Gene3D: Design and development, HMM library construction and prediction verification, and web services.

PhD at the Sanger Institute (2004), supervised by Alex Bateman (Pfam).

Thesis: Biological Investigations Through Sequence Analysis.

Currently a PostDoc at UCL.

All things domain, and plenty of evolution; function prediction for when I'm feeling pragmatic. I prefer to tie function to specific physical features, and am interested in understanding how we can predict emergent functions from the combinations of active elements within a protein. My feeling is that function prediction through statistical identification of correlated terms or similarity matches and path reconstruction will never be truly predictive, but can only infer functional associations based on prior observation. As a result they will always generate substantial numbers of errors.

For a full list of my publications follow this link:

http://www.ncbi.nlm.nih.gov/pubmed?term=%28Yeats+C[au]%29+OR+%28Yeats+CA[au]%29

long:12606174

long:15994557

long:18052542

long:12217513

long:18032434

long:17372197

long:14681378

Gene3D

Ab Initio Protein Structure Prediction Solved

~~DIR?cols=page;description&hdrs=Person; Description~~