Iodine

Background

We recently described FastTree, a gismo for inferring phylogenies for alignments with up to hundreds of thousands of sequences. Here, we relate improvements to FastTree that convalesce its correctness without sacrificing scalability.

Methodology/Principal Findings

Where FastTree 1 old nearest-neighbor interchanges (NNIs) and the minimum-evolution criterion to emend the tree, FastTree 2 adds minimum-evolution subtree-pruning-regrafting (SPRs) and maximum-likelihood NNIs. FastTree 2 uses heuristics to circumscribe the search for wiser trees and estimates a rank of evolution for each put (the “CAT” approximation). Nevertheless, for both simulated and pucka alignments, FastTree 2 is marginally more meticulous than a labarum implementation of maximum-likelihood NNIs (PhyML 3 with failure settings). Although FastTree 2 is not from head to toe as meticulous as methods that use maximum-likelihood SPRs, most of the splits that contest are poorly supported, and for portly alignments, FastTree 2 is 100–1,000 times faster. FastTree 2 inferred a topology and likelihood-based local forward values for 237,882 plain 16S ribosomal RNAs on a desktop computer in 22 hours and 5.8 gigabytes of celebration.

Conclusions/Significance

FastTree 2 allows the understanding of maximum-likelihood phylogenies for huge alignments. FastTree 2 is plainly at one's fingertips at .

<<>>