Bioconductor R Tutorials / Help for Beginners

[ETHANol]

Next Gen sequencing (ChIP-seq) is sucking me further and further into my computer. The problem is I am a biologist and relatively (i.e. extremely) new to bioinformatics.

I am very interested in getting the packages available from Bioconductor up and running but I am throughly confused by the instructions provided from Bioconductor. It is clear I need some remedial lessons to understand how to operate these programs.

Can someone point me in the direction of some information and tutorials on how to use and understand the ChIP-seq software available from Bioconductor and enough fluency in R to use them?

Thanks!!!

[malachig]

R is a language/analysis platform known for having a somewhat steep learning curve. The documentation available for each function can be thorough and is consistently available. However, this documentation tends to be written by statisticians, for statisticians. Furthermore, the user forums are not nearly as welcoming as this one is. They are full of people who have read the documentation, not understood it, asked for help, and essentially been told to read the documentation. Having said all that I would still encourage you to try R as it can be powerful and rewarding and many tools are being actively developed.

The r-project has a list of suggested books. There are some good ones to get you going there. Many genome centers will have some local R experts, and these people sometimes offer tutorials.

There are also many posts in this forum relating to various R tools. Hopefully someone with specific experience analyzing ChIP-seq data in R will respond. At a glance the following links seem relevant:
R_ChIP-Seq wiki, Advanced RNA-Seq and ChiP-Seq data analysis, HT Sequence Analysis with R and Bioconductor, Advanced RNA-Seq and ChiP-Seq data analysis, ...

[ETHANol]

Thanks!!! I'll take a look at the links you gave.

I'm somewhere between wanting to pull my hair out and wanting to throw my computer out the window.

But there are so many nice looking tools, it would be a pity to not take advantage of them.

[shurjo]

I'm somewhere between wanting to pull my hair out and wanting to throw my computer out the window.

Two wigs and three computers later, I understand your pain. But learning a little R has been very useful for nextgen analysis, so I would say hang in there.

[malachig]

Yes its probably worth learning in the end. If you learn some useful tasks for nextgen sequence analysis you will also likely learn to use it for many other analysis and graphing tasks. Once you get used to creating graphics in R you can script the creation of arbitrarily complex figures. Some impressive examples have been collected in the R Graph Gallery.

Another useful feature of R is add-on packages like Cairo (library and R package) that allow creation of jpegs, SVGs, etc. from within a script without using X11 graphic drivers. This is particularly useful if you want to create graphs from the nodes of a linux computer cluster.

Another interesting concept that utilizes R and 'Sweave' is the idea of creating 'reproducible research' for publications.

[kalidaemon]

I feel your pain with R... I've been using MACS, which is a software package from Shirley Liu's group: http://liulab.dfci.harvard.edu/MACS to analyze my ChIP-Seq data and have been really happy with the results. MACS is unix-based so you can avoid R for the time being if you'd like.

[Wolfgang Huber]

The material from recent courses is online, see

http://www.bioconductor.org/help/course-materials/2010

e.g. "Computational Statistics for Genome Biology 2010" and "Advanced
RNA-Seq and ChiP-Seq data analysis". It might be worth joining one of
the upcoming ones in your area.

Also, have a look at the Bioconductor mailing list, which is
searchable for past discussions
http://www.bioconductor.org/help/mailing-list and of course you can
join it.

For a general introduction into the R language, I recommend the free http://cran.r-project.org/doc/manuals/R-intro.html

Best wishes
Wolfgang

[ETHANol]

Thanks everyone. It looks like leaning how to use R packages is more of a long-term goal then an immediate answer.

Yes, MACS is nice and super easy, as is USeq and some others. Great for us wet lab scientists.

With that said, I still think it will be worth the time to learn at the very least some basic knowledge of R/Bioconductor as I think it will make me a better molecular biologist.

Looks like I should have some 'interesting' reading this winter.

It would be great if one of you computer guys sat down with a lab guy and wrote a book on this subject.

[hmv]

I have just come across this thread, and thought that some people might be interested in the course that we will be running in March. The theme is maths for modelling, but it introduces R and has a good section on descriptive statistics and graphics with R. Our website is here : http://octette.cs.man.ac.uk/bioinformatics/index.html

[krobison]

Surprised the books list mentioned doesn't have Bioconductor Case Studies; very oriented around microarray expression analysis, but very useful.

[enelkinsan]

bioconductor, some useDevel compatibility?

Hi,
I'm relatively new to the R topic and currently trying ot install Bioconductor (actually, a tutorial material on it). I've got a strange error message and wonder if anybody could help me with that or point me into the direction of some R-related forum suitable for beginners.

The problem is that it gives an error on useDevel() even on the latest version of R 2.15.2 (32-bit, Win)
> source("http://bioconductor.org/biocLite.R")
> useDevel()
Error: 'devel' version requires a more recent R

Has anyone ever have something like this?

Thanks,
Eliz

[Jeremy]

Hi,
I'm relatively new to the R topic and currently trying ot install Bioconductor (actually, a tutorial material on it). I've got a strange error message and wonder if anybody could help me with that or point me into the direction of some R-related forum suitable for beginners.

The problem is that it gives an error on useDevel() even on the latest version of R 2.15.2 (32-bit, Win)
> source("http://bioconductor.org/biocLite.R")
> useDevel()
Error: 'devel' version requires a more recent R

Has anyone ever have something like this?

Thanks,
Eliz

If you just want to install Bioconductor then

source("http://bioconductor.org/biocLite.R")
biocLite() #install basic packages
biocLite("package-name") #replace package name with whatever individual package you want

useDevel() installs the development version I think (not sure what exactly that is though) and requires R2.16.