Table 3.

Performance on Illumina datasets

		Original	Allpaths-LG		Coral		ECHO		Fiona-H		HiTEC		Quake
Dataset		e-rate	e-rate	gain	e-rate	gain	e-rate	gain	e-rate	gain	e-rate	gain	e-rate	gain
C.elegans	30×	0.4404	0.3152	28.42	0.5173	−17.45	—a		0.3292	25.24	—b		0.3881	14.02
D.melanogaster	5×	1.3244	0.8900	32.80	0.7187	45.73	0.7652	42.22	0.6587	50.26	1.0080	23.89	0.7342	46.95
D.melanogaster	28×	1.1065	0.7509	32.13	0.7727	30.16	—a		0.7592	31.39	—b		0.8005	29.09
E.coli K-12	30×	0.2070	0.0015	99.29	0.0178	91.40	0.0184	91.12	0.0043	97.91	0.0141	93.18	0.0094	95.55
E.coli K-12	490×	0.2070	0.0036	98.24	0.0084	95.69	—b		0.0028	98.66	0.0128	93.81	0.0077	96.35
P.syringae	42×	0.5984	0.1531	74.41	0.1491	75.08	0.0533	91.09	0.0517	91.36	0.1351	77.42	0.1219	79.65
S.cerevisae	22×	0.4101	0.1803	56.04	0.2713	33.86	0.2777	32.30	0.1686	58.89	0.5045	−23.00	0.1963	52.15

Note: This table shows the same metrics as Table 1. ^aThe program ran too long—The problems were Coral and HiTEC produced a segfault, requiring more than 72 GB of memory. ^bThe program crashed. ECHO was killed after running more than 4 days in the case of C.elegans and D.melanogaster datasets and the subprogram NeighborJoin crashed on the full E.coli dataset;