The human transcriptome: Evolution in action
Rotem Sorek
TAU
The estimated number of genes in the human genome is boiling down to ~25,000
or less. The transcriptome, however, is much more complex, with hundreds of
thousands of possible transcripts functioning throughout the lifetime of
each individual.
In the talk I will present two studies of the human transcriptome. In the
first, we have identified a set of unique, natural chimeric transcripts, which fuse
between pairs of consecutive genes on the human genome. We show that such
transcripts can create novel fusion proteins that share the domains of both
original proteins, or affect the regulation of the original proteins. We
provide evidence that transcription mediated gene fusion can be a mechanism
contributing to the evolution of protein complexes.
In the second study we used processed pseudogenes (PPGs) to generate a map
of the transcriptome. By analyzing thousands of human PPGs we were able to
discover hundreds of novel splice variants, and experimentally verified that
many of them are still active in the human transcriptome. Furthermore, we
demonstrate that PPGs enable the identification of ancient splice variants
that are now extinct. Our results show that the genome carries a "virtual cDNA
library" that can readily be used to analyze both the ancestral and present
transcriptomes.
These two studies were conducted in collaboration with Pini Akiva, Amir
Toporik, Ronen Shemesh and Amit Novik, from Compugen Ltd.