dislib.cluster.Daura

class dislib.cluster.daura.base.Daura(cutoff)

Bases: BaseEstimator

Daura clustering.

Distributed implementation of the distances-based Daura clustering, introduced on Daura et al. 1. A description of the algorithm can be found on:

http://lockhartlab.squarespace.com/blog/2018/1/14/clustering-with-daura-et-al.

Parameters

cutoff (float) – Distance to determine the neighbors of a sample.

References

1

Daura, X., Gademann, K., Jaun, B., Seebach, D., van Gunsteren, W.F. and Mark, A.E. (1999). Peptide Folding: When Simulation Meets Experiment. In Angewandte Chemie International Edition, 38 (pp. 236-240).

fit_predict(distances)

Compute Daura clustering.

Parameters

distances (ds-array (n_samples, n_samples)) – Pairwise distances between the samples.

Returns

clusters – A list of clusters. Each cluster is a list of sample indices, starting with the cluster center.

Return type

List[ List[ int ] ]