CLARK: fast and accurate classification of metagenomic and genomic sequences using discriminative k -mers

BMC Genomics

Table 2 Species-level classification accuracy and speed of CLARK, KRAKEN , and NBC for four simulated metagenomes

	HiSeq			MiSeq			simBA-5			simHC.20.500
	*Prec*	*Sens*	*Speed*	*Prec*	*Sens*	*Speed*	*Prec*	*Sens*	*Speed*	*Prec*	*Sens*	*Speed*
NBC (k=15)	68.67	68.70	0.008	68.33	68.33	0.007	91.74	91.74	0.007	94.32	94.32	0.005
Clark (k=20)	69.44	61.46	272	70.72	62.45	239	91.32	82.48	269	94.34	94.32	96
Kraken (k=31)	74.00	53.49	2,332	77.72	58.72	1,361	92.99	78.70	1,976	84.67	84.31	237
Clark (k=31)	86.74	58.59	3,011	89.49	61.84	1,566	98.85	76.80	2,855	94.67	94.26	251
Kraken-Q (k=31)	75.88	50.78	6,224	78.07	53.68	5,308	92.67	74.39	7,023	82.40	74.84	3,809
Clark-E (k=31)	90.08	55.18	30,976	94.31	58.36	24,029	98.92	66.02	24,996	92.78	92.38	15,583
Clark-l (k=27)	85.35	53.95	1,676	85.89	64.91	904	85.55	46.28	1,702	94.06	93.53	141

Precision and sensitivity are expressed as percentages, while speed is expressed in 10³ reads per minute for NBC, KRAKEN, and CLARK on the classification of “HiSeq”, “MiSeq”, “simBA-5” and “simHC.20.500” metagenome datasets against the 1473 species-level targets, in single-threaded mode.