A DNA-based clustering method based on statistics adapted to heterogeneous coordinate data

A cluster analysis is often used in social sciences, management, general science and engineering, etc. with the objective of characterising structures in heterogeneous data sets. In this case, collections of information granules are obviously constructed through clustering techniques. However, clustering problems are intractable and NP-complete problems with a number of patterns. In this article, we discuss the use of DNA computing as a vehicle of heterogeneous coordinated data clustering, and elaborate on the fundamentals of DNA computing in the context of clustering tasks. A novel DNA-based clustering method is proposed, using statistics-based encoding of DNA strands, for clustering coordinated data from simulated DNA studies and experiments. The results also show the capabilities of this method when adapted to heterogeneous coordinate data.