Tweet
Hello, I am new to GO analysis and I am a little bit confused.
I have some mass spec data. The output data is a list of gene names (UNIPROT) and a value for the control and experimental. I have normalized all data to the control for each protein in the study. After taking log2, I have a +/- list of expression data for all proteins in my study for the three experimental conditions.
I want to see what genes show up in what major cellular/biochemical/disease categories and visual their enrichment level. So basically if say 10 different genes are known to be involved in a specific disease phenotype or biochemical pathway they would cluster together in a dendogram. From what I understand then, I need a GO clustering tool to see my data and a program that can generate clustered heat maps to visualize the expression level for each gene.
The problem is that there are so many programs available and I don't really understand the differences. I have been told that DAVID is probably the best tool for clustering my data (and well-accepted in the literature). But then I have seen people use Cluster 3.0 - which does not seems to be the same kind of clustering that DAVID does - and then view in Java TreeView.
Do I use the DAVID "functional annotation clustering" and then import into TreeView? This is what seems to make the most sense to me but I am not sure...
You can probably tell I am really confused. Any help would be appreciated.
EDIT: I am now aware that Cluster 3.0 apparently clusters hierarchically based on expression data not gene ontology so that explains my first bit of confusion. I found a nice Nature summary of expression clustering so I think I understand how it works now.
As for GO clustering though, I would still like to know if using DAVID + TreeView is a good idea. And is there a way to show both expression clustering and GO clustering on the same heatmap? I can't visualize what this would look like but I don't think it would work.
I have some mass spec data. The output data is a list of gene names (UNIPROT) and a value for the control and experimental. I have normalized all data to the control for each protein in the study. After taking log2, I have a +/- list of expression data for all proteins in my study for the three experimental conditions.
I want to see what genes show up in what major cellular/biochemical/disease categories and visual their enrichment level. So basically if say 10 different genes are known to be involved in a specific disease phenotype or biochemical pathway they would cluster together in a dendogram. From what I understand then, I need a GO clustering tool to see my data and a program that can generate clustered heat maps to visualize the expression level for each gene.
The problem is that there are so many programs available and I don't really understand the differences. I have been told that DAVID is probably the best tool for clustering my data (and well-accepted in the literature). But then I have seen people use Cluster 3.0 - which does not seems to be the same kind of clustering that DAVID does - and then view in Java TreeView.
Do I use the DAVID "functional annotation clustering" and then import into TreeView? This is what seems to make the most sense to me but I am not sure...
You can probably tell I am really confused. Any help would be appreciated.
EDIT: I am now aware that Cluster 3.0 apparently clusters hierarchically based on expression data not gene ontology so that explains my first bit of confusion. I found a nice Nature summary of expression clustering so I think I understand how it works now.
As for GO clustering though, I would still like to know if using DAVID + TreeView is a good idea. And is there a way to show both expression clustering and GO clustering on the same heatmap? I can't visualize what this would look like but I don't think it would work.