Soft Fuzzy Model for Mining Amino Acid Associations in Peptide Sequences of Mycobacterium tuberculosis Complex

Analysis of biological data plays an important role in medical and bioinformatics industry. However, uncertainty in this biological information is the most unavoidable challenge of this era. The existing algorithms for association rule mining are inadequate to address the issues of uncertainty in the molecular data. Variation in the length of the sequences leads to variation in the degree of relationships among amino acids. Ignorance of the parameters leads to uncertainty due to the dependencies of the objects and their patterns on the parameters. The degree of relationships among various amino acids present in the molecular sequences also depends on the parameters like length ranges and species, etc. In this article, a soft fuzzy set approach has been proposed for mining fuzzy amino acid associations in peptide sequences of Mycobacterium tuberculosis complex (MTBC). The approach is employed to incorporate the degree of relationships among amino acids present in the peptide sequences. The soft sets are employed to model relationships of amino acids with the parameters like length range, species etc. The amino acid associations and their relationships with various parameters in the peptide sequences of MTBC obtained in the present study will be of great use in developing signatures that will provide better insights into the structures, functions and interactions of proteins.