Novel Distances for Dollo Data

TL;DR

This paper introduces the Additive Dollo Distance (ADD), a new binary distance measure tailored for Dollo process data, demonstrating its theoretical advantages and superior performance over existing distances through simulations and real data applications.

Contribution

The paper presents the ADD, a novel distance for Dollo data, with proven consistency, theoretical properties, and improved performance over existing methods in simulations and empirical datasets.

Findings

ADD outperforms other binary distances on Dollo data.

LogDet distance performs poorly with Dollo data, affecting genome reconstruction.

ADD provides consistent and theoretically sound results for Dollo process data.

Abstract

We investigate distances on binary (presence/absence) data in the context of a Dollo process, where a trait can only arise once on a phylogenetic tree but may be lost many times. We introduce a novel distance, the Additive Dollo Distance (ADD), which is consistent for data generated under a Dollo model, and show that it has some useful theoretical properties including an intriguing link to the LogDet distance. Simulations of Dollo data are used to compare a number of binary distances including ADD, LogDet, Nei Li and some simple, but to our knowledge previously unstudied, variations on common binary distances. The simulations suggest that ADD outperforms other distances on Dollo data. Interestingly, we found that the LogDet distance performs poorly in the context of a Dollo process, which may have implications for its use in connection with conditioned genome reconstruction. We apply the ADD to two Diversity Arrays Technology (DArT) datasets, one that broadly covers Eucalyptus species and one that focuses on the Eucalyptus series Adnataria. We also reanalyse gene family presence/absence data on bacteria from the COG database and compare the results to previous phylogenies estimated using the conditioned genome reconstruction approach.