TY - JOUR T1 - Convert Your Favorite Protein Modeling Program Into A Mutation Predictor: “MODICT” JF - bioRxiv DO - 10.1101/038992 SP - 038992 AU - Ibrahim Tanyalcin AU - Katrien Stouffs AU - Dorien Daneels AU - Carla Al Assaf AU - Willy Lissens AU - Anna Jansen AU - Alexander Gheldof Y1 - 2016/01/01 UR - http://biorxiv.org/content/early/2016/04/15/038992.abstract N2 - Motivation: Predict whether a mutation is deleterious based on the custom 3D model of a protein.Methods: We have developed MGDIGT, a mutation prediction tool which isbased on per residue rmsd (root mean square deviation) values of superimposed3D protein models. Our mathematical algorithm was tested for 42 describedmutations in multiple genes including renin, beta-tubulin, biotinidase,sphingomyelin phosphodiesterase-1, phenylalanine hydroxylase and medium chainAcyl-Coa dehydrogenase. Moreover, modict scores corresponded toexperimentally verified residual enzyme activities in mutated biotinidase,phenylalanine hydroxylase and medium chain Acyl-CoA dehydrogenase. Severalcommercially available prediction algorithms were tested and results werecompared. The modict PERL package and the manual can be downloaded from https://github.com/MODICT/MODICT.Conclusion: We show here that modict is capable tool for mutation effectprediction at the protein level, using superimposed 3D protein models instead ofsequence based algorithms used by PGLYPHEN and SIFT. ER -