msmtools.analysis.pcca(T, m)

Compute meta-stable sets using PCCA++ _[1] and return the membership of all states to these sets.

  • T ((n, n) ndarray or scipy.sparse matrix) – Transition matrix
  • m (int) – Number of metastable sets

clusters – Membership vectors. clusters[i, j] contains the membership of state i to metastable state j

Return type:

(n, m) ndarray


Perron cluster center analysis assigns each microstate a vector of membership probabilities. This assignement is performed using the right eigenvectors of the transition matrix. Membership probabilities are computed via numerical optimization of the entries of a membership matrix.


[1]Roeblitz, S and M Weber. 2013. Fuzzy spectral clustering by PCCA+: application to Markov state models and data classification. Advances in Data Analysis and Classification 7 (2): 147-179