bdpn 0.1.5

Estimation of BDPN parameters from phylogenetic trees.

bdpn

Estimator of BDPN model parameters from phylogenetic trees.

BD-PN model

BD-PN model has 5 parameters:

• λ -- transmission rate
• ψ -- removal rate
• p -- sampling probability upon removal
• p_n -- probability to notify the last partner upon sampling
• φ -- removal (and sampling) rate after notification

As in the basic birth-death model [Stadler 2009], all the individuals are in the same state. They can transmit with a constant rate λ, get removed with a constant rate ψ, and their pathogen can be sampled upon removal with a constant probability p. On top of that, in the BD-PN model, at the moment of sampling the sampled individual might notify their most recent partner with a constant probability p_n. Upon notification, the partner is removed almost instantaneously (modeled via a constant notified removal rate φ >> ψ) and their pathogen is sampled.

BD-PN model makes 3 assumptions:

1. only observed individuals can notify (instead of any removed individual);
2. notified individuals are always observed upon removal;
3. only the most recent partner can get notified.

Epidemiological parameters:

• R₀=λ/ψ -- reproduction number
• 1/ψ -- infectious time

Installation

To install bdpn:

python3 setup.py install

Usage in command line

BDPN parameter estimation

The following command estimated the BDPN parameters for a given tree tree.nwk and a given sampling probability p=0.4, and saves the estimated parameters to a comma-separated file estimates.csv:

bdpn_infer --p 0.4 --nwk tree.nwk --log estimates.csv

To see detailed options, run:

bdpn_infer --help

PN test

The following command applies the PN test to a given tree tree.nwk and saves the PN-test value to the file cherry_test.txt:

pn_test --nwk tree.nwk --log cherry_test.txt

To see detailed options, run:

pn_test --help