MAMMOTH (Matching molecular models obtained from theory): An automated method for model comparison

• DOI: 10.1110/ps.0215902
• Publication Date: 2002-10-15T19:21:58Z
• Authors: Angel R. Ortiz; Charlie E.M. Strauss; Osvaldo Olmea
• Keywords: Structural genomics; Similarity (geometry); Protein structure database; Structural alignment

Advances in structural genomics and protein structure prediction require the design of automatic, fast, objective, well benchmarked methods capable comparing assessing similarity low-resolution three-dimensional structures, via experimental or theoretical approaches. Here, a new method for sequence-independent alignment is presented that allows comparison an with arbitrary tertiary model. The heuristic algorithm given then used to show it can describe random alignments proteins different folds good accuracy by extreme value distribution. From this observation, score between two conformations same derived from likelihood obtaining chance. performance compared established, consensus manual-based scores data sets. We found approach correlates better than other tools gold standard provided human evaluator. Timings indicate fast enough routine use large databases models. Overall, our results program (MAMMOTH) will be tool comparisons applications. MAMMOTH available web site at http://physbio.mssm.edu/~ortizg/.