Nucleic Acids Research Advance Access published online on June 15, 2009
Nucleic Acids Research, doi:10.1093/nar/gkp444
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Computational Biology |
Structural implication of splicing stochastics
1Center for Advanced Research in Biotechnology, University of Maryland Biotechnology Institute, 9600 Gudelsky Drive, Rockville, MD 20850 and 2Molecular and Cell Biology Program, University of Maryland, College Park, MD 20742, USA
*To whom correspondence should be addressed. Tel: +1 240 314 6240; Fax: +1 240 314 6253; Email: melamud{at}umbi.umd.edu
Received May 8, 2009. Revised May 8, 2009. Accepted May 11, 2009.
Even though nearly every human gene has at least one alternative splice form, very little is so far known about the structure and function of resulting protein products. It is becoming increasingly clear that a significant fraction of all isoforms are products of noisy selection of splice sites and thus contribute little to actual functional diversity, and may potentially be deleterious. In this study, we examine the impact of alternative splicing on protein sequence and structure in three datasets: alternative splicing events conserved across multiple species, alternative splicing events in genes that are strongly linked to disease and all observed alternative splicing events. We find that the vast majority of all alternative isoforms result in unstable protein conformations. In contrast to that, the small subset of isoforms conserved across species tends to maintain protein structural integrity to a greater extent. Alternative splicing in disease-associated genes produces unstable structures just as frequently as all other genes, indicating that selection to reduce the effects of alternative splicing on this set is not especially pronounced. Overall, the properties of alternative spliced proteins are consistent with the outcome of noisy selection of splice sites by splicing machinery.