Multi-Factorial Rna-Seq Dataset

Question

[Tom De Medts]

We have a dataset from 4 genotypes, each quadruplicated, and 6 time points.

So that is a LOT of data points. 446=96!

I have never worked with R yet (have some limited Perl / Unix experience though)

How best and from where should I teach myself the rationale for choosing one method of RNA-Seq data analysis (or more), and the practical knowledge for how to implement theory into practice?

SOme of things I want to do are: 1. to know the different ways to normalize (and understand the stat behind how and why to choose one method from them all) 2. to know how to hierarchical cluster into groups that have identical temporal profiles of expression 3. to know how to compare more than just 2 genotypes at a time 4. to know how to compare across time course and not just one time point etc..

Could someone please help me out with how to go about this learning process.

Thanks, TdM

Answer 1

[Michael]

If you don't have experience with R and want to get to use Bioconductor in your project quickly, a good way might be to take a course, like this one at the EBI. Now, here comes the downside: such courses are not taking place very often are offered by a few institutions, and places are scarce. There is plenty of online material but I wasn't able to find an online classroom-like course. Courses are often announced on the Bioconductor mailing list which you could subscribe to. What I would like is a page where one could register and list all R&Bioconductor events and courses worldwide, from small courses and seminars to large training programs.

Otherwise, have a look at the 'second best' options:

You can have a look at some course materials from bioconductor courses. Look at this question and answers which has loads of links to useful material.

[Olbzn]

I recomment getting a couple books on R to get started, like statistics with R and bioconductor case studies. The course materials are great, along with the bioconductor mailing list.