I need to learn R - Any advice? (Apr/23/2008 )
Hey hey 
Turns out our "Bioinformatics expert" can't (or won't) help us analyse our chip-chip results! <insert huge drama here>
Sooooooo...I need to learn R! Does anyone have any advice, or can you recommend a dummy's guide to R? As a molecular biologist, I am unsure whether my brain is designed for programming!!!
Thanks in advance!
Clare
Hi!
Well, a colleague and me also have/had to get into that story a bit for array analyses.
What we did was following basically the instructions that are given in the respective Vignettes (help documents telling you about the commands you need).
Basically you can start from scratch by loading in the .cel files (I suppose it is cel-format) and start with doing a check for the quality of the data and go on and normalise it. For all of these procedures you can get Vignettes that kind of guide you through the process.
Once you are done with quality control and normalisation steps you can read out these expression data as Tab delimited file.
This file can then be further process by TigrTool, which also allows you to do SAM-analysis.
I have to say, we are also no experts at all (!), however we managed to get to certain points ourselves pretty good.
What makes it difficult is the possibilites that you have to analyse Chip-data - e.g which normalisation method one can use and what impact this would have on the Chip-data.
However, once you know which method might work best or is used in your lab anyways it can be really helpful.
I've sent respective links to your PM-box
Cheers
i am also looking for an easy way to learn 'R'. all i know is that you need to know the 'S' language of computer to run the program. i downloaded the help file of R and didn't understand a word. so i would also be very interested in any solutions you might come up with about using R
'S' language!!?? Is this why most of the R instructions made no sense at all? Oh dear. This is exactly why I think Bioinformatics divisions in an institute are essential. They love this stuff! <my poor brain>
hello clare,
in a previous post i added the manual for R. i don't know about chip-chip analysis but i'm aware there are preformed packages for R that deal with that kind of stuff, or at least for microarrays, maybe a few modifications on those packages would do.
R doesn't look friendly at all. doesn't excel work for your analysis? you could check out this page:
http://phoenix.phys.clemson.edu/tutorials/excel/
hope this helps.
cheers.
in a previous post i added the manual for R. i don't know about chip-chip analysis but i'm aware there are preformed packages for R that deal with that kind of stuff, or at least for microarrays, maybe a few modifications on those packages would do.
R doesn't look friendly at all. doesn't excel work for your analysis? you could check out this page:
http://phoenix.phys.clemson.edu/tutorials/excel/
hope this helps.
cheers.
hi hi,
Yes, we are going to use a package in R called Ringo, which is specifically designed for analysing histone-modification chip-chip. At the moment I don't even know how to put my data in (even though I have instructions)!!! We can't open all of our data using excel because we have 244K features, and I think R is the way to go with our data anyway. But cheers for the link - it will come in handy for other things
Clare
Exactly,
these packages (they come along with the help-documents or Vignettes) you can take for the pre-analysis of your data. All packages for normalisation, boxplotting etc..., are available for R.
You don't have to learn the complete R stuff (well, might be much more beneficial of course 
) for doing this kind of analysis since many,many,many packages are already available.
It is rather important to see what these packages do to your data.
As an example: (I assume you have .cel-files available from the Chip-analyses; which contain the raw expression values)
There is a package called 'affy' or 'simpleaffy' which you would need to install into R (instruction is given on the bioconductor website - basically copy and paste command).
Once installed, there is a command like 'ReadAffy ()' (or very similar) which you need to read in these cel.files in R.
Define the directory in which the cel.-files are located that you want to analyse and use the command 'ReadAffy ()', which will load all cel.files from that respective directory you determined before into R.
Once they are read in, they are ready to be manipulated.
Cheers
i suppose you'll have to tabulate your results, you can do that in any simple text program (not office)
If you are in Cambridge, uk you could try the people at the Sanger or the people at the biochem department (Blundell labs). If you are at the sanger or ebi, try chat or foo-mongers.