treePL: divergence time estimation using penalized likelihood for large phylogenies

Ever larger phylogenies are being constructed due to the explosion of genetic data and development of high-performance phylogenetic reconstruction algorithms. However, most methods for calculating divergence times are limited to datasets that are orders of magnitude smaller than recently published l...

Full description

Saved in:
Bibliographic Details
Published in:Bioinformatics (Oxford, England) England), 2012-10, Vol.28 (20), p.2689-2690
Main Authors: SMITH, Stephen A, O'MEARA, Brian C
Format: Article
Language:English
Subjects:
Algorithms
Biological and medical sciences
Fundamental and applied biological sciences. Psychology
General aspects
Likelihood Functions
Mathematics in biology. Statistical analysis. Models. Metrology. Data processing in biology (general aspects)
Phylogeny
Software
Citations: Items that this one cites
Items that cite this one
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:Ever larger phylogenies are being constructed due to the explosion of genetic data and development of high-performance phylogenetic reconstruction algorithms. However, most methods for calculating divergence times are limited to datasets that are orders of magnitude smaller than recently published large phylogenies. Here, we present an algorithm and implementation of a divergence time method using penalized likelihood that can handle datasets of thousands of taxa. We implement a method that combines the standard derivative-based optimization with a stochastic simulated annealing approach to overcome optimization challenges. We compare this approach with existing software including r8s, PATHd8 and BEAST. Source code, example files, binaries and documentation for treePL are available at https://github.com/blackrim/treePL.
ISSN:1367-4803
1367-4811
DOI:10.1093/bioinformatics/bts492