Best way to normalize Human BodyMap 2.0 data

Question

[Dataman]

Hi,

I have a list of interesting genes and I want to compare their expression levels within different tissues. To do so, I am looking at gene expression levels of the genes in question in the BodyMap 2.0 dataset. However, I am not sure whether this dataset is normalized and my search in the web to see whether this data is normalized was unfruitful! So, I am wondering if we need to normalize the BodyMap data (just to be able to compare the expression levels in different tissues), which normalization method is the best? Currently, I am thinking of the 'median of ratios' normalization which was used in DESeq. However, I would like to hear your ideas as well.

Thanks in advance for your thoughts!