identifying human liver specific genes

[cariboudoug]

Hello,

I'm interested in identifying human genes overexpressed and/or specific to the liver. I was hoping this would be straightforward with databases like Expression Atlas, but I haven't been able to get a simple list of genes using that site and others.

I then decided to download the Human Body Map 2.0 RNA-Seq mapped reads and identify differentially expressed genes by using cuffdiff (After getting FPKM with cufflinks and a corresponding gtf file, Homo_sapiens.GRCh37.75.gtf).

Unfortunately since the mapped reads are separated by tissue, but not by replicates, cuffdiff wasn't able to confidently call any genes differentially expressed between any tissues.

Does anyone have any suggestions for how I could use this data (or other data) to identify liver DE genes? Please let me know if it would be helpful to have more details.

Currently I'm thinking of simply clustering the genes and choosing groups of genes by eye, but I'm not sure what the best approach would be.

The reason I would prefer to use the HumanBodyMap data is that I am also interested in doing similar analyses of DE liver genes in mouse and dog, which have similar multi-tissue RNA-seq data available.

Thanks,
Gavin

[mbblack]

Have you looked at the TIGER database - http://bioinfo.wilmer.jhu.edu/tiger/ ?

[cariboudoug]

Thanks for the reply mbblack.

Turns out my Adblocker was blocking the expression atlas from working :s
It's working now and the baseline expressions are exactly what I needed!