FastTree infers approximately-maximum-likelihood phylogenetic trees from alignments of 

nucleotide or protein sequences. FastTree can handle alignments with up to a million 

of sequences in a reasonable amount of time and memory. For large alignments, FastTree 

is 100-1,000 times faster than PhyML 3.0 or RAxML 7. FastTree is open-source software -- you 

can download the code below.

FastTree is more accurate than PhyML 3 with default settings, 

and much more accurate than the distance-matrix methods that are 

traditionally used for large alignments. FastTree uses the Jukes-Cantor or generalized time-

reversible (GTR) models of nucleotide evolution and the JTT (Jones-Taylor-Thornton 1992) model 

of amino acid evolution. To account for the varying rates of evolution across sites, FastTree 

uses a single rate for each site (the "CAT" approximation). To quickly estimate the 

reliability of each split in the tree, FastTree computes local support values with 

the Shimodaira-Hasegawa test (these are the same as PhyML 3's "SH-like local supports").