BACKGROUND: A new method for the prediction of protein structural classes is constructed based on Rough Sets algorithm, which is a rule-based data mining method. Amino acid compositions and 8 physicochemical properties data are used as conditional attributes for the construction of decision system. After reducing the decision system, decision rules are generated, which can be used to classify new objects. RESULTS: In this study, self-consistency and jackknife tests on the datasets constructed by G.P. Zhou (Journal of Protein Chemistry, 1998, 17: 729-738) are used to verify the performance of this method, and are compared with some of prior works. The results showed that the rough sets approach is very promising and may play a complementary role to the existing powerful approaches, such as the component-coupled, neural network, SVM, and LogitBoost approaches. CONCLUSION: The results with high success rates indicate that the rough sets approach as proposed in this paper might hold a high potential to become a useful tool in bioinformatics.

MOTIVATION: Simple sequence repeats or microsatellites have been found abundantly in many genomes. However, the significance of distribution preference has not been completely understood. Completion of the Arabidopsis genome sequencing allows us to better understand and characterize microsatellites. RESULTS: Microsatellite distribution was more abundant in 5'-flanking regions of genes compared with that expected in the whole genome, with an over-representation of AG and AAG repeats; there were clear differences from distributions in 3'-flanks and coding fractions, where triplet frequencies evidently corresponded to codon usage. We identified 1140 full-length genes that contained at least one locus of AG or AAG repeats in their upstream sequences, and whose functional characteristics were significantly associated with the repeats. This observation indicates that selective pressure markedly differed in the three transcribed regions, with positive selection of AG and AAG repeats in 5'-flanks close to those genes whose products are preferentially involved in transcription.