Cufflinks for Arabidopsis.. Ensembl or TAIR 10

Question

[mfahim]

Hi there,

I am about to run cufflinks on my TOPHAT bam files for Arabidopsis. I guess I will have to use mapped.bam files for my samples.

What reference genome should I use for cufflinks?

ENSEMBL19

or

TAIR10

I want to make sure that I get gene names instead of XLOC id in cummeRbund..

Your advice is very precious for me. Thank you.

Answer 1

[Antonio R. Franco]

There is a paper around "A comprehensive evaluation of ensembl, RefSeq and UCSC annotations in the context of RNA-seq read mapping and gene quantification. Zhao and Zhang BMC Genomics (2015) 16:97 that evaluates the effect of different annotations in your data

This affects to only a small set of genes, those that are annotated in a different way in different assemblies

If this is something concerning you, use both reference genomes, and compare

[mfahim]

Thanks Antonio..

That's a great help.. one last question.. I am interested in Alternative Splicing and novel isoforms (if any).. Any further suggestion?..

Cheers

Fahim

[Antonio R. Franco]

If using RNA-Seq, I would suggest you to use the latest version of limma under R

The vignette this program provides is very clear and intuitive

[mfahim]

Thanks Antonio..

I will try Limma.. I hope there is GUI.. I have little understanding of commands and programming..

Have a great day.

Fahim

[Antonio R. Franco]

I don't think there is a GUI for limma. But you can follow the vignette (help)

There is a nice GUI for working with R and several programs for RNA-Seq analysis, but this does not include limma.

You can find it here, RNASeqGUI.