ModelMaker

ModelMaker

Version 0.1, a VMD plugin for creating complete models with Rosetta. ModelMaker is an all-in-one modeling solution combining the strength of established but often complicated to use modeling suits like Rosetta and Modeller with MDFF in an easy-to-use manner in VMD. Integrative modeling approaches usually aim to automate the process of structure analysis to avoid human bias, yet the experience of structural biologists may be a desirable factor in structure refinement. Therefore, ModelMaker uniquely allows for incorporation of user expertise by taking advantage of interactive MDFF, allowing for manual manipulation of the structure.

Installation
To use ModelMaker, you will need to download and install (in addition to VMD) Rosetta. Installation details for ModelMaker itself can be found here in the wiki.
Once ModelMaker is installed, in VMD's TkConsole, type `modelmaker` to see a list of commands. You can also type `modelmaker command` where 'command' is one of the available ModelMaker commands to see more information about it.

ModelMaker Tutorial pdf (required tutorial files [.tar.gz, 76M]) A tutorial demonstrating a variety of modelmaker commands.

Source Code and more information can be found on GitHub.

ModelMaker Workflow
ModelMaker interactively builds complete structural models guided by incomplete structural data from experiments, automated structure prediction, and user expertise. With ModelMaker, incomplete models are completed by generating ensembles of models of the missing segments with de novo structure prediction in Rosetta. Then, a single complete model is obtained by ensemble filtering through sorting, clustering, or secondary structure analysis in VMD. This complete model is refined to a mid-resolution density (4.5 to 8 Angstrom) by interactive molecular dynamical flexible fitting (MDFF) resulting in a structure with a reliable fit of the secondary structure elements. To further refine the backbone and side chains to high-resolution densities (3.5 to 4.5 Angstrom) monte carlo backbone and sidechain rotamere search algorithms from Rosetta are combined with MDFF in an iterative manner.

PDB entries generated with ModelMaker: 5L4G, 5L4K (in silico human proteasome); 5MP9, 5MPD, 5MPE, 5MPB, 5MPC, (in silico yeast proteasome); 6EPF, 6EPC, 6EPD, 6EPE (in situ rat proteasome).

Publications

  • Structural Studies of Neuronal C9ORF72 Poly-GA Aggregates Reveal Proteasome Accumulation and Impairment. Guo Q., Lehmer C.*, Martinez-anchez A.*, Rudack T*, Beck F., Hartmann H.,erez Berlanga M., Frottin F., Hipp M ., Hart F. U., Edbauer D., Baumeister W., Fernandez-Busnadiego R.Cell 2018, in press.

  • Structural insights into the functional cycle of the ATPase module of the 26S proteasome. Wehmer M.*, Rudack T.*, Beck F.*, Aufderheide A, Pfeifer G., Plitzko J.,Foerster F., Schulten K., Baumeister W., Sakata E.

    • PNAS 2017 114(4):1305-1310.

  • The structure of the 26S proteasome at a resolution of 3.9 A. Schweitzer A.*, Aufderheide A.*, Rudack T.*, Beck F., Pfeifer G., Plitzko J., Sakata E., Schulten K., Foerster F., Baumeister W.

    • PNAS 2016 113(28):7816-7821.