Table 1. ML scores (as log likelihoods) of the solutions obtained by the ML methods on the three largest biological datasets.

Alignment	ML Method	16S.B.ALL	16S.T	16S.3
TrueAln	RAxML	−1,589,345	−1,727,411	−1,376,372
	FastTree	−1,607,277	−1,731,963	−1,380,308
	RAxML-Limited	−1,601,407	−1,733,148	−1,380,946
SATé	RAxML	n.d.	−1,650,778	−1,316,340
	FastTree	n.d.	−1,654,334	−1,319,623
	RAxML-Limited	n.d.	−1,655,854	−1,317,604
MAFFT	RAxML	n.d.	−1,645,509	n.d.
	FastTree	n.d.	−1,649,615	n.d.
	RAxML-Limited	n.d.	−1,647,176	n.d.
PartTree	RAxML	−1,973,592	−1,760,582	−1,346,922
	FastTree	−2,032,424	−1,766,371	−1,350,858
	RAxML-Limited	−2,013,115	−1,762,611	−1,351,822
ClustalW	RAxML	n.d.	−1,716,012	−1,365,614
	FastTree	n.d.	−1,719,138	−1,369,027
	RAxML-Limited	n.d.	−1,720,420	−1,368,776
Quicktree	RAxML	−1,878,299	−3,439,265	−2,451,782
	FastTree	−1,903,840	−3,449,574	−2,459,658
	RAxML-Limited	−1,891,570	−3,448,084	−2,460,095

Some alignments were missing due to high memory requirements of an alignment method on datasets of this size, preventing ML methods from running; these entries are marked “n.d.”. n = 1 for each reported value.