Science for Health
In the division of Mathematical Biology, we develop methods of sequence analysis to aid in the identification of novel genes from the sequences of the human and other genomes.
Our methods work mainly with the protein sequence that is derived directly from the gene sequence. The most effective method we employ involves comparing protein sequences: if a novel protein is similar to a known protein, it may also share similarities in structure and function. If the novel gene has no similarity with a characterised protein, we use alternative approaches based on predictions of folded (3D) protein structure. We use a variety of methods for this, including sequence comparison methods that have been modified to take account of structure (threading) and statistical methods (such as Artificial Neural Nets) to infer the most likely structure.
We are continually improving and expanding these methods through both fundamental research on algorithms and continual comparison and analysis of the sequences and structures that are known.
© MRC National Institute for Medical Research
The Ridgeway, Mill Hill, London NW7 1AA