GTEx data and R scripts
GTEx consortium provided RNAseq data of human tissues. We normalized the data is performed as described previously1,2. RNA transcripts were filtered with RPKM value > 0.1 in at least 80% of the samples, followed by a log2 transformation of the RPKM values (using a pseudocount of 1), and finally a zero-mean normalization. This resulted in RNA transcript profiles from a total of 16,142 protein-coding genes, pseudogenes, and long noncoding RNAs (lncRNAs).
- Download here or alternatively here the normalized GTEx data and R scripts to create the t-SNE map and the clustering.
Note that t-SNE uses random start initializations. The final t-SNE map can therefore be slightly different then shown in the manuscript.