Supplementary Materials

The human transcriptome across tissues and individuals

Marta Melé, Pedro G. Ferreira, Ferran Reverter, David S. DeLuca, Jean Monlong, Michael Sammeth, Taylor R. Young, Jakob M Goldmann, Dmitri D. Pervouchine, Timothy J. Sullivan, Rory Johnson, Ayellet V. Segrè, Sarah Djebali, Anastasia Niarchou, The GTEx Consortium, Fred A. Wright, Tuuli Lappalainen, Miquel Calvo, Gad Getz, Emmanouil T. Dermitzakis, Kristin G. Ardlie, Roderic Guigó

Materials/Methods, Supplementary Text, Tables, Figures, and/or References

Download Supplement
  • Materials and Methods
  • Figs. S1 to S25
  • Tables S1 to S20
  • Captions for data tables S4 to S7 and S9 to S18
  • References
Data tables S4 to S7 and S9 to S18
Table S4 Top Expressed genes. The hundred most expressed genes in each tissue and their cumulative contribution to the global amount of gene expression in that tissue. Table is an excel file.
Table S5. Tissue preferentially expressed genes. We used NOISeq to call a gene tissue preferentially expressed by comparing the samples from a given tissue to those samples from the rest of the tissues. The table shows all instances where the mean expression of the gene in the tested tissue was significantly higher (FDR<0.01 and a log2 fold change >= 4) than in the samples from the rest of the tissues.
Table S6. Tissue exclusive expressed genes. Genes with high tissue specificity (phi>=0.95) or tissue anti-specificity (phi < -0.95). For each pair (gene, tissue) the phi statistic is computed from a 2x2 contingency table that includes the number of samples from the tissue in which the gene is expressed (RPKM>0.1) and not expressed (RPKM<0.1), and the number of samples from the rest of the tissues in which the gene is expressed and not expressed. Values of phi >=0.95 indicate that the gene is expressed in (nearly) all samples from the tissue (it is exclusive of the tissue), and in (nearly) no samples from the rest of the tissues. Genes with phi < -0.95 indicate that it is (nearly) not expressed in samples from the tissue and it is expressed in (nearly) all samples form the rest of the tissues.
Table S7. Number of tissue exclusive expressed genes with different expression thresholds. Number of genes with high tissue specificity (phi>=0.95) or tissue anti-specificity (phi < -0.95) using different thresholds to consider a gene expressed (threshold expressed) or not expressed (threshold not expressed). For each pair (gene, tissue) the phi statistic is computed from a 2x2 contingency table that includes the number of samples from the tissue in which the gene is expressed (RPKM> threshold expressed) and not expressed (RPKM< threshold not expressed), and the number of samples from the rest of the tissues in which the gene is expressed and not expressed. For all thresholds, most tissue exclusive genes are from Testis.
Table S9. Contribution of individual and tissue to variation in gene expression of protein coding and lncRNAs. Relative contribution of individual, tissue and residual to the variance in gene expression.
Table S10. Genes differentially expressed between males and females (FDR <0.05)
Table S11. Genes differentially expressed between African American and individuals of European ancestry (FDR<0.05).
Table S12. Genes differentially expressed across age. Genes that changed (FDR<0.05) expression with age across all GTEx tissues based on LMM analysis.
Table S13. Genes differentially expressed between males and females within each tissue (FDR<0.05).
Table S14. Genes differentially expressed between African American and individuals of European ancestry within each tissue (FDR<0.05).
Table S15. Tissue preferential Exon Inclusion. Number of exons differentially included between a given tissue and the remaining tissues. An exon is considered to be differentially included if FDR < 0.01 and ΔPSI>0.1
Table S16. Tissue exclusive Exon Inclusion. Number of exons that have high tissue exclusivity (phi>=0.95) or tissue anti-exclusivity (phi < -0.95)
Table S17. Individual and Tissue contribution to variation of splicing in protein coding genes.
Table S18. GO enrichment analysis for genes with high contribution of individual to splicing variation. Analysis was carried out with the 139 genes that could be mapped in the DAVID database. Twenty-nine clusters were detected. The figure shows the first 2 most enriched clusters. The first cluster has 9-fold enrichment for functions related to translation and the ribosomes.