Dr. Jonathan Ward

During his time in CoMPLEX, Jonathan worked on a Kernel-based classification of protein structure and function from amino acid sequences. His work describes the application of kernel methods and, in particular, the support vector machine (SVM) to the classification of protein structure and function. The data sources used by Jonathan include structure predictions and other properties that can be derived directly from amino acid sequences.

In his work, he describes a new method for the prediction of secondary structure using an SVM learning algorithm. He presented this as a guide to the application of SVMs to problems in bioinformatics. He showed that the final prediction method is comparable performance to several of the most accurate modern methods.

He also worked on the development of a method to recognize native disorder from amino acid sequences. This predictor (DISOPRED2) was shown to be the most accurate contemporary method on targets from the fifth CASP experiment. In his method, the false positive rate of DISOPRED2 is determined using ordered structures from the Protein Data Bank, and the classifier is then used to estimate the frequency of disorder in complete genomes. The final part of this project focused on the design and implementation of a publicly available web service for disorder prediction.

Page last modified on 24 aug 09 10:48