > On Tue, Jun 19, 2012 at 8:18 PM, Aaron Mackey <amackey at="" virginia.edu=""> > wrote: >> yes, it should be possible with a voom()-based analysis to get the >> variances >> "right". >> >> -Aaron >> >> On Tue, Jun 19, 2012 at 12:47 PM, Yong Li <mail.yong.li at="" googlemail.com="">> > >> wrote: >>> >>> Hello, >>> >>> limma has been so valuable in microarray data analysis, but has >>> anyone >>> used limma for finding differentially expressed proteins from >>> quantitative proteomics data? The data I got has tumor/normal ratios >>> of thousands proteins, and both tumor and normal have a number of >>> replicates. Could such data be analyzed with limma? >>> >>> If limma can not be used here, what statistics method is suitable so >>> that we can get statistically significant proteins with p-values? >>> Any >>> suggestion is appreciated. >>> >>> Kind regards, >>> Yong >>> >>> _______________________________________________ >>> 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 >> >> > > _______________________________________________ > 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
Bernd Fischer ▴ 550
@bernd-fischer-5348
Last seen 5.1 years ago
Germany / Heidelberg / DKFZ
Dear Yong! I used limma for ion count data. First I computed log-ratios per peptide and then summarized log-ratios per protein. Protein log-ratios were then analyzed by limma. Have a lock at our paper: Castello, Fischer, et al., Insights into RNA Biology from an Atlas of Mammalian mRNA-Binding Proteins, CELL, 2012 Best, Bernd On 06/19/2012 06:47 PM, Yong Li wrote: > Hello, > > limma has been so valuable in microarray data analysis, but has anyone > used limma for finding differentially expressed proteins from > quantitative proteomics data? The data I got has tumor/normal ratios > of thousands proteins, and both tumor and normal have a number of > replicates. Could such data be analyzed with limma? > > If limma can not be used here, what statistics method is suitable so > that we can get statistically significant proteins with p-values? Any > suggestion is appreciated. > > Kind regards, > Yong > > _______________________________________________ > 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
Dear Bernd, thank you for your answer and the paper. It was mentioned in the paper that the R script used for the analysis was in a Bioconductor data package named mRNAinteractomeHeLa, however I couldn't find it on the Bioc web site. The paper is new so probably I should wait for the next Bioc release? Best regards, Yong On Wed, Jun 20, 2012 at 8:09 AM, Bernd Fischer <bernd.fischer at="" embl.de=""> wrote: > Dear Yong! > > I used limma for ion count data. First I computed log-ratios per peptide and > then summarized log-ratios per protein. Protein log-ratios were then > analyzed > by limma. > Have a lock at our paper: > Castello, Fischer, et al., Insights into RNA Biology from an Atlas of > Mammalian > mRNA-Binding Proteins, CELL, 2012 > > Best, > Bernd > > > On 06/19/2012 06:47 PM, Yong Li wrote: >> >> Hello, >> >> limma has been so valuable in microarray data analysis, but has anyone >> used limma for finding differentially expressed proteins from >> quantitative proteomics data? The data I got has tumor/normal ratios >> of thousands proteins, and both tumor and normal have a number of >> replicates. Could such data be analyzed with limma? >> >> If limma can not be used here, what statistics method is suitable so >> that we can get statistically significant proteins with p-values? Any >> suggestion is appreciated. >> >> Kind regards, >> Yong >> >> _______________________________________________ >> 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 > > > _______________________________________________ > 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
Yong Li ▴ 80
@yong-li-5277
Last seen 7.4 years ago
Dear Axel, thanks for your answer. You are right, a matrix can be given to lmFit() and in this case just the Amean is not calculated in the returned object. Best regards, Yong On Tue, Jun 26, 2012 at 8:19 AM, <axel.klenk at="" actelion.com=""> wrote: > Dear Yong, > > I don't think you need an MAList -- all limma functions will accept a > simple matrix > of your log2 ratios... or at least, all limma functions I have ever used, > will do that... :-) > > Cheers, > > ?- axel > > > Axel Klenk > Research Informatician > Actelion Pharmaceuticals Ltd / Gewerbestrasse 16 / CH-4123 Allschwil / > Switzerland > > > > > From: > Yong Li <mail.yong.li at="" googlemail.com=""> > To: > bioconductor at r-project.org > Date: > 26.06.2012 00:01 > Subject: > Re: [BioC] Using limma for quantitative proteomics data > Sent by: > bioconductor-bounces at r-project.org > > > > Dear Aaron, > > thank you and others for suggestions. My data is really ratios and not > absolute values for normal and tumor. Sorry that I am still not quite > sure how to move forward with limma when I take log2 of the ratios. It > looks like I then will have the M component of the MAList, but how can > I construct the A to make an MAList? Or I am missing something here? > > Kind regards, > Yong > > On Tue, Jun 19, 2012 at 11:09 PM, Aaron Mackey <amackey at="" virginia.edu=""> > wrote: >> There's a thread on the bioconductor mailing list about using voom for >> RSEM-based RNA-seq quantification, in which ?Gordon Smythe explained > that >> while voom() was designed for count data, it doesn't require it. ?As Tim >> Triche has suggested, if you're raw data is really ratios (and not > absolute >> values for normal and tumor), then you should take log2 of those ratios > and >> use limma from there; you can then also hijack the arrayQualityMetrics >> package to check QC (MA plots, mean-variance relationships, etc.) >> >> -Aaron >> >> On Tue, Jun 19, 2012 at 3:39 PM, Yong Li <mail.yong.li at="" googlemail.com=""> >> wrote: >>> >>> Dear Aaron, >>> >>> thank you for your quick answer! I have checked the help page of >>> voom() but it seems to be used for count data. My data are just >>> tumor/normal ratios. I am wondering if you could provide more details? >>> >>> Best regards, >>> Yong >>> >>> On Tue, Jun 19, 2012 at 8:18 PM, Aaron Mackey <amackey at="" virginia.edu=""> >>> wrote: >>> > yes, it should be possible with a voom()-based analysis to get the >>> > variances >>> > "right". >>> > >>> > -Aaron >>> > >>> > On Tue, Jun 19, 2012 at 12:47 PM, Yong Li > <mail.yong.li at="" googlemail.com=""> >>> > wrote: >>> >> >>> >> Hello, >>> >> >>> >> limma has been so valuable in microarray data analysis, but has > anyone >>> >> used limma for finding differentially expressed proteins from >>> >> quantitative proteomics data? The data I got has tumor/normal ratios >>> >> of thousands proteins, and both tumor and normal have a number of >>> >> replicates. Could such data be analyzed with limma? >>> >> >>> >> If limma can not be used here, what statistics method is suitable so >>> >> that we can get statistically significant proteins with p-values? > Any >>> >> suggestion is appreciated. >>> >> >>> >> Kind regards, >>> >> Yong >>> >> >>> >> _______________________________________________ >>> >> 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 >>> > >>> > >>> >>> _______________________________________________ >>> 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 >> >> > > _______________________________________________ > 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 of this email and in any file transmitted with it is strictly confidential and may be legally privileged. > It is intended solely for the addressee. If you are not the intended recipient, any copying, distribution or any other use of this email is prohibited and may be unlawful. In such case, you should please notify the sender immediately and destroy this email. > The content of this email is not legally binding unless confirmed by letter. > Any views expressed in this message are those of the individual sender, except where the message states otherwise and the sender is authorised to state them to be the views of the sender's company. For further information about Actelion please see our website at http://www.actelion.com >
ADD COMMENT

Login before adding your answer.

Traffic: 481 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