Home

pyNBS: A Python 2.7 implementation for network based stratification of tumor mutations

Welcome to the documentation homepage for pyNBS. This wiki will explain in detail all of the modules and functions available in pyNBS as well as describe how to execute the pyNBS code from the command line. pyNBS walkthroughs using Jupyter notebooks can be found in the Examples/ folder in the repository.

pyNBS Overview

pyNBS is a Python 2.7 implementation of the network-based stratification (NBS) algorithm formerly described by Hofree et al. (Nature Methods 2013). The NBS algorithm leverages information from molecular interaction networks to guide the stratification of tumor somatic mutation profiles into clinically relevant subtypes. Tumor somatic mutation profiles are generally quite sparse, and clustering tumors based only on this information has proven to be difficult due to their genomic heterogeneity between samples. The NBS algorithm attempts to address the challenges of stratifying sparse binary somatic mutation profiles by using a consensus clustering framework and molecular network information to aggregate samples with mutations in genes near one another in the molecular network. The underlying assumption driving NBS is that the same exact genes do not need to be disrupted to drive cancer, but rather simply genes involved in the same pathways and molecular mechanisms need to be disrupted to give rise to cancer. pyNBS adapts the original Matlab code provided by Hofree et al in 2013 into Python while improving the transparency, modularity, and flexibility of the algorithm.