Using Data Mining and Computational Approaches to Study Intermediate Filament Structure and Function

Experimental and theoretical research aimed at determining the structure and function of the family of intermediate filament proteins has made significant advances over the past 20 years. Much of this has either contributed to or relied on the amino acid sequence databases that are now available online, and the data mining approaches that have been developed to analyze these sequences. As the quality of sequence data is generally high, it follows that it is the design of the computational and graphical methodologies that are of especial importance to researchers who aspire to gain a greater understanding of those sequence features that specify both function and structural hierarchy. However, these techniques are necessarily subject to limitations and it is important that these be recognized. In addition, no single method is likely to be successful in solving a particular problem, and a coordinated approach using a suite of methods is generally required. A final step in the process involves the interpretation of the results obtained and the construction of a working model or hypothesis that suggests further experimentation. While such methods allow meaningful progress to be made it is still important that the data are interpreted correctly and conservatively. New data mining methods are continually being developed, and it can be expected that even greater understanding of the relationship between structure and function will be gleaned from sequence data in the coming years.