schrodinger.application.transforms.ligprep.neutralizer module

Apache Beam transform for neutralizing structures.

A thin Beam adapter around the legacy ligprep3 neutralizer driver (schrodinger.application.macromodel.packages.ligprep3.neutralizer._run), which converts charged groups to their neutral protonation state. The legacy chemistry is reused verbatim (LIGPREP-2231).

class schrodinger.application.transforms.ligprep.neutralizer.NeutralizerConfig(*, max_atom: int = 500, retain: bool = False, retain_lab: bool = False)

Bases: BaseModel

Configuration for the neutralizer transform.

Defaults match the legacy ligprep workflow’s invocation, which coincide with the standalone CLI defaults.

max_atom: int

retain: bool

retain_lab: bool

model_config: ClassVar[ConfigDict] = {}

Configuration for the model, should be a dictionary conforming to [ConfigDict][pydantic.config.ConfigDict].

class schrodinger.application.transforms.ligprep.neutralizer.Neutralizer(*, _config: pydantic.main.BaseModel | None = None, **kwargs)

Bases: PTransformWithConfig

Neutralize structures, wrapping the legacy ligprep3 neutralizer logic.

Structures that fail neutralization (e.g. exceeding the atom limit) are dropped. See NeutralizerConfig for the parameters.

Example usage:

>>> from schrodinger import adapter
>>> import apache_beam as beam
>>> st = adapter.to_structure('CC(=O)[O-]')
>>> with beam.Pipeline() as p:
...     neutralized = (p
...         | beam.Create([st])
...         | Neutralizer())

config_class

alias of NeutralizerConfig

expand(pcoll) → OutcomePCollections

Neutralize a PCollection of structures.

Parameters:

pcoll – PCollection of Structure objects

Returns:

OutcomePCollections with the neutralized structures (.main) and DroppedRecord objects for those that failed (.dropped)