This is a list of every application in Rosetta that is compiled with weekly releases as of June 15, 2015.


Runs the performance benchmarks, which determine whether recent changes to the Rosetta codebase have substantially altered its performance. Users do not, generally speaking, benefit from running this application unless they are attempting to make major modifications to Rosetta.


Runs a scientific test that computes the delta-delta-G of mutation to alanine for a number of residues with known experimental values. Similar to the performance benchmark, it is very unlikely that any user modifications to the codebase of a protocol could propagate backward to affect this benchmark.


Runs fixbb design on several PDBs to test the overall amino acid type distribution returned by that algorithm.


Repacks a PDB and reports how many rotamers were preserved by the packer: this statistic of scoring function quality is intended to be run on very high quality PDBs because recapitulation of native good rotamers would be good.


Performs ab initio structure determination using fragments files and secondary structure predictions.


Performs the backrub mover, which involves a particular type of backbone sampling intended to maintain consistent sidechain interactions and orientations


Selects the first 400 models for creating clusters based on CA distance according to cluster radius. Assign remaining structures one at a time by computing distance to prior created clusters; outside the cluster radius, a new cluster is created.


Combines multiple silent files into one.


Extracts the models with provided tags from a silent file into a PDB file.


Given a protein sequence and optionally secondary structure or other information, produces fragments file of the desired length(s) by drawing on the "vall," the Rosetta fragment library.


Replace each residue in a pose with the idealized coordinates drawn from the database.


Locally minimize a structure using a given scoring function and minimization type.


Minimize a ligand at an interface and dump a lot of analysis information.


Computes advanced cavity detection statistics on PDB structures to analyze packing quality.


Perform cycles of all-atom minimization and sidechain repacking with a ramping repulsive term. Responds to a provided relaxation script.


Massively versatile application for expanding and contracting structures; it essentially brings the sampling framework of loop modeling to scenarios that require design.


This application uses a genetic algorithm and repeated backrub sampling with a resfile to optimize the sequence of a protein and explore the range of residue identities that that structure permits.


Dock a set of symmetric oligomers.


Locate buried voids using an early version of RosettaHoles and attempt point mutants that might reduce void volume.


Determine what residues contact each other.


Obtain weighted RMSD of a set of decoys, align them, and write them out.


Obtain weighted RMSD of a set of decoys and then cluster them, using a more traditional clustering algorithm than cluster.


Performs extensive analysis on the residues present at the interface, generating a wide variety of metrics.


Form a peptide macrocycle from a conformation of a peptide in a binding pocket.

per_residue_energies, residue_energy_breakdown

Prints out the per residue energy breakdown for every residue in a pose.

score, score_jd2

Scores a given PDB with the requested scoring function. Can also be used to extract models with a given tag from a silent file into a PDB. score_jd2 takes advantage of the modern job distributor.


Legacy code to perform antibody structure determination.


Graft new loop onto an existing antibody framework.


Models the CDR H3 loop of an antibody in conjunction with optimizing surrounding loops and VH/VL orientation.


A catchall app that, in versions of Rosetta before three, did just about everything. It is distributed with BOINC to perform Rosetta@Home tasks. You probably don't need to run this application.


Minirosetta, but with a graphics viewer.


Cluster components of multiple sequence alignments based on sequence similarity.


Set up numbering of alignments so that they match provided PDBs.


Obtain full length model of a protein from an alignment.


Thread each input pose onto the template pose.

score_aln, score_aln2

Score alignments using a provided PSSM; score_aln2 uses machinery specialized for comparative modeling and is preferred.


Superimpose two given poses either by provided target residues or by all residues.


Manually remove any sequence mismatches or failed simulations from the silent structure provided.


Perform coupled moves, in which the Boltzmann rotamer mover (which, in contrast to the packer, makes a concerted effort to achieve sequence diversity) repacks and small backbone moves are made at the same time.


Employs ray casting algorithms (optimized for execution on GPUs) to achieve ligand docking.


Creates input files for using ray casting for alignment. This application is separate in large part because DARC benefits from GPU acceleration and this would not.


Finds the energy of mutation for point mutations (or multiple mutations) of a given PDB. Samples ensembles of wild type and mutant and gives the energies of each as either the average of the lowest three energies, the lowest energy, or the average energy.


Creates an ensemble of structures using minimization with all-CA-pair constraints, optionally small and shear moves, and rotamer trials. Notionally, scoring and computing statistics on wild type and mutant ensembles generated by this function would be another mode of ddG prediction; importantly for interface ddG, it would not explicitly be modeling the bound and unbound states.


Minimize with constraints on all pairs of CA atoms. I


Repacks a pose using a resfile. Naturally, fixbb's iconic use is for fixed-backbone design; this is identical internally to repacking a pose using a resfile and merely allowing rotamers of other amino acids at one or more positions.


Perform multistate design using a genetic algorithm with a structure ensuring that corresponding residues in different states are packed to the same rotamer, using MPI.


Scan for stabilizing mutations in a protein. This application can run in a double mutant mode that requires the mutated residues in question to have at least one atom in contact with each other. It can also take only a given list of mutations, output structures, accept a ddg cutoff, and instead try to destabilize an interface.


Performs fixed backbone design and analyzes the resulting produced sequences for a statistic called sequence recovery: how often was the native sequence recapitulated for different residue types (polar, aliphatic, ...) and environments (core, surface, ...).


Repacks, redesigns, or minimizes a complex comprising one or more beta-peptide chains. (Importantly, a beta-peptide here does not refer to a conformation, but rather being composed of beta amino acids.)


Perform design on RNA poses. This application may be preferred for RNA design because it can process critical options such as whether to sample a particular proton chi on O2' and dump useful information, such as whether residues are in particular environments (single stranded, double stranded, or tertiary contact).


Intended to mutate exposed residues to predominantly have positive or negative charge due to useful resulting biological properties. Can set a target charge or not; can be run in a deterministic AvNAPSA mode or a stochastic pack_rotamers mode. By default, only mutates to lysine for positive mode or glutamate (except starting from asparagine) in negative mode; can accept custom reference energies in Rosetta mode.


Designs three residues of a zinc binding site on ankyrin, then use a residue from the other protein to define the interface orientation, sample rigid body space, then design.


Set up a homodimeric interface mediated by coordinating two zinc atoms: grafts in two two-residue zinc binding matches onto the protein surface, rotates around the zinc-zinc axis and the perpendicular axis. Grid searches zinc-zinc axis rotations and dumps uneclipsed poses with good metal geometry.


Symmetric design on a zinc-mediated homodimer, with minimization of sidechains, backbones, and the jump.


Prepacks the interface between a pair of chains to prepare for a docking simulation.


The main workhorse of classical rigid body docking, which can perform both the low resolution global centroid protocol and the high resolution refinement protocol.


Given a structure and electron density data to which it has a poor fit, generate loops files that would make Rosetta remodel up to the given fraction of the starting structure.


Uses fragments to perform a molecular replacement protocol into electron density data.


Given an enzyme constraint file, produces an inverse rotamer tree of the residues in question (which form a geometrically ideal "theoretical enzyme") and dump it as a multimodel PDB. As "stealth functionality," can take a PDB using -s that will be aligned to the theozyme created.


Performs fixed or flexible backbone enzyme design with ligand rotamer sampling.


Clustering algorithm specialized for working with RNA


Turns provided RNA containing PDB files into the RNA equivalent of a "vall" fragments file, which instead just stores torsions. Similarly, stores rigid body orientations


De novo design of RNA structures; this differs from ab initio in the protein context by using "fragments files" that are actually just torsions and relative rigid body orientations.


Extracts RNA structures from silent files


Minimizes RNA structures using the RNAMinimizer (preferred); can vary bond geometry and skip trials on the O2' chi.


Scores RNA structures.


Solves RNA structures using RNA constraints.


Employs fiber diffraction data to refine protein structures.


Builds a peptide of a given sequence in extended conformation.


Dock a peptide in a binding pocket with full peptide backbone flexibility. A starting pose with the peptide near its putative binding pocket is preferred, though in the ab initio mode, no knowledge about the peptide's final conformation is needed.


Design a protein-protein interface anchored by a portion grafted from a native interface.


Grafts an "anchor" region from one protein onto another to help the latter protein successfully bind the former's native partner.


Locate interfaces in the PDB that have "anchor" regions of high loopiness and excellent interface binding energies, to be grafted onto other proteins to aid in dimerization.


Ensure that the Rosetta database is internally consistent (for example, the Dunbrack libraries produce correct binaries).


Atomtree_diff silent files are a particular format of silent file; this application converts them into many PDB files.


Dock ligands to proteins. DEPRECATED BY RosettaScripts movers. Use is not supported and is only relevant for consistency with legacy workflows.


Minimize and repack sidechains, for preparing structures prior to ligand docking; thus, analogous to docking_prepack_protocol.


Take top 5% of poses by total score, sorted by interface energy. Work through one by one and keep poses that are at least min_rmsd away from the poses kept thus far.


Model flexible protein loops with one of many algorithms, including CCD (cyclic coordinate descent) and KIC (kinematic closure).


Generate grids for the apo (unliganded) form of the protein by searching the chemistry around the pocket.


Generate grids for the holo protein-ligand structure.


Orient residues around a ligand to match particular scaffold atoms.


Ab initio fold a membrane protein. Functionally, this is just a wrapper around Abrelax that is responsive to to to membrane-specific options.


Extract and score key motifs from residue side chains from a given PDB and output them.


Identify flexible regions in a pose using the loops framework and use a collection of inverse rotamers drawn from the motif library to attempt to close the loop.


Calculates a rotamer library for a given set of backbone angles for a given ResidueType.


Calculates the unfolded state energy (a type of reference energy used by the mm_std energy function) for a ResidueType.


Generate constraint files based on a provided file of well-formatted NOEs and write assignments based on the structure of the native pose provided.


Design a fixed complex of protein and hydrogen bond surrogate scaffold. HBS are constrained helix mimetics made of canonical or noncanonical amino acids with a macrocycle constraining the first several residues of the backbone.


Design a fixed complex of protein and oligooxopiperazine scaffold. Oligooxopiperazines are helix surface mimetics made of canonical or noncanonical amino acids with six-membered rings constraining the backbone


Design a fixed complex of protein and peptoid scaffold. Peptoids are N-alkyl or N-aryl glycine residues; many peptoid-specific side chains are incorporated into Rosetta.


Optionally extend a peptide a number of residues; perform design and minimization to massage it into the binding pocket. Has both low and high resolution phases.


Anchored, flexible backbone peptide docking to set up structures for pepspec.


Relax a shell of pocket residues around a target exemplar residue.


Obtain target pocket by analyzing a ddg file and evaluating individual residue pockets therein.


Measure statistics and grid data for a desired pocket.


Evaluates a pocket around a desired residue intended to form an "exemplar." See documentation for the series of applications: pocket_relax, pocket_suggest_target_residues_by_ddg, pocket_measure, and make_exemplar.


Sample probe atoms or molecules of desired size around a nucleobase.


Extensive RNA structural sampling and analysis.


Much as the [FeaturesReporter] framework exports a variety of protein features to databases, this application will extract critical features from RNA structures.


Graft multiple RNA sequences together.


Produce an idealized RNA helix from a sequence, for use in subsequent grafting.


Predict what the chem map (a common piece of data used for experimental RNA structure determination) would be for a given structural model.


Given a set of poses, assign and output their suite names (a structural classification).


Thread an RNA sequence onto a template.


Add structural constraints to an RNA pose, including specific constraints on phosphate locations and other atoms, followed by minimization.


Dock a protein to a mineral surface.


Performs stepwise assembly to build protein loops.


Performs stepwise assembly on RNA structures.


Alignment, O2' packing, RMSD, mutation, slicing, silent file creation, and minimization utility algorithms.


General code that will perform Monte Carlo stepwise assembly on anything.


Neural network-enhanced protein structure prediction, using a multiple sequence alignment as input. Requires extras=tensorflow during compilation.


If a design is sufficiently comparable to the native score, we revert


Runs an XML-specified protocol composed of particular Movers, Filters, and TaskOperations that have been made compatible with RosettaScripts.


Runs the DockDesignParser, allows access to protein-DNA information and code, and uses a customized PDB output to add protein-DNA specific information.


Locates exposed beta strands within the provided PDB as a potential target for homodimerization with a designed beta strand.


Actually performs homodimer design on a fixed beta strand scaffold.


Orients two copies of one chain so that a previously exposed beta strand became aligned to each other to form a homodimer.


Given a low resolution structure that therefore only contains a trace of the alpha carbons, expand the CA atoms to full residues and perform extensive sampling.


A dock-design application originally written to exemplify the wonders of what the new job distributor and object-oriented Rosetta 3.0+ can do for you. Docks and designs and minimizes a peptide in a binding pocket with minimal flexibility.


Explores large swaths of conformational space to enumerate possible conformations for a disordered tail of a protein. Does not provide positive information; there is not a fixed conformation for this tail in any event, but it can provide information about whether a region of conformational space is accessible or inaccessible.


Given a starting PDB with a cysteine and a PDB of ubiquitin (though truly applicable to any protein), conjugate the C-terminus of the ubiquitin PDB to the cysteine indicated and then dock the two proteins together subject to that constraint.


Given a starting PDB with a cysteine and a PDB of ubiquitin (though truly applicable to any protein containing a cysteine), form a disulfide from the two cysteines indicated and then dock the two proteins together subject to that constraint.


Given a starting PDB with a lysine and a PDB of ubiquitin (though truly applicable to any protein), conjugate the C-terminus of the ubiquitin PDB to the lysine indicated and then dock the two proteins together subject to that constraint.


Optimize the weights of a provided energy functions to produce excellent sequence recovery and rotamer recovery.


Test an XML script written for the rostta_scripts application against Rosetta's internally-generated XML Schema defintion without running the script.


Test an XML script written for the rostta_scripts application against Rosetta's internally-generated XML Schema defintion and that the Movers/Filters/TaskOperations/etc. that are defined within it can be constructed and initialized without running the script. A strictly more rigorous test than the validate_rosetta_script application.