Revising our view of the mammalian trascriptome


Brendan J. Frey
University of Toronto and
Canadian Institute for Advanced Research

In the past 10 years, researchers have been trying to get a clear
picture of what the library of expressed protein-coding and
non-protein-coding mammalian genes looks like, and what the functions
of these gene variants are.  In this talk, I'll describe how we've
used DNA microarrays and machine learning techniques to obtain new
biological results on protein-coding and non-protein-coding genes.
Our results reveal the existence of thousands of putative new
protein-coding regions, explain discrepancies between existing gene
libraries, and appear to be in conflict with recent results obtained
by others.  Taking a closer look at our revised view of the mammalian
transcriptome, we've explored the global impact of alternative
splicing and how it introduces functional diversity, and the question
of how non-protein-coding genes, in particular microRNAs, act to
regulate genes.

References: Huang et al., RECOMB, 2006; Pan et al., Genes &
Development 20, 2006; Frey et al., Nature Genetics 37:9, 2005; Pan et
al., Molecular Cell 16, 2004.