Skip to main content

Clonal deconvolution in cancer using longitudinal NGS data

Project description

Introduction

Tumours are mixtures of phylogenetically related cancer cell populations or clones, which are subject to a process of Darwinian evolution in response to selective pressures in their local micro-environment. clonosGP is a statistical methodology for tracking this latent heterogeneity continuously in time based on longitudinally collected tumour samples. In technical terms, it combines Dirichlet Process Mixture Models with Gaussian Process Priors to identify clusters of mutations and track their cellular prevalence continuously in time. If only cross-sectional data are available, then it performs standard non-parametric clustering of mutations based on their observed frequency, similarly to PyClone and other software in the same category. The statistical models underlying clonosGP were implemented in the excellent probabilistic programming system PyMC3 on which we also rely for inference using variational methods.

Installation

clonosGP requires Python 3.7 or later. It can be easilly installed as follows:

  1. Create a virtual environment: python3 -m venv myenv
  2. Activate the newly created environment: source myenv/bin/activate
  3. Install clonosGP as follows: pip install -U clonosGP

All necessary dependencies will also be installed.

Usage

A guide to start using clonosGP quickly is available here. A more thorough tutorial can be found here.

Citation

For citation information check http://github.com/dvav/clonosGP

Project details


Release history Release notifications | RSS feed

This version

1.0

Download files

Download the file for your platform. If you're not sure which to choose, learn more about installing packages.

Source Distribution

clonosGP-1.0.tar.gz (12.5 kB view hashes)

Uploaded Source

Supported by

AWS AWS Cloud computing and Security Sponsor Datadog Datadog Monitoring Fastly Fastly CDN Google Google Download Analytics Microsoft Microsoft PSF Sponsor Pingdom Pingdom Monitoring Sentry Sentry Error logging StatusPage StatusPage Status page