schrodinger.seam.transforms package¶

Seam Transform Catalog¶

See also:

Transform	Description
`ReadStructuresFromFile`	Read a file (or files) containing structures and return a PCollection of `Structure` objects.
`WriteStructuresToFile`	Write a PCollection of `Structure` objects to a file.
`ReadMolsFromFile`	Read a file (or files) containing molecules.
`WriteMolsToFile`	Write a PCollection of `Mol` objects to a file.
`FixedSample`	Split a PCollection into two PCollections: a random sample of the input PCollection and the remaining elements.
`Top`	Split a PCollection into two PCollections: the largest n elements of the input PCollection and the remaining elements.
`RandomSample`	Create a PCollection made up of a random sample of the input PCollection.
`Smallest`	Create a PCollection containing the smallest element(s) of the input PCollection.
`IPythonInspect`	Starts an IPython shell to interactively inspect the contents of a PCollection.
`ParetoRank`	Associate inputs with the Pareto front for parameter minimization.

Transform	Description
`LigPrep`	Prepare ligands using LigPrep, turning 2D `Mol`s into 3D structures. LigPrep works by expanding tautomeric and ionization states, ring conformations, and stereoisomers consistent with the input mol to fully capture the relevant states of the molecule in 3D. For each input mol, multiple output structures may be generated.

Transforms for generating 3d structures in a variety of conformations and poses.

Transform	Description
`Dock`	Dock ligands into a receptor using Glide.
`FastConfGen`	Generate conformers using Fast3D.
`ShapeScreen`	Generate conformers aligned to shape(s).

Transforms for minimizing structures with various methods.

Transform	Description
`FFMinimize`	Minimize a structure using a force field minimization. Supports independent structures, homogeneous conformers, or custom grouping.
`SetMinimizationEnergy`	Extract energy from `MinimizationResult` and set it as a property on the structure.
`CollectConformers`	Group structures by a conformer key function into lists.
`RunMLFFMinimize`	Run MLFF minimization and yield an MLFFMinimizeResult for each input, regardless of success or failure.
`MLFFMinimize`	Minimize a structure using an MLFF model. Backward-compatible wrapper around RunMLFFMinimize with dropped output for failed structures.

Transforms for preparing and refining protein structures.

Transform	Description
`ProteinPrep`	Clean up and prepare a protein using PrepWizard.
`Refine`	Refine a protein receptor using Prime.

Transforms for taking molecules and enumerating new molecules based on them.

Transform	Description
`Decorate`	Decorates a PCollection of `Mol` objects using `RGroup` objects.
`Fragment`	Generate molecular fragments from input `Mol` objects.
`Synthesize`	Combinatorial retrosynthesis of `Mol` objects.

Transforms for filtering collections of molecules based on various criteria.

Transform	Description
`PropertySpaceFilter`	Filters a PCollection of `Structure` or `Mol` objects based on physicochemical property ranges.
`SubstructureFilter`	Filters a PCollection of `Structure` or `Mol` objects based on substructure matches.

Transforms for training and applying ML models to molecular data.

Transform	Description
`TrainModel`	Train an ML model to predict numeric properties from molecular structure fingerprints.
`PredictProperties`	Predict properties on structures using a trained ML model. Accepts a direct `MLModel` instance or a PCollection side input.