isoCNV: in silico optimization of copy number variant detection from targeted or exome sequencing data

BMC Bioinformatics

Table 2 Benchmark results for the individual callings and the in silico validation dataset

Dataset	Method	TP	TN	FP	FN	Total	Sensitivity	Specificity	PPV	NPV	F-score
ICR96	DECoN	247	27,330	60	49	27,686	0.8345	0.9978	0.8046	0.9982	0.8192
	CNVkit	205	27,225	91	165	27,686	0.5541	0.9967	0.6827	0.9940	0.6156
	panelcn.MOPS	278	27,061	18	329	27,686	0.4580	0.9993	0.9392	0.9880	0.6157
	In silico validation	58	27,390	0	238	27,686	0.1959	1	1	0.9914	0.3277
NimbleGen	DECoN	220	7236	43	1138	8637	0.1620	0.9941	0.8365	0.8641	0.2714
	CNVkit	777	7274	582	4	8637	0.9949	0.9259	0.5717	0.9995	0.7262
	panelcn.MOPS	736	6891	619	391	8637	0.6531	0.9176	0.5432	0.9463	0.5931
	In silico validation	30	7278	0	1329	8637	0.0220	1	1	0.8456	0.0432

ISSN: 1471-2105