# Deep learning from phylogenies to uncover the epidemiological dynamics of outbreaks

## TL;DR

This paper introduces a deep learning method to analyze genetic data and understand how epidemics spread, using phylogenies to infer outbreak dynamics quickly and accurately.

## Contribution

The novel contribution is a likelihood-free deep learning approach that uses phylogenies for fast and accurate epidemiological inference.

## Key findings

- The method outperforms state-of-the-art techniques in speed and accuracy on simulated data.
- It successfully identifies the impact of superspreading in an HIV dataset from Zurich.
- The approach supports model selection and parameter estimation for large phylogenies.

## Abstract

Widely applicable, accurate and fast inference methods in phylodynamics are needed to fully profit from the richness of genetic data in uncovering the dynamics of epidemics. Standard methods, including maximum-likelihood and Bayesian approaches, generally rely on complex mathematical formulae and approximations, and do not scale with dataset size. We develop a likelihood-free, simulation-based approach, which combines deep learning with (1) a large set of summary statistics measured on phylogenies or (2) a complete and compact representation of trees, which avoids potential limitations of summary statistics and applies to any phylodynamics model. Our method enables both model selection and estimation of epidemiological parameters from very large phylogenies. We demonstrate its speed and accuracy on simulated data, where it performs better than the state-of-the-art methods. To illustrate its applicability, we assess the dynamics induced by superspreading individuals in an HIV dataset of men-having-sex-with-men in Zurich. Our tool PhyloDeep is available on github.com/evolbioinfo/phylodeep.

Widely applicable, accurate and fast inference methods in phylodynamics are needed to fully profit from the richness of genetic data in uncovering the dynamics of epidemics. Here, the authors develop a likelihood-free, simulation-based deep learning approach.