Analysis of genome signature strength of SARS coronavirus using Self-Organizing Map neural network

The nucleotide usage patterns vary not only from organism to organism, but also between genes in the same genome. Each genome has its own characteristics. This unique identity, called genome signature, of a genome is multidimensional. One of the ways to probe into this area is to analyze the nucleotide sequence composition of the genome. In this paper, the nucleotide compositional structure of SARS Corona virus which is the cause of the Severe Acute Respiratory Syndrome (SARS) is analyzed. Both the mono, di and tri nucleotides compositions are explored to find out the genomic nucleotide pattern. The Kohonen´s self-organizing map neural network model is used as a tool to analyze the strengths of different nucleotide signatures of the genome. The analysis reveals that SARS virus is Thymine dominated, AT-rich and has the dinucleotide signature as qualitatively best signature, although codon and RSCU based SOM results in clearer cluster maps.