Dr Chris Bryant
I was the Principal Investigator on the EPSRC project "Efficient Biological Grammar Acquisition" (GR/S68682, £110K). I worked on the EPSRC project "Closed Loop Machine Learning" (GR/M56067) which culminated in The Robot Scientist (see Nature 427(6971):247-252, 2004).
The development and application of machine learning algorithms. Areas of machine learning of interest include rule induction, relational data mining and inductive logic programming. The main focus of the applications are contemporary, challenging problems in molecular biology.
Specific interests include using machine learning for:
- Forming hypotheses, devising trials to discriminate between these competing hypotheses, and then using the results of these trials to converge upon an accurate hypothesis.
- Automatically generating grammars for biological sequences.
- Discovering refinements to biological networks, such as metabolic pathways.
Previous real-world applications include:
- Predicting which of the upstream Open Reading Frames in S.cerevisiae regulate gene expression.
- Discovering how genes participate in the aromatic amino acid pathway of S.cerevisiae.
- Predicting the coupling preference of GPCR proteins.
- Recognising human neuropeptide precursors.
Recommending chiral stationary phases based on the structural features of an enantiomer pair.
Qualifications and Memberships
PhD Data Mining for Chemistry: the Application of Three Machine Induction Tools to a Database of Enantioseparations (University of Manchester Institute of Science and Technology) 1996
MSc Applied Artificial Intelligence (University of Aberdeen) 1993
BSc (Hons) Combined Studies in Science: Chemistry and Computing (Sunderland University) 1990
Full Member of Data Mining & Pattern Recognition Research Centre (from 2009 to 2012)
Full Member of Informatics Research Centre (from 2013 to present)
Associate Member of Biochemistry, Drug Design & Cancer Research Centre (from October 2009 to September 2011)
Associate Member of the British Computer Society.