Table 2.

Large structural variants found in biological data sets

Variant	Sample	VALOR2		Long Ranger		LUMPY		DELLY		TARDIS
		Pred.	Known∗	Pred.	Known∗	Pred.	Known∗	Pred.	Known ∗	Pred.	Known∗
Deletions	NA19238	8	8	1	1	81	49	192	127	14	13
	NA19239	10	10	3	3	104	64	232	157	17	14
	NA19240	11	11	2	2	95	59	228	157	15	14
	NA12878	14	14	18	18	138	62	273	170	20	20
	CHM1	9	8	109	72	106	47	226	113	20	19
	CHM13	7	7	95	65	78	43	660	423	10	8
Inversions	NA19238	56	17	2	2	3	0	407	37	14	1
	NA19239	49	15	1	1	4	0	406	33	11	0
	NA19240	89	25	3	2	4	0	435	31	9	1
	NA12878	33	12	5	1	3	0	415	37	43	1
	CHM1	35	26	2	2	3	0	259	23	22	1
	CHM13	40	28	2	2	5	0	1496	65	50	0
Duplications ‡	NA19238	9	5	3	3	142	91	307	183	77	46
	NA19239	9	5	0	0	158	96	298	189	79	42
	NA19240	19	8	2	2	139	91	284	187	82	47
	NA12878	6	4	20	19	196	93	341	184	293	133
	CHM1	5	3	0	0	164	83	289	138	131	64
	CHM13	7	3	0	0	519	276	1425	784	329	196
Translocations	NA19238	1	0	0	0	336	0	8788	0	N/A	N/A
	NA19239	3	0	0	0	368	0	8946	0	N/A	N/A
	NA19240	1	0	0	0	362	0	9250	0	N/A	N/A
	NA12878	1	0	1	0	842	0	9770	0	N/A	N/A
	CHM1	0	0	0	0	320	0	6511	0	N/A	N/A
	CHM13	0	0	0	0	184	0	117667	0	N/A	N/A

Similar to Table 1, we only report large SVs we discovered in real data sets (> 80 kbp for inversions, > 40 kbp for duplications, > 100 kbp for deletions, and > 100 kbp for translocations). We ran LUMPY using the smoove wrapper as recommended by the authors. Note that TARDIS does not predict translocations. ^‡We merged tandem and interspersed duplications in this table since Long Ranger, LUMPY, and DELLY do not differentiate between them. ^∗For CNVs (deletions and duplications), known variants refer to those that are reported in dbVar [52] non-redundant call set (https://ftp.ncbi.nlm.nih.gov/pub/dbVar/sandbox/sv_datasets/nonredundant/). For balanced rearrangements (inversions and translocations), we used the gnomAD [53] v2.1.1 call set, lifted over to GRCh38 (https://storage.googleapis.com/gnomad-public/papers/2019-sv/gnomad_v2.1_sv.sites.vcf.gz)