SaveProbabilitiesMetricMover

Back to Mover page.

SaveProbabilitiesMetricMover

General description

Save a PerResidueProbabilitiesMetric to a weights and/or position-specific-scoring-matrix (PSSM) file. The weights file can be loaded as LoadedProbabilitiesMetric and the PSSM can be used with the FavorSequenceProfileMover to add ResidueType constraints for a design run.

Autogenerated Tag Syntax Documentation:

A class to save the probabilities from a PerResidueProbabilitiesMetric in a weights file.

References and author information for the SaveProbabilitiesMetricMover mover:

SaveProbabilitiesMetricMover Mover's author(s): Moritz Ertelt, University of Leipzig moritz.ertelt@gmail.com

<SaveProbabilitiesMetricMover name="(&string;)" metric="(&string;)"
        filename="(&string;)" filetype="(weights &string;)"
        use_cached_data="(false &bool;)" cache_prefix="(&string;)"
        cache_suffix="(&string;)" fail_on_missing_cache="(true &bool;)" />

metric: (REQUIRED) A PerResidueProbabilitiesMetric to calculate the pseudo-perplexity from.
filename: (REQUIRED) The name of the output weights file storing the probabilities.
filetype: The output filetype, either psi-blast pssm (pssm), weights (weights) file or both (both) (default is weights)
use_cached_data: Use any data stored in the datacache that matches the set metrics name (and any prefix/suffix.) Data is stored during a SimpleMetric's apply function, which is called during RunSimpleMetrics
cache_prefix: Any prefix used during apply (RunSimpleMetrics), that we will match on if use_cache is true
cache_suffix: Any suffix used during apply (RunSimpleMetrics), that we will match on if use_cache is true
fail_on_missing_cache: If use_cached_data is True and cache is not found, should we fail?

Example

Here we use the ESM language model to predict amino acid probabilities and save them both as PSSM and weights file.

<ROSETTASCRIPTS>
    <RESIDUE_SELECTORS>
        <Chain name="res" chains="A" />
    </RESIDUE_SELECTORS>
    <SIMPLE_METRICS>
        ----------------- Define model to use for prediction -----------------------------
        <PerResidueEsmProbabilitiesMetric name="esm" residue_selector="res" model="esm2_t30_150M_UR50D" multirun="true"/>
    </SIMPLE_METRICS>
    <FILTERS>
    </FILTERS>
    <MOVERS>
        <RunSimpleMetrics name="inference" metrics="esm"/>
        -------------------------- Setup saving ---------------------------------------------
        <SaveProbabilitiesMetricMover name="save" metric="esm" filename="probs" filetype="both" use_cached_data="true"/>
    </MOVERS>
    <PROTOCOLS>
        <Add mover_name="inference"/>
        <Add mover_name="save"/>
    </PROTOCOLS>
</ROSETTASCRIPTS>

Reference

This is currently unpublished.

SaveProbabilitiesMetricMover