Computational analysis of sugarcane ESTs for high-quality clusters and SSR mining

Abstract

Expressed sequence tags (ESTs) have provided opportunity for development of powerful marker SSR when high-quality EST clusters are available. EST clustering is commonly performed on the basis of nucleotide similarity to reduce redundancy and increase the sequence quality. The degree of similarity is one of the important parameters affecting the EST cluster quality. This work aimed to determine EST cluster quality with various degrees of nucleotide similarity and identifying SSR locus within the defined EST clusters. A collection of 2,268 ESTs from mature stalk of sugarcane ( Saccharum spp.) hybrid cultivar CP72-2036, available in dbEST of GenBank, was passed into pre-processing steps to eliminate the sequencing errors and contaminant sequences. This resulted in 2,167 clean ESTs. EST clustering with sequence identity P = 85, 90, 95 and 100% reduced the EST data set. The lowest number of clusters was obtained at P = 85%. Exploring of SSR locus also yielded the lowest number of SSR in EST clusters defined at the P value = 85%.