Functional genomics research

The Brazma research group complements the Functional Genomics service team, and focuses on developing new methods and algorithms and integrating new types of data across multiple platforms. We are particularly interested in cancer genomics and relationships between transcriptomics and proteomics. We collaborate closely with several groups at EMBL-EBI, including the Marioni, Stegle and Saez-Rodriguez groups.

Our group led the analysis of transcript isoform use and fusion gene discovery from lymphoblastoid cell lines of 465 individuals who participated in the 1000 Genomes Project, as a part of the GEUVADIS project. The human transcriptome contains in excess of 100,000 different transcripts. We analysed transcript composition in 16 human tissues and five cell lines to show that, in a given condition, most protein coding genes have one major transcript expressed at significantly higher level than others, and that in human tissues the major transcripts contribute almost 85% to the total mRNA.

With our collaborators from Canada, France, UK, Latvia and other countries we co-led a European renal cancer project CAGEKID, a part of the International Cancer Genome Consortium (ICGC). In addition to supporting previous reports on frequent aberrations in the epigenetic machinery and PI3K/mTOR signaling, we uncovered novel pathways and genes affected by recurrent mutations and abnormal transcriptome patterns including focal adhesion, components of extracellular matrix (ECM) and genes encoding FAT cadherins.

Future plans

Large-scale data integration and systems biology will remain in the focus or our research. We will be extending our work on cancer genomics as a part of the pan-cancer project of the International Cancer Genome Consortium, in which we are co-leading the transcriptomics/genomics integration working group that aims to study aberrant transcription patterns across many cancer types. We will extend our research on dominant transcripts to newer, much larger datasets to study how dominant transcripts switch over between different tissues and what implication this has on proteome.

Brazma Research, EMBL-EBILoss of function through alternative splicing in renal cancer, using the example of the protein phosphatase 2A activator (PPP2R4). Ensembl data shows a switch between two PC transcripts. The APPRIS database, which houses protein structure, function and conservation information for splice variants (developed at CNIO and INB in Spain), shows the principal transcript in N, but not in T. Using the EMBOSS, Needle and UniPDB tools, we see there is less than 35% protein overlap.

Experimental / computational fellowships

ESPOD postdoctoral programme

Interdisciplinary fellowships

EIPOD: Interdisciplinary postdoctoral fellowships at EMBL

Selected publications

Gonzàlez-Porta MFrankish ARung J, et al. (2013) Transcriptome analysis of human tissues and cell lines reveals one dominant transcript per geneGenome Biology 14:R70.

Lappalainen TSammeth MFriedländer MR, et al. (2013) Transcriptome and genome sequencing uncovers functional variation in humans. Nature 501:506-511.

Scelo G, Riazalhosseini Y, Greger L, et al. (2014) Variation in genomic landscape of clear cell renal cell carcinoma across EuropeNature Communications 5:5135.