Award details

Computational approaches to identifying gene regulatory networks in Arabidopsis

Reference	BBS/E/J/0000A059
Principal Investigator / Supervisor	Professor Michael Bevan
Co-Investigators / Co-Supervisors
Institution	John Innes Centre
Department	John Innes Centre Department
Funding type	Research
Value (£)	132,328
Status	Completed
Type	Institute Project
Start date	10/02/2003
End date	09/02/2006
Duration	36 months

Abstract

This grant brought together three groups to establish a multidisciplinary approach to the problem of interpreting transcriptional regulatory sequences in genomes. As the regulation of gene expression underlies most biological processes the problem is highly relevant. The availability of complete genome sequences, careful identification of gene sequences and whole genome microarray data for gene expression in many tissues and conditions provides the resources to start to tackle this problem. In this project groups involved in microarray data generation and analysis, promoter analysis and mathematics worked together to analyse gene expression data and prediction of promoter element functions in the model plant Arabidopsis. Using a branch of mathematics called machine learning we classified gene expression responses to treatment with glucose according to promoter sequence composition. This method successfully predicted the expression response of about 78% of genes in the test set. We then developed computationally and numerically efficient machine- learning procedures as a general tool for promoter analysis. The resulting algorithm, BLogReg, provides better test set accuracy than previous models. Furthermore a new learning method was developed (OR-LS-SVM) was developed that permits classification of more complex promoter features as the classifier in a higher dimensional feature space. An example of this increased functionality is the capability of searching a feature space of sequences of any length, with mismatches, and combinations of such features. This feature is very well suited for other applications in computational biology involving pattern recognition.

Summary

unavailable

Committee	Closed Committee - Genes & Developmental Biology (GDB)
Research Topics	X – not assigned to a current Research Topic
Research Priority	X – Research Priority information not available
Research Initiative	X - not in an Initiative
Funding Scheme	X – not Funded via a specific Funding Scheme

I accept the terms and conditions of use (opens in new window)

export PDF file

back to list new search