MolProbity is a web-based tool for validating the quality of 3D structures of proteins, nucleic acids, and their complexes. It provides detailed all-atom contact analysis, updated dihedral-angle diagnostics, and calculates H-bond and van der Waals contacts between components. It adds and optimizes all hydrogen atoms, including polar and nonpolar ones, and detects and fixes flipped sidechains of Asn, Gln, and His. It also includes new analysis functions for RNA, interfaces, and NMR ensembles. The web site and major programs have been rewritten for improved speed, convenience, and integration with other resources. MolProbity results are reported in multiple forms, including numeric scores, lists, charts, downloadable files, and 3D kinemage graphics. It is free and available at http://molprobity.biochem.duke.edu. MolProbity is unique in offering all-atom contact analysis and up-to-date, high-accuracy Ramachandran and rotamer distributions. It applies to both X-ray and NMR structures, and to both proteins and nucleic acids. It is useful to both 'consumers' and 'producers' of structural models. Consumers can check regions of interest for accuracy, while producers can find and fix errors during fitting and refinement. MolProbity focuses producers' efforts on areas that need attention, making accurate structure determination faster. Most flagged problems are worth examining, and most errors can be corrected offline. MolProbity uses physics- and knowledge-based algorithms to analyze structures. It adds hydrogens to files without them using Reduce, and can redo and optimize preexisting hydrogens. It also uses high-accuracy Ramachandran and rotamer distributions to check mainchain and sidechains for conformational outliers. It reports on geometric indicators of misfitting, such as Cβ deviation and base-phosphate perpendicular distance. MolProbity's primary input is structural models in PDB format. It can fetch models from the PDB or NDB. Additional data may be uploaded to supplement the analysis. Kinemage files can be uploaded and viewed online in KiNG. MolProbity provides several types of output, including modified PDB files, 3D kinemage graphics, tabular summaries, and analyses of molecular interface contacts. MolProbity is accessed via a web browser and offers a user-friendly interface. It is also available via command-line scripts and private servers. It has been continuously operating for over five years, with hundreds of users per month and thousands of sessions. It has seen many improvements, including the consolidation of reports into 'multi-criterion' evaluations, the ability to handle NMR ensembles and analyze interface contacts, and the expansion of RNA functionality. MolProbity provides a summary of validation criteria, listing numbers of outliers and percentile rankings. It also provides a 'MolProbity score' as a single number for overallMolProbity is a web-based tool for validating the quality of 3D structures of proteins, nucleic acids, and their complexes. It provides detailed all-atom contact analysis, updated dihedral-angle diagnostics, and calculates H-bond and van der Waals contacts between components. It adds and optimizes all hydrogen atoms, including polar and nonpolar ones, and detects and fixes flipped sidechains of Asn, Gln, and His. It also includes new analysis functions for RNA, interfaces, and NMR ensembles. The web site and major programs have been rewritten for improved speed, convenience, and integration with other resources. MolProbity results are reported in multiple forms, including numeric scores, lists, charts, downloadable files, and 3D kinemage graphics. It is free and available at http://molprobity.biochem.duke.edu. MolProbity is unique in offering all-atom contact analysis and up-to-date, high-accuracy Ramachandran and rotamer distributions. It applies to both X-ray and NMR structures, and to both proteins and nucleic acids. It is useful to both 'consumers' and 'producers' of structural models. Consumers can check regions of interest for accuracy, while producers can find and fix errors during fitting and refinement. MolProbity focuses producers' efforts on areas that need attention, making accurate structure determination faster. Most flagged problems are worth examining, and most errors can be corrected offline. MolProbity uses physics- and knowledge-based algorithms to analyze structures. It adds hydrogens to files without them using Reduce, and can redo and optimize preexisting hydrogens. It also uses high-accuracy Ramachandran and rotamer distributions to check mainchain and sidechains for conformational outliers. It reports on geometric indicators of misfitting, such as Cβ deviation and base-phosphate perpendicular distance. MolProbity's primary input is structural models in PDB format. It can fetch models from the PDB or NDB. Additional data may be uploaded to supplement the analysis. Kinemage files can be uploaded and viewed online in KiNG. MolProbity provides several types of output, including modified PDB files, 3D kinemage graphics, tabular summaries, and analyses of molecular interface contacts. MolProbity is accessed via a web browser and offers a user-friendly interface. It is also available via command-line scripts and private servers. It has been continuously operating for over five years, with hundreds of users per month and thousands of sessions. It has seen many improvements, including the consolidation of reports into 'multi-criterion' evaluations, the ability to handle NMR ensembles and analyze interface contacts, and the expansion of RNA functionality. MolProbity provides a summary of validation criteria, listing numbers of outliers and percentile rankings. It also provides a 'MolProbity score' as a single number for overall