You should note that a number of the tools in R and BioConductor require raw count data, not normalized count data. If the sequencing center did not include raw count data, you will need to get that.

If the data is already normalized count data, you can compute a simple ANOVA for statistical differences between samples. However, one of the single greatest sources of variation in RNA-seq statistical results derives from differences in the normalization approach used, so you really should explore potential normalization schemes yourself to decide which one you prefer (although again, to do that you will need raw count data).

If you do decide to use R/BioConductor tools, I would strongly encourage you to explore several of them., as you may find results vary a great deal depending on choice of normalization and statistical significance test used. There is no "one-size fits all" for RNA-seq data analysis at this point in time, so you need to explore your particular data a bit to get a sense of its specific variance and behavior, across your samples and across your obtained range of mapped reads.

You can use R bioconductor to statistically analyse RNA-seq data.

http://www.bioconductor.org/packages/release/BiocViews.html#___RNAseq

Based on your requirements , you can use different packages available.

You can find the paper published in nature which cites the above:

http://www.nature.com/nprot/journal/v8/n9/full/nprot.2013.099.html

I second Poornima: R is a good choice.

Agreed..R is a good choice indeed

You should note that a number of the tools in R and BioConductor require raw count data, not normalized count data. If the sequencing center did not include raw count data, you will need to get that.

If the data is already normalized count data, you can compute a simple ANOVA for statistical differences between samples. However, one of the single greatest sources of variation in RNA-seq statistical results derives from differences in the normalization approach used, so you really should explore potential normalization schemes yourself to decide which one you prefer (although again, to do that you will need raw count data).

If you do decide to use R/BioConductor tools, I would strongly encourage you to explore several of them., as you may find results vary a great deal depending on choice of normalization and statistical significance test used. There is no "one-size fits all" for RNA-seq data analysis at this point in time, so you need to explore your particular data a bit to get a sense of its specific variance and behavior, across your samples and across your obtained range of mapped reads.

R and Bioconductor. Though number of freely available tools are also available.

I would recommend Tophat (maping) -> Cufflinks (Assembly) -> EdgeR (differetial expression) for analysis of RNAseq data. The choice of softwares is based upon number of citations and the flexibility provided by the multitude of parameters in each program, which allows customizing the analysis according to the sample, treatment or organism.

However these programs require you to be well versed with the unix and R environment. If you are new to the field you may as well give a look to this publication, which provides a user interface to to well known RNAseq analysis softwares

http://www.ncbi.nlm.nih.gov/pubmed/22684630

If the excel contains genes / transcripts and tag counts / counts per millions (but *not* RPKM) use Bioconductor edgeR or Deseq libraries.

I prefer Trinity (for denovo assembly)-RSEM (uses Bowtie) for mapping and R-Bioconductor DEseq for differential analysis.

I will appreciate if you can help me with RobiNA, .

I have some problems to use RobiNA as described below.

I reached Read mapping but I could not proceed to next step.

Because it does not recognise my sequence file.

Further, I need to mention about the other problem. for just training I have downloaded "hg19" files from "UCSC bioinformatics" website, which has so many files for the sequences. I would like to combine them but I could not perform it. would you able to help me in solving this problem?

Thanks in advance.