Primary sequence analysis and representation techniques in carbohydrates
Centre for Parallel Computing 115 New Cavendish Street, London W1M 8JS, UK
1Division of Biotechnology 115 New Cavendish Street London W1M 8JS, UK
2Artificial Intelligence Division, University of Westminster 115 New Cavendish Street London W1M 8JS, UK
Sequence similarity calculations of carbohydrates present several problems which must be addressed if a computer implementation is to be achieved. These problems range from the computational representation of the complex carbohydrate structrure to the method by which the comparison of residue and linkage is to be made. This paper therefore discusses the form of this representation and how two or more carbohydrates can be meaningfidly compared. An example set of results using this approach is presented and discussed to illustrate how similarity comparison can show relationships between carbohydrates, features that are otherwise hidden by the sheer volume of data which must be considered.
Received on March 25, 1993; accepted on June 18, 1993