Protein Function Prediction
Our knowledge of the protein repertoire is expanding rapidly as the international genomics initiatives continue. Since experimental characterisation of such huge amounts of data is not feasible, computational function prediction methods are required to predict the functions of proteins. CATH, the in-house protein family resource is being used to functionally classify proteins and predict the functions of whole proteins. These 'functional families' are derived using a new protocol that combines sequence clustering with supervised cluster evaluation based on sequence information. An earlier version of the protocol relying on GO annotation data for cluster evaluation has already been shown to perform well in CAFA, an international function prediction competition (Nature Methods, 2013). We have participated in CAFA in order to test the new protocol for classifying relatives into functional families. Further, a web server will be built to make these function/functional site prediction tools available to biologists and clinicians.
Research Area: Protein Function Prediction