Kpax3 is a Julia package for inferring the group structure of genetic sequences. In general, any multivariate categorical dataset (such as presence/absence data) can be analyzed by Kpax3. Output consists of a clustering of both the rows (statistical units) and columns (statistical variables) of the provided data matrix. It is an improved version of kpax2, providing an MCMC algorithm for a proper Bayesian approach and a genetic algorithm for MAP estimation.

Code available on GitHub.

Alberto Pessia
Postdoctoral researcher


(2018). Kpax3: Bayesian bi-clustering of large sequence datasets. Bioinformatics, 34(12): 2132–2133. doi: 10.1093/bioinformatics/bty056.

PDF Code Project DOI URL


Bayesian bi-clustering of categorical data

Cluster analysis is a common statistical technique for partitioning the observed data into disjoint homogeneous groups. In the presence …