af2rave.amino module

Wrapper module for AMINO.

This module defines a AMINO class that stores the model parameters. The default parameters are best suited for af2rave use and may not be universially applicable. For more general use, please use the cli module by calling af2rave amino

class AMINO(**kwargs)[source]

Bases: object

AMINO module reduces the redudancy in choice of collective variables. Usually the default parameters are best suited for af2rave use. If you have memory issues, consider reducing the number of bins or striding the data. The performance of the code is O(N^2M), where N is the number of order parameters and M is the number of data points. The memory bottleneck is the mutual information calculation.

Parameters:

  • n (int) – The maximum number of order parameters to consider. Default is 20.

  • bins (int) – The number of bins for the computing the mutual information. Default is 50.

  • kde_bandwidth (float) – The bandwidth for the kernel density estimation. Default is 0.02.

  • verbose (bool) – Whether to print the progress. Default is False.

classmethod from_colvar(colvar: Colvar, **kwargs) → AMINO[source]

Run AMINO from a Colvar object. For keyword arguments, see __init__.

Parameters:

colvar (Colvar or str) – Colvar object.

Returns:

AMINO object, with the result stored in result.

Return type:

AMINO

classmethod from_file(filename: str | list[str], **kwargs) → AMINO[source]

Run AMINO from a COLVAR file. For keyword arguments, see __init__.

Parameters:

filename (str | list[str]) – The COLVAR file or files to read.

Returns:

AMINO object, with the result stored in result.

Return type:

AMINO

property result: list[str]

The resulting order parameters from AMINO.

Returns:

The list of order parameters.

Return type:

list[str]

run(label: list[str], data) → None[source]

Run AMINO on the given data. This is not the recommended way to use AMINO. Consider using the class methods from_file and from_colvar instead.

Parameters:

  • label (list[str]) – The list of order parameter labels.

  • data (NDArray) – The data of the order parameters, shaped (n_order_parameters, n_data_points)

to_colvar() → Colvar[source]

Output the chosen order parameters and their input values as a Colvar object. Equivalent to input_colvar.choose(self.result).

Returns:

The Colvar object with the chosen order parameters.

Return type:

Colvar