Skip to main navigationSkip to main content
The University of Southampton
Biological Sciences

Research project: Transcriptome-wide prediction of eukaryotic translation initiation

Currently Active: 
Yes

This project couples bioinformatics screens with experimental data to identify possible novel sites of eukaryotic translation initiation and predict cases of transcripts with multiple alternative initiation codons, which may be an important generator of N-terminal protein diversity with important roles in protein regulation and disease.

It is clear that gene expression only provides a very limited window into the activity of the main drivers of biological processes, namely proteins, and that a lot of regulation occurs at the level of protein translation. The Kozak consensus sequence of mammalian initiation codons (GCC[A/G]CCAUGG) was established twenty-five years ago but we still know surprisingly little about the full repertoire of "rules" for translation initiation. Examples now abound of initiation from sub-optimal initiation codons, including a number of non-canonical non-AUG codons. Proposed models of "leaky scanning" and internal ribosome entry imply that some transcripts may have multiple alternative initiation codons (AIC) and there is increasing experimental evidence to indicate that this may be a widespread phenomenon. AIC usage of this nature may be an important generator of N-terminal peptide diversity, which might in turn be playing an important role in (a) disease (if the AIC usage is "accidental"), or (b) protein regulation (if AIC usage alters in response to signals).
In an attempt to get a handle on how widespread this phenomenon is - and how tightly regulated - we are performing transcriptome-wide bioinformatics analyses of annotated initiation codons and potential AIC sites in humans and mice. In addition to attempting to assess the robustness of initiation codon annotation, which itself relies somewhat on the Kozak consensus, we are identifying candidates for possible AIC, both 5' of annotated start sites and 3' of those in an apparently "weak" context. Computational predictions are being coupled to laboratory experiments to investigate the role of cis- and trans¬- factors in AIC selection and inform future screens.

Funding: University of Southampton and BBSRC

Related research groups

Molecular and Cellular Biosciences
Share this research project Share this on Facebook Share this on Twitter Share this on Weibo
Privacy Settings