Hi Jabez, Jabez Wilson wrote: > Sorry.... sent too soon. > > Dear all, I know that this question has been asked in a couple of > forms, but I haven't noticed a full reply given. I'm hoping that > someone will be able to give me the exact answer. I'm comparing > stimulated cells vs unstimulated cells on each slide at two time > points (4 hrs and 8 hrs). Suppose there are 4 samples at each time > point the targets file will look like this: > > FileName cy3 cy5 1 stim4 neg4 2 neg4 > stim4 3 stim4 neg4 4 neg4 stim4 > > 5 stim8 neg8 6 neg8 stim8 7 > stim8 neg8 8 neg8 stim8 > > There is no common reference (pool) as there is in the weaver example > in the limma guide, so should I use e.g. neg4 as the reference i.e. > > design <- modelMatrix(targets,ref="neg4") > > If I do that then when I fit the model using lmFit(MA, design) I get > > "Coefficients not estimable: stim8 " You can't use neg4 as a reference when it isn't actually a reference (e.g., it has to be on every slide). If you create a design matrix this way you will get > modelMatrix(targets, ref="neg4") Found unique target names: neg4 neg8 stim4 stim8 neg8 stim4 stim8 [1,] 0 -1 0 [2,] 0 1 0 [3,] 0 -1 0 [4,] 0 1 0 [5,] 1 0 -1 [6,] -1 0 1 [7,] 1 0 -1 [8,] -1 0 1 Which is not of full rank (e.g., the stim8 column is a linear combination of the neg8 column). You don't give any information about your experiment, so it is difficult to help. In addition, people are in general hesitant to help people with experimental design or analysis questions because a.) that is what many of us do for a living, so you are in effect asking for pro bono work, and b.) without knowing more about a given experiment it isn't reasonable for people to give analysis advice anyway. So, without knowing more about your experiment other than the short names you gave your treatments, can you not simply analyze the '4' samples separately from the '8' samples, using a reference design in each case? Or if the negX samples are all supposed to be similar in expression (and you can show they are), you could rename them 'neg' and then have a true reference design. Best, Jim > > Can anyone help me from this point (apart from advising me to do the > microarray expt with affymetrix chips)? > > > > > > [[alternative HTML version deleted]] > > > > -------------------------------------------------------------------- ---- > > > _______________________________________________ Bioconductor mailing > list Bioconductor at stat.math.ethz.ch > 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 Douglas Lab University of Michigan Department of Human Genetics 5912 Buhl 1241 E. Catherine St. Ann Arbor MI 48109-5618 734-615-7826

The simplest way to handle this type of design is to use the single channel analysis in limma. This allows you to use array as a block, and then do the treatment contrasts in the usual way. There is nothing wrong with the experimental design. It is an incomplete block design that allows the greatest precision for the stimulant effect within time. However, the main effect of time and the time by stimulant interaction can all be assessed. --Naomi At 10:24 AM 5/22/2009, Jabez Wilson wrote: >Thanks for you prompt reply, James. I'm familiar with you from your >postings on the SAM mailing lists, so am honoured that you can reply >to my query. >Whilst I don't want to take bread out of honest statisticians >mouths, I've been asked to analyse using limma an experiment that's >already taken place - the design was down to someone else and >obviously from your reply should have been done in a different way. >The experiment is quite simple in that spleen cells are incubated >with or without a stimulant (I think it's the ppd of the TB vaccine) >for either 4 or 8 hours, and the contrast of interest is whether >there is a difference in gene expression in the stimulated cells >between the two time periods. >I thought about analysing it using the "negs" as a common reference >as you suggest, but that would assume that there is no change in >expression in the unstimulated samples, which it not an assumption >that now can be proved one way or the other. >Up until now, I had been analysing them seperately, and just >comparing the lists of expressed genes, but thought that there could >be more information obtained by trying to "link" them. If it could >be done, then I could also answer the question of whether the >unstimulated "negs" are in fact different at the two time points. >I take it from your reply that this cannot actually be done now. > >Thanks again for replying, > >Jabez >--- On Fri, 22/5/09, James W. MacDonald <jmacdon at="" med.umich.edu=""> wrote: > > >From: James W. MacDonald <jmacdon at="" med.umich.edu=""> >Subject: Re: [BioC] Limma analyse factorial data with two-color arrays >To: "Jabez Wilson" <jabezwuk at="" yahoo.co.uk=""> >Cc: "bioconductor" <bioconductor at="" stat.math.ethz.ch=""> >Date: Friday, 22 May, 2009, 2:12 PM > > >Hi Jabez, > >Jabez Wilson wrote: > > Sorry.... sent too soon. > > > > Dear all, I know that this question has been asked in a couple of > > forms, but I haven't noticed a full reply given. I'm hoping that > > someone will be able to give me the exact answer. I'm comparing > > stimulated cells vs unstimulated cells on each slide at two time > > points (4 hrs and 8 hrs). Suppose there are 4 samples at each time > > point the targets file will look like this: > > > > FileName cy3 cy5 1 stim4 neg4 2 neg4 > > stim4 3 stim4 neg4 4 neg4 stim4 > > > > 5 stim8 neg8 6 neg8 stim8 7 > > stim8 neg8 8 neg8 stim8 > > > > There is no common reference (pool) as there is in the weaver example > > in the limma guide, so should I use e.g. neg4 as the reference i.e. > > > > design <- modelMatrix(targets,ref="neg4") > > > > If I do that then when I fit the model using lmFit(MA, design) I get > > > > "Coefficients not estimable: stim8 " > >You can't use neg4 as a reference when it isn't actually a reference >(e.g., it has to be on every slide). If you create a design matrix >this way you will get > > > modelMatrix(targets, ref="neg4") >Found unique target names: >neg4 neg8 stim4 stim8 > neg8 stim4 stim8 >[1,] 0 -1 0 >[2,] 0 1 0 >[3,] 0 -1 0 >[4,] 0 1 0 >[5,] 1 0 -1 >[6,] -1 0 1 >[7,] 1 0 -1 >[8,] -1 0 1 > >Which is not of full rank (e.g., the stim8 column is a linear >combination of the neg8 column). > >You don't give any information about your experiment, so it is >difficult to help. In addition, people are in general hesitant to >help people with experimental design or analysis questions because >a.) that is what many of us do for a living, so you are in effect >asking for pro bono work, and b.) without knowing more about a given >experiment it isn't reasonable for people to give analysis advice anyway. > >So, without knowing more about your experiment other than the short >names you gave your treatments, can you not simply analyze the '4' >samples separately from the '8' samples, using a reference design in >each case? Or if the negX samples are all supposed to be similar in >expression (and you can show they are), you could rename them 'neg' >and then have a true reference design. > >Best, > >Jim > > > > > > Can anyone help me from this point (apart from advising me to do the > > microarray expt with affymetrix chips)? > > > > > > > > > > > > [[alternative HTML version deleted]] > > > > > > > > ------------------------------------------------------------------ ------ > > > > > > _______________________________________________ Bioconductor mailing > > list Bioconductor at stat.math.ethz.ch > 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 >Douglas Lab >University of Michigan >Department of Human Genetics >5912 Buhl >1241 E. Catherine St. >Ann Arbor MI 48109-5618 >734-615-7826 > > > > > [[alternative HTML version deleted]] > >_______________________________________________ >Bioconductor mailing list >Bioconductor at stat.math.ethz.ch >https://stat.ethz.ch/mailman/listinfo/bioconductor >Search the archives: >http://news.gmane.org/gmane.science.biology.informatics.conductor Naomi S. Altman 814-865-3791 (voice) Associate Professor Dept. of Statistics 814-863-7114 (fax) Penn State University 814-865-1348 (Statistics) University Park, PA 16802-2111

Limma codes for only one level of blocking. So, I guess you will need to stick with 2 channel analysis with blocks to deal with this. Fortunately, for a 2x2 design, things are still (pretty) good. Assuming that M4=stim4-neg4 and M8=stim8-neg8 and A4 and A8 are defined with +. The main effect of stim is the average of M4 and M8. The interaction of stim with time is M8-M4. But I do not think you can estimate the time main effect, which is A8-A4 in this analysis. Note that A is not normalized. --Naomi At 07:51 AM 5/27/2009, Jabez Wilson wrote: >Thanks for that suggestion, Naomi, I take it you mean "separate" >channel analysis in the limma users guide. There's a complication, >in that there are technical dye-swap replicates for each sample - I >can't currently see how I can integrate this into the code. Could >you perhaps suggest how and whether I can do this. >TIA >Jabez > >--- On Sat, 23/5/09, Naomi Altman <naomi at="" stat.psu.edu=""> wrote: > > >From: Naomi Altman <naomi at="" stat.psu.edu=""> >Subject: Re: [BioC] Limma analyse factorial data with two-color arrays >To: "Jabez Wilson" <jabezwuk at="" yahoo.co.uk="">, "James W. MacDonald" ><jmacdon at="" med.umich.edu=""> >Cc: "bioconductor" <bioconductor at="" stat.math.ethz.ch=""> >Date: Saturday, 23 May, 2009, 5:07 AM > > >The simplest way to handle this type of design is to use the single >channel analysis in limma. This allows you to use array as a block, >and then do the treatment contrasts in the usual way. > >There is nothing wrong with the experimental design. It is an >incomplete block design that allows the greatest precision for the >stimulant effect within time. However, the main effect of time and >the time by stimulant interaction can all be assessed. > >--Naomi > >At 10:24 AM 5/22/2009, Jabez Wilson wrote: > >Thanks for you prompt reply, James. I'm familiar with you from your > >postings on the SAM mailing lists, so am honoured that you can reply > >to my query. > >Whilst I don't want to take bread out of honest statisticians > >mouths, I've been asked to analyse using limma an experiment that's > >already taken place - the design was down to someone else and > >obviously from your reply should have been done in a different way. > >The experiment is quite simple in that spleen cells are incubated > >with or without a stimulant (I think it's the ppd of the TB vaccine) > >for either 4 or 8 hours, and the contrast of interest is whether > >there is a difference in gene expression in the stimulated cells > >between the two time periods. > >I thought about analysing it using the "negs" as a common reference > >as you suggest, but that would assume that there is no change in > >expression in the unstimulated samples, which it not an assumption > >that now can be proved one way or the other. > >Up until now, I had been analysing them seperately, and just > >comparing the lists of expressed genes, but thought that there could > >be more information obtained by trying to "link" them. If it could > >be done, then I could also answer the question of whether the > >unstimulated "negs" are in fact different at the two time points. > >I take it from your reply that this cannot actually be done now. > > > >Thanks again for replying, > > > >Jabez > >--- On Fri, 22/5/09, James W. MacDonald <jmacdon at="" med.umich.edu=""> wrote: > > > > > >From: James W. MacDonald <jmacdon at="" med.umich.edu=""> > >Subject: Re: [BioC] Limma analyse factorial data with two-color arrays > >To: "Jabez Wilson" <jabezwuk at="" yahoo.co.uk=""> > >Cc: "bioconductor" <bioconductor at="" stat.math.ethz.ch=""> > >Date: Friday, 22 May, 2009, 2:12 PM > > > > > >Hi Jabez, > > > >Jabez Wilson wrote: > > > Sorry.... sent too soon. > > > > > > Dear all, I know that this question has been asked in a couple of > > > forms, but I haven't noticed a full reply given. I'm hoping that > > > someone will be able to give me the exact answer. I'm comparing > > > stimulated cells vs unstimulated cells on each slide at two time > > > points (4 hrs and 8 hrs). Suppose there are 4 samples at each time > > > point the targets file will look like this: > > > > > > FileName cy3 cy5 1 stim4 neg4 2 neg4 > > > stim4 3 stim4 neg4 4 neg4 stim4 > > > > > > 5 stim8 neg8 6 neg8 stim8 7 > > > stim8 neg8 8 neg8 stim8 > > > > > > There is no common reference (pool) as there is in the weaver example > > > in the limma guide, so should I use e.g. neg4 as the reference i.e. > > > > > > design <- modelMatrix(targets,ref="neg4") > > > > > > If I do that then when I fit the model using lmFit(MA, design) I get > > > > > > "Coefficients not estimable: stim8 " > > > >You can't use neg4 as a reference when it isn't actually a reference > >(e.g., it has to be on every slide). If you create a design matrix > >this way you will get > > > > > modelMatrix(targets, ref="neg4") > >Found unique target names: > >neg4 neg8 stim4 stim8 > > neg8 stim4 stim8 > >[1,] 0 -1 0 > >[2,] 0 1 0 > >[3,] 0 -1 0 > >[4,] 0 1 0 > >[5,] 1 0 -1 > >[6,] -1 0 1 > >[7,] 1 0 -1 > >[8,] -1 0 1 > > > >Which is not of full rank (e.g., the stim8 column is a linear > >combination of the neg8 column). > > > >You don't give any information about your experiment, so it is > >difficult to help. In addition, people are in general hesitant to > >help people with experimental design or analysis questions because > >a.) that is what many of us do for a living, so you are in effect > >asking for pro bono work, and b.) without knowing more about a given > >experiment it isn't reasonable for people to give analysis advice anyway. > > > >So, without knowing more about your experiment other than the short > >names you gave your treatments, can you not simply analyze the '4' > >samples separately from the '8' samples, using a reference design in > >each case? Or if the negX samples are all supposed to be similar in > >expression (and you can show they are), you could rename them 'neg' > >and then have a true reference design. > > > >Best, > > > >Jim > > > > > > > > > > Can anyone help me from this point (apart from advising me to do the > > > microarray expt with affymetrix chips)? > > > > > > > > > > > > > > > > > > [[alternative HTML version deleted]] > > > > > > > > > > > > ---------------------------------------------------------------- -------- > > > > > > > > > _______________________________________________ Bioconductor mailing > > > list Bioconductor at stat.math.ethz.ch > > 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 > >Douglas Lab > >University of Michigan > >Department of Human Genetics > >5912 Buhl > >1241 E. Catherine St. > >Ann Arbor MI 48109-5618 > >734-615-7826 > > > > > > > > > > [[alternative HTML version deleted]] > > > >_______________________________________________ > >Bioconductor mailing list > >Bioconductor at stat.math.ethz.ch > >https://stat.ethz.ch/mailman/listinfo/bioconductor > >Search the archives: > >http://news.gmane.org/gmane.science.biology.informatics.conductor > >Naomi S. Altman 814-865-3791 (voice) >Associate Professor >Dept. of Statistics 814-863-7114 (fax) >Penn State University 814-865-1348 (Statistics) >University Park, PA 16802-2111 > > > > > > [[alternative HTML version deleted]] > >_______________________________________________ >Bioconductor mailing list >Bioconductor at stat.math.ethz.ch >https://stat.ethz.ch/mailman/listinfo/bioconductor >Search the archives: >http://news.gmane.org/gmane.science.biology.informatics.conductor Naomi S. Altman 814-865-3791 (voice) Associate Professor Dept. of Statistics 814-863-7114 (fax) Penn State University 814-865-1348 (Statistics) University Park, PA 16802-2111