Project Large Scale Protein Family Classification

Project Large-scale protein family classification (Thomas).

Project Large-scale protein family classification (Thomas)

Project Name: Large-scale protein family classification

Project Start Date: 2009/03/01

Collaborators: Cedric, Peter Meinicke (Gobics, [email protected])

Description:

Rapid protein family classification is a viable tool for automated protein function prediction in the post-genomic era. In this project I concentrate on the development of alignment-free methods for fast protein domain family assignment.

These methods include different approaches to represent protein sequences in a uniform vector space and machine learning techniques that are suitable for large-scale classification.

The latest development also includes an evaluation of a discriminative PSI-BLAST approach based on these techniques.

Lab-Book:

Links:

Bibliography: