Search
Question: heatmap Clustering help, two class (Control vs Exp) Experiments,
0
gravatar for SAURIN
7.1 years ago by
SAURIN1.1k
SAURIN1.1k wrote:
Hi BioC, I am using below function for heatmap. heatmap.2(FeatureX,col=gmpalette,Colv= FALSE,Rowv = as.dendrogram(hclu st(row.dist,method="centroid")),scale="row",key=TRUE,keysize=0.60,symk ey=FALSE,density.info="none",trace="none",margins=c(5,5),cexRow=1,cexC ol=1,cex.sub=1); Above function, I am getting Genes clustered but not samples and when I try to cluster samples, I get all control and exp. mixed up. how can I cluster samples but still be able to keep Controls and Exp. apart in terms of order. please let me know if anybody performed something like that. Thank you so much for your time and help in advance, Saurin
ADD COMMENTlink modified 7.1 years ago by Steve Lianoglou12k • written 7.1 years ago by SAURIN1.1k
0
gravatar for Steve Lianoglou
7.1 years ago by
Genentech
Steve Lianoglou12k wrote:
Hi, On Wed, Nov 3, 2010 at 12:16 PM, Saurin D. Jani <saurin_jani at="" yahoo.com=""> wrote: > Hi BioC, > > I am using below function for heatmap. > > heatmap.2(FeatureX,col=gmpalette,Colv= FALSE,Rowv = as.dendrogram(hc lust(row.dist,method="centroid")),scale="row",key=TRUE,keysize=0.60,sy mkey=FALSE,density.info="none",trace="none",margins=c(5,5),cexRow=1,ce xCol=1,cex.sub=1); > > > Above function, I am getting Genes clustered but not samples That's because you are setting Colv=FALSE in your call to heatmap.2(...) > and when I try to cluster samples, I get all control and exp. mixed up. > > how can I cluster samples but still be able to keep Controls and Exp. apart in terms of order. > > please let me know if anybody performed something like that. It seems your data is trying to tell you something that you don't want to hear ... -- 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 COMMENTlink written 7.1 years ago by Steve Lianoglou12k
Hi, You are right but When I do this: heatmap.2(FeatureX,col=gmpalette,Colv=as.dendrogram(hclust(col.dist,me thod="average")), Rowv=as.dendrogram(hclust(row.dist,method="average") ),scale="row",key=TRUE,keysize=0.60,symkey=FALSE,density.info="none",t race="none",margins=c(5,MapMargin),cexRow=1,cexCol=1,cex.sub=1); my control and exp. samples get mixed up..!! is there anyway I can pass a parameter ..not to do that just cluster samples on control and then exp. so, sorted view will be there. May be something like this: cluster control samples then exp. samples and then cluster based on Signal Intensity. so, I keep the order ctrl1,ctrl5,ctrl6,ctrl2,... and then Exp1,Exp5,Ex2,Exp10 ....so on... so, ctrl1 and ctrl 5 are next to each other and Exp1 and Exp5 are next to each other. Thanks, Saurin --- On Wed, 11/3/10, Steve Lianoglou <mailinglist.honeypot at="" gmail.com=""> wrote: > From: Steve Lianoglou <mailinglist.honeypot at="" gmail.com=""> > Subject: Re: [BioC] heatmap Clustering help, two class (Control vs Exp) Experiments, > To: saurin_jani at yahoo.com > Cc: "Bioconductor Bioconductor" <bioconductor at="" stat.math.ethz.ch=""> > Date: Wednesday, November 3, 2010, 12:34 PM > Hi, > > On Wed, Nov 3, 2010 at 12:16 PM, Saurin D. Jani <saurin_jani at="" yahoo.com=""> > wrote: > > Hi BioC, > > > > I am using below function for heatmap. > > > > heatmap.2(FeatureX,col=gmpalette,Colv= FALSE,Rowv = > as.dendrogram(hclust(row.dist,method="centroid")),scale="row",key=TR UE,keysize=0.60,symkey=FALSE,density.info="none",trace="none",margins= c(5,5),cexRow=1,cexCol=1,cex.sub=1); > > > > > > Above function, I am getting Genes clustered but not > samples > > That's because you are setting Colv=FALSE in your call to > heatmap.2(...) > > > and when I try to cluster samples, I get all control > and exp. mixed up. > > > > how can I cluster samples but still be able to keep > Controls and Exp. apart in terms of order. > > > > please let me know if anybody performed something like > that. > > It seems your data is trying to tell you something that you > don't want > to hear ... > > -- > 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 REPLYlink written 7.1 years ago by SAURIN1.1k
Hi Saurin, On Wed, Nov 3, 2010 at 12:43 PM, Saurin D. Jani <saurin_jani at="" yahoo.com=""> wrote: > Hi, > > You are right but When I do this: > > heatmap.2(FeatureX,col=gmpalette,Colv=as.dendrogram(hclust(col.dist, method="average")), Rowv=as.dendrogram(hclust(row.dist,method="average ")),scale="row",key=TRUE,keysize=0.60,symkey=FALSE,density.info="none" ,trace="none",margins=c(5,MapMargin),cexRow=1,cexCol=1,cex.sub=1); > > my control and exp. samples get mixed up..!! is there anyway I can pass a parameter ..not to do that just cluster samples on control and then exp. so, sorted view will be there. But why would you cluster the samples to begin with, if you just want to reorder them in some (your) arbitrary way? Assuming your data is properly nomralised, etc. and clustering your samples "mixes them up," then the heatmap is showing you visually that your treatment examples aren't "strikingly different" than your controls. Your data is trying to tell you that (apparently) all of these experiments kind of look (expression wise) like each other. Maybe that's telling you something about the quality of your data, or its annotation? Maybe you can try the plotPCA function in the affycoretools package as another way to see how your experiments "cluster together". I'm not sure that it would change things, but what happens if you remove all probes w/ low variance across your entire dataset and re-cluster them? > May be something like this: cluster control samples then exp. samples and then cluster based on Signal Intensity. so, I keep the order ctrl1,ctrl5,ctrl6,ctrl2,... and then Exp1,Exp5,Ex2,Exp10 ....so on... But then this is kind of misrepresenting what one would expect to see in such a plot .. you could, of course, plot and save heatmaps over just your control data, then again with just your experiment, then photoshop them together, but ... what's the point? I guess the question is: what are you trying to show in the heatmap you are trying to produce? Given that, people might be able to then suggest things you could try. -- 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 REPLYlink written 7.1 years ago by Steve Lianoglou12k
Hi Steve, you are right..ordering in my way...i thought clustering is same is ordering..! sorry, may be my description of problem was not clear. Here is why i want to do such a thing: for Example: I like to know if treatment T is making effect in listening music. I don't know which control samples were regularly listening to music at home. so, if geneX is over expressed. in Ctrl1 and Ctrl5 so they are grouped together and geneX is expressed normal levels in Ctrl20 then in heatmap those two ctrl samples are together. in Exp. sample..situation..same thing...if treatment T is given how music loving gene's intensity is behaving in different in exp. samples. I mean this is not new at all ..I am sure...somebody already has done this but I can't find something like that in order to display using heatmap !! Saurin --- On Wed, 11/3/10, Steve Lianoglou <mailinglist.honeypot at="" gmail.com=""> wrote: > From: Steve Lianoglou <mailinglist.honeypot at="" gmail.com=""> > Subject: Re: [BioC] heatmap Clustering help, two class (Control vs Exp) Experiments, > To: saurin_jani at yahoo.com > Cc: "Bioconductor Bioconductor" <bioconductor at="" stat.math.ethz.ch=""> > Date: Wednesday, November 3, 2010, 12:56 PM > Hi Saurin, > > On Wed, Nov 3, 2010 at 12:43 PM, Saurin D. Jani <saurin_jani at="" yahoo.com=""> > wrote: > > Hi, > > > > You are right but When I do this: > > > > > heatmap.2(FeatureX,col=gmpalette,Colv=as.dendrogram(hclust(col.dist, method="average")), > Rowv=as.dendrogram(hclust(row.dist,method="average")),scale="row",ke y=TRUE,keysize=0.60,symkey=FALSE,density.info="none",trace="none",marg ins=c(5,MapMargin),cexRow=1,cexCol=1,cex.sub=1); > > > > my control and exp. samples get mixed up..!! is there > anyway I can pass a parameter ..not to do that just cluster > samples on control and then exp. so, sorted view will be > there. > > But why would you cluster the samples to begin with, if you > just want > to reorder them in some (your) arbitrary way? > > Assuming your data is properly nomralised, etc. and > clustering your > samples "mixes them up," then the heatmap is showing you > visually that > your treatment examples aren't "strikingly different" than > your > controls. Your data is trying to tell you that (apparently) > all of > these experiments kind of look (expression wise) like each > other. > > Maybe that's telling you something about the quality of > your data, or > its annotation? > > Maybe you can try the plotPCA function in the affycoretools > package as > another way to see how your experiments "cluster > together". > > I'm not sure that it would change things, but what happens > if you > remove all probes w/ low variance across your entire > dataset and > re-cluster them? > > > May be something like this: cluster control samples > then exp. samples and then cluster based on Signal > Intensity. so, I keep the order ctrl1,ctrl5,ctrl6,ctrl2,... > and then Exp1,Exp5,Ex2,Exp10 ....so on... > > But then this is kind of misrepresenting what one would > expect to see > in such a plot .. you could, of course, plot and save > heatmaps over > just your control data, then again with just your > experiment, then > photoshop them together, but ... what's the point? > > I guess the question is: what are you trying to show in the > heatmap > you are trying to? produce? > > Given that, people might be able to then suggest things you > could try. > > -- > 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 REPLYlink written 7.1 years ago by SAURIN1.1k
Hi Steve, May be something like this: d d u | u u d n geneY u u d | u u d n geneX ---------------------- c1 c5 c10 | e1 e4 e8 e2 where: d = down, u = up, n = normal , c= control , e = exprimental original expression set has this order: c1,c2,c3...c10 e1,e2,e3..e10 Thank you so much, Saurin --- On Wed, 11/3/10, Steve Lianoglou <mailinglist.honeypot at="" gmail.com=""> wrote: > From: Steve Lianoglou <mailinglist.honeypot at="" gmail.com=""> > Subject: Re: [BioC] heatmap Clustering help, two class (Control vs Exp) Experiments, > To: saurin_jani at yahoo.com > Cc: "Bioconductor Bioconductor" <bioconductor at="" stat.math.ethz.ch=""> > Date: Wednesday, November 3, 2010, 12:56 PM > Hi Saurin, > > On Wed, Nov 3, 2010 at 12:43 PM, Saurin D. Jani <saurin_jani at="" yahoo.com=""> > wrote: > > Hi, > > > > You are right but When I do this: > > > > > heatmap.2(FeatureX,col=gmpalette,Colv=as.dendrogram(hclust(col.dist, method="average")), > Rowv=as.dendrogram(hclust(row.dist,method="average")),scale="row",ke y=TRUE,keysize=0.60,symkey=FALSE,density.info="none",trace="none",marg ins=c(5,MapMargin),cexRow=1,cexCol=1,cex.sub=1); > > > > my control and exp. samples get mixed up..!! is there > anyway I can pass a parameter ..not to do that just cluster > samples on control and then exp. so, sorted view will be > there. > > But why would you cluster the samples to begin with, if you > just want > to reorder them in some (your) arbitrary way? > > Assuming your data is properly nomralised, etc. and > clustering your > samples "mixes them up," then the heatmap is showing you > visually that > your treatment examples aren't "strikingly different" than > your > controls. Your data is trying to tell you that (apparently) > all of > these experiments kind of look (expression wise) like each > other. > > Maybe that's telling you something about the quality of > your data, or > its annotation? > > Maybe you can try the plotPCA function in the affycoretools > package as > another way to see how your experiments "cluster > together". > > I'm not sure that it would change things, but what happens > if you > remove all probes w/ low variance across your entire > dataset and > re-cluster them? > > > May be something like this: cluster control samples > then exp. samples and then cluster based on Signal > Intensity. so, I keep the order ctrl1,ctrl5,ctrl6,ctrl2,... > and then Exp1,Exp5,Ex2,Exp10 ....so on... > > But then this is kind of misrepresenting what one would > expect to see > in such a plot .. you could, of course, plot and save > heatmaps over > just your control data, then again with just your > experiment, then > photoshop them together, but ... what's the point? > > I guess the question is: what are you trying to show in the > heatmap > you are trying to? produce? > > Given that, people might be able to then suggest things you > could try. > > -- > 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 REPLYlink written 7.1 years ago by SAURIN1.1k
On Wed, Nov 3, 2010 at 2:07 PM, Saurin D. Jani <saurin_jani@yahoo.com>wrote: > Hi Steve, > > > May be something like this: > > d d u | u u d n geneY > u u d | u u d n geneX > ---------------------- > c1 c5 c10 | e1 e4 e8 e2 > > > where: d = down, u = up, n = normal , c= control , e = exprimental > > original expression set has this order: > c1,c2,c3...c10 e1,e2,e3..e10 > > Thank you so much, > Saurin > > > Hi, Saurin. You're going to need to write some code to do this, I think. I would suggest clustering your two sample groups, c and e, using hclust. Grab the order from that for each of the two groups, combine them in whatever way you like, figuring out the appropriate indexes for ALL the samples based on the separate hclust results, and then feeding that custom ordering to the heatmap function. I don't think this is as common a use case as you might think, so I doubt there is a simple canned solution. Sean > --- On Wed, 11/3/10, Steve Lianoglou <mailinglist.honeypot@gmail.com> > wrote: > > > From: Steve Lianoglou <mailinglist.honeypot@gmail.com> > > Subject: Re: [BioC] heatmap Clustering help, two class (Control vs Exp) > Experiments, > > To: saurin_jani@yahoo.com > > Cc: "Bioconductor Bioconductor" <bioconductor@stat.math.ethz.ch> > > Date: Wednesday, November 3, 2010, 12:56 PM > > Hi Saurin, > > > > On Wed, Nov 3, 2010 at 12:43 PM, Saurin D. Jani <saurin_jani@yahoo.com> > > wrote: > > > Hi, > > > > > > You are right but When I do this: > > > > > > > > > heatmap.2(FeatureX,col=gmpalette,Colv=as.dendrogram(hclust(col.dist, method="average")), > > > Rowv=as.dendrogram(hclust(row.dist,method="average")),scale="row",ke y=TRUE,keysize=0.60,symkey=FALSE, > density.info > ="none",trace="none",margins=c(5,MapMargin),cexRow=1,cexCol=1,cex.su b=1); > > > > > > my control and exp. samples get mixed up..!! is there > > anyway I can pass a parameter ..not to do that just cluster > > samples on control and then exp. so, sorted view will be > > there. > > > > But why would you cluster the samples to begin with, if you > > just want > > to reorder them in some (your) arbitrary way? > > > > Assuming your data is properly nomralised, etc. and > > clustering your > > samples "mixes them up," then the heatmap is showing you > > visually that > > your treatment examples aren't "strikingly different" than > > your > > controls. Your data is trying to tell you that (apparently) > > all of > > these experiments kind of look (expression wise) like each > > other. > > > > Maybe that's telling you something about the quality of > > your data, or > > its annotation? > > > > Maybe you can try the plotPCA function in the affycoretools > > package as > > another way to see how your experiments "cluster > > together". > > > > I'm not sure that it would change things, but what happens > > if you > > remove all probes w/ low variance across your entire > > dataset and > > re-cluster them? > > > > > May be something like this: cluster control samples > > then exp. samples and then cluster based on Signal > > Intensity. so, I keep the order ctrl1,ctrl5,ctrl6,ctrl2,... > > and then Exp1,Exp5,Ex2,Exp10 ....so on... > > > > But then this is kind of misrepresenting what one would > > expect to see > > in such a plot .. you could, of course, plot and save > > heatmaps over > > just your control data, then again with just your > > experiment, then > > photoshop them together, but ... what's the point? > > > > I guess the question is: what are you trying to show in the > > heatmap > > you are trying to produce? > > > > Given that, people might be able to then suggest things you > > could try. > > > > -- > > 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 > > > > > > > _______________________________________________ > Bioconductor mailing list > Bioconductor@stat.math.ethz.ch > https://stat.ethz.ch/mailman/listinfo/bioconductor > Search the archives: > http://news.gmane.org/gmane.science.biology.informatics.conductor > [[alternative HTML version deleted]]
ADD REPLYlink written 7.1 years ago by Sean Davis21k
Hi Sean, Thank you so much for your help and guidance in this. Your solution sounds good...I will try my best to code this one out..!! Thank you again, Saurin --- On Wed, 11/3/10, Sean Davis <sdavis2@mail.nih.gov> wrote: From: Sean Davis <sdavis2@mail.nih.gov> Subject: Re: [BioC] heatmap Clustering help, two class (Control vs Exp) Experiments, To: saurin_jani@yahoo.com Cc: "Steve Lianoglou" <mailinglist.honeypot@gmail.com>, "Bioconductor Bioconductor" <bioconductor@stat.math.ethz.ch> Date: Wednesday, November 3, 2010, 2:30 PM On Wed, Nov 3, 2010 at 2:07 PM, Saurin D. Jani <saurin_jani@yahoo.com> wrote: Hi Steve, May be something like this: �d �d �u � | u � u �d n geneY �u �u �d � | u � u �d n geneX ---------------------- c1 c5 c10 �| e1 e4 e8 e2 where: d = down, u = up, n = normal , c= control , e = exprimental original expression set has this order: c1,c2,c3...c10 e1,e2,e3..e10 Thank you so much, Saurin Hi, Saurin. You're going to need to write some code to do this, I think. �I would suggest clustering your two sample groups, c and e, using hclust. �Grab the order from that for each of the two groups, combine them in whatever way you like, figuring out the appropriate indexes for ALL the samples based on the separate hclust results, and then feeding that custom ordering to the heatmap function. �I don't think this is as common a use case as you might think, so I doubt there is a simple canned solution. Sean � --- On Wed, 11/3/10, Steve Lianoglou <mailinglist.honeypot@gmail.com> wrote: > From: Steve Lianoglou <mailinglist.honeypot@gmail.com> > Subject: Re: [BioC] heatmap Clustering help, two class (Control vs Exp) Experiments, > To: saurin_jani@yahoo.com > Cc: "Bioconductor Bioconductor" <bioconductor@stat.math.ethz.ch> > Date: Wednesday, November 3, 2010, 12:56 PM > Hi Saurin, > > On Wed, Nov 3, 2010 at 12:43 PM, Saurin D. Jani <saurin_jani@yahoo.com> > wrote: > > Hi, > > > > You are right but When I do this: > > > > > heatmap.2(FeatureX,col=gmpalette,Colv=as.dendrogram(hclust(col.dist, method="average")), > Rowv=as.dendrogram(hclust(row.dist,method="average")),scale="row",ke y=TRUE,keysize=0.60,symkey=FALSE,density.info="none",trace="none",marg ins=c(5,MapMargin),cexRow=1,cexCol=1,cex.sub=1); > > > > my control and exp. samples get mixed up..!! is there > anyway I can pass a parameter ..not to do that just cluster > samples on control and then exp. so, sorted view will be > there. > > But why would you cluster the samples to begin with, if you > just want > to reorder them in some (your) arbitrary way? > > Assuming your data is properly nomralised, etc. and > clustering your > samples "mixes them up," then the heatmap is showing you > visually that > your treatment examples aren't "strikingly different" than > your > controls. Your data is trying to tell you that (apparently) > all of > these experiments kind of look (expression wise) like each > other. > > Maybe that's telling you something about the quality of > your data, or > its annotation? > > Maybe you can try the plotPCA function in the affycoretools > package as > another way to see how your experiments "cluster > together". > > I'm not sure that it would change things, but what happens > if you > remove all probes w/ low variance across your entire > dataset and > re-cluster them? > > > May be something like this: cluster control samples > then exp. samples and then cluster based on Signal > Intensity. so, I keep the order ctrl1,ctrl5,ctrl6,ctrl2,... > and then Exp1,Exp5,Ex2,Exp10 ....so on... > > But then this is kind of misrepresenting what one would > expect to see > in such a plot .. you could, of course, plot and save > heatmaps over > just your control data, then again with just your > experiment, then > photoshop them together, but ... what's the point? > > I guess the question is: what are you trying to show in the > heatmap > you are trying to� produce? > > Given that, people might be able to then suggest things you > could try. > > -- > 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 > _______________________________________________ Bioconductor mailing list Bioconductor@stat.math.ethz.ch https://stat.ethz.ch/mailman/listinfo/bioconductor Search the archives: http://news.gmane.org/gmane.science.biology.informatics.conductor [[alternative HTML version deleted]]
ADD REPLYlink written 7.1 years ago by SAURIN1.1k
Hi, heatmap clustering two class ..Solved:  Thanks Sean,Thoma, Steve...for your help. For Thread info. here is code:  FeatureX <- esetSub2X; # DE genes Exp. set   eR <- nrow(FeatureX);   eC <- ncol(FeatureX);   #HSET <- matrix(data = 0, nrow = eR, ncol = eC);    HSET <- numeric();    Hcols <- character();   #row.dist <- as.dist(1 - cor(t(FeatureX)));   row.dist <- as.dist(1 - cor(t(FeatureX)));   col.dist <- as.dist(1 - cor(FeatureX)); ############### building custom Heatmap set   clust_col <- hclust(col.dist,method="average");   cORD <- as.numeric(clust_col$order);   smps <- colnames(FeatureX);   n_smps <- smps[w:x];   t_smps <- smps[y:z]; #----------------------------------------------------------- # build Normal samples  for(f in 1:cORD)   {      ns = colnames(FeatureX)[cORD[f]];     if(is.element(ns,n_smps))        {         HSET <- cbind(HSET,FeatureX[,ns]);         Hcols <- c(Hcols,ns);        }    } # build Tumor samples  for(f in 1:cORD)   {      ts = colnames(FeatureX)[cORD[f]];     if(is.element(ts,t_smps))        {         HSET <- cbind(HSET,FeatureX[,ts]);         Hcols <- c(Hcols,ts);        }    } colnames(HSET) <- Hcols; FeatureX <- HSET; heatmap.2(FeatureX); #----------------------------------------------------------- On Wed, Nov 3, 2010 at 2:07 PM, Saurin D. Jani <saurin_jani@yahoo.com> wrote: Hi Steve, May be something like this:  d  d  u   | u   u  d n geneY  u  u  d   | u   u  d n geneX ---------------------- c1 c5 c10  | e1 e4 e8 e2 where: d = down, u = up, n = normal , c= control , e = exprimental original expression set has this order: c1,c2,c3...c10 e1,e2,e3..e10 Thank you so much, Saurin Hi, Saurin. You're going to need to write some code to do this, I think.  I would suggest clustering your two sample groups, c and e, using hclust. Grab the order from that for each of the two groups, combine them in whatever way you like, figuring out the appropriate indexes for ALL the samples based on the separate hclust results, and then feeding that custom ordering to the heatmap function.  I don't think this is as common a use case as you might think, so I doubt there is a simple canned solution. Sean --- On Wed, 11/3/10, Steve Lianoglou <mailinglist.honeypot@gmail.com> wrote: > From: Steve Lianoglou <mailinglist.honeypot@gmail.com> > Subject: Re: [BioC] heatmap Clustering help, two class (Control vs Exp) Experiments, > To: saurin_jani@yahoo.com > Cc: "Bioconductor Bioconductor" <bioconductor@stat.math.ethz.ch> > Date: Wednesday, November 3, 2010, 12:56 PM > Hi Saurin, > > On Wed, Nov 3, 2010 at 12:43 PM, Saurin D. Jani <saurin_jani@yahoo.com> > wrote: > > Hi, > > > > You are right but When I do this: > > > > > heatmap.2(FeatureX,col=gmpalette,Colv=as.dendrogram(hclust(col.dist, method="average")), > Rowv=as.dendrogram(hclust(row.dist,method="average")),scale="row",ke y=TRUE,keysize=0.60,symkey=FALSE,density.info="none",trace="none",marg ins=c(5,MapMargin),cexRow=1,cexCol=1,cex.sub=1); > > > > my control and exp. samples get mixed up..!! is there > anyway I can pass a parameter ..not to do that just cluster > samples on control and then exp. so, sorted view will be > there. > > But why would you cluster the samples to begin with, if you > just want > to reorder them in some (your) arbitrary way? > > Assuming your data is properly nomralised, etc. and > clustering your > samples "mixes them up," then the heatmap is showing you > visually that > your treatment examples aren't "strikingly different" than > your > controls. Your data is trying to tell you that (apparently) > all of > these experiments kind of look (expression wise) like each > other. > > Maybe that's telling you something about the quality of > your data, or > its annotation? > > Maybe you can try the plotPCA function in the affycoretools > package as > another way to see how your experiments "cluster > together". > > I'm not sure that it would change things, but what happens > if you > remove all probes w/ low variance across your entire > dataset and > re-cluster them? > > > May be something like this: cluster control samples > then exp. samples and then cluster based on Signal > Intensity. so, I keep the order ctrl1,ctrl5,ctrl6,ctrl2,... > and then Exp1,Exp5,Ex2,Exp10 ....so on... > > But then this is kind of misrepresenting what one would > expect to see > in such a plot .. you could, of course, plot and save > heatmaps over > just your control data, then again with just your > experiment, then > photoshop them together, but ... what's the point? > > I guess the question is: what are you trying to show in the > heatmap > you are trying to  produce? > > Given that, people might be able to then suggest things you > could try. > > -- > 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 > _______________________________________________ Bioconductor mailing list Bioconductor@stat.math.ethz.ch https://stat.ethz.ch/mailman/listinfo/bioconductor Search the archives: http://news.gmane.org/gmane.science.biology.informatics.conductor [[alternative HTML version deleted]]
ADD REPLYlink written 7.1 years ago by SAURIN1.1k
Please log in to add an answer.

Help
Access

Use of this site constitutes acceptance of our User Agreement and Privacy Policy.
Powered by Biostar version 2.2.0
Traffic: 258 users visited in the last hour