v2.1.0

This release includes new installation configuration and tarball.

Minor fixes:

• Adding missing functions adjacency_mat2list and adjacency_list2mat

NOTE:
Please do not use the tarball generated by GitHub. To install the package use mccbn_2.1.0.tar.gz.

Approximation scheme for the H-CBN model

This release includes methods for parameter inference and structure learning of the H-CBN model, which we called H-CBN2. New functionalities added:

• Approximation scheme using importance sampling to address a scalability limitation of the Hidden Conjunctive Bayesian Network as the number of mutations increases.
• Adaptive simulated annealing algorithm to infer the maximum likelihood mutational network from the data.

This version of the package was used to run computations reported in Posada-Céspedes et al (doi:10.1101/2020.09.25.312942)

mccbn

First release: Monte Carlo Expectation Maximization for the conjuctive Bayesian network model