![]() Here we introduce AUTO-MUTE 2.0, a portable alternative to the web-based server, with platform-specific and command-line driven versions designed for Windows (PC) and Linux/Unix (Mac), as well as for Cygwin, a Unix working environment emulator for Windows (free downloads available from ). Our models were developed by implementing classification and regression statistical machine learning algorithms using the Java-based Weka software package. Large sets of diverse mutations that have been studied experimentally for their functional effects, which occur in proteins that share low sequence similarity, were used to train the AUTO-MUTE predictors. For proteins with known 3D structures, any mutation defined by a single residue replacement can be represented as a vector of features that include data derived from our in silico mutagenesis procedure. ![]() We previously developed the AUTO-MUTE server, an online set of tools for predicting protein functional consequences upon mutation, by implementing a computational mutagenesis technique that employs a four-body, knowledge-based statistical potential function derived via the coarse graining of protein structures at the amino acid level. Each approach uniquely employs evolutionary, sequence, or structural information to characterize residue substitutions in proteins, and predictions of functional effects are obtained via mathematical, rule-based, or statistical learning methods. A number of in silico mutagenesis methodologies have been developed in recent years, yielding efficient computational analogues to complement experimental methods from the wet laboratory at a fraction of the time and cost, as well as reliable and immediate predictions for functional effects of single residue replacements. Site-directed mutagenesis experiments provide researchers with opportunities to evaluate their effects on protein stability, activity, or disease potential, to annotate structural or functional roles of residues, to gain insights into mechanisms of protein folding, and to accumulate data needed for engineering new proteins with desired thermodynamic and physicochemical properties. Included among these upgrades is the ability to perform three highly requested tasks: to run “big data” batch jobs to generate predictions using modified protein data bank (PDB) structures, and unpublished personal models prepared using standard PDB file formatting and to utilize NMR structure files that contain multiple models. Nevertheless, all the codes have been rewritten and substantially altered for the new portable software, and they incorporate several new features based on user feedback. These five command-line driven tools, as well as all the supporting programs, complement those that run our AUTO-MUTE web-based server. Two additional classifiers are available, one for predicting activity changes due to residue replacements and the other for determining the disease potential of mutations associated with nonsynonymous single nucleotide polymorphisms (nsSNPs) in human proteins. Three of the predictors evaluate changes to protein stability upon mutation, each complementing a distinct experimental approach. The AUTO-MUTE 2.0 stand-alone software package includes a collection of programs for predicting functional changes to proteins upon single residue substitutions, developed by combining structure-based features with trained statistical learning models.
0 Comments
Leave a Reply. |
Details
AuthorWrite something about yourself. No need to be fancy, just an overview. ArchivesCategories |