Removing reference mapping biases using limited or no genotype data identifies allelic differences in protein binding at disease-associated loci

Buchkovich, Martin L.; Eklund, Karl; Duan, Qing; Li, Yun; Mohlke, Karen L.; Furey, Terrence S.

doi:10.1186/s12920-015-0117-x

Table 2 Allelic imbalance detection accuracy in alignments using partial or no genotypes compared to complete genotypes

From: Removing reference mapping biases using limited or no genotype data identifies allelic differences in protein binding at disease-associated loci

Factor/Assay (Condition)	Complete^a			Partial Genotype^b Imbalances									No Genotype^c Imbalances
	Total	Partial	None	Total			Known variants			Predicted variants			Total			Known variants			Predicted variants
	N_t	N_imp	N_com	N_t	Sens	Prec	N_imp	Sens	Prec	N_t-N_imp	Sens	Prec	N_t	Sens	Prec	N_com	Sens	Prec	N_t-N_com	Sens	Prec
CREB1 (50 bp)	200	125	141	190	73.0	76.8	134	96.8	90.3	56	33.3	44.6	203	76.0	74.9	160	93.6	82.5	43	33.9	46.5
CREB1 (35 bp)	106	70	81	104	73.6	75.0	74	97.1	91.9	30	27.8	33.3	107	77.4	76.6	87	92.6	86.2	20	28.0	35.0
CREB1 (20 bp)	26	16	16	24	69.2	75.0	17	100.0	94.1	7	20.0	28.6	22	69.2	81.8	17	100.0	94.1	5	20.0	40.0
CTCF (35 bp)	267	187	192	300	83.1	74.0	198	98.4	92.9	102	47.5	37.3	298	85.0	76.2	210	97.9	89.5	88	52.0	44.3
DNase (20 bp)	104	43	47	138	51.0	38.4	42	97.7	100.0	96	18.0	11.5	144	51.9	37.5	55	97.9	83.6	89	14.0	9.0
CREB1 (2 alns)^d	200	125	141	195	78.5	80.5	135	97.6	90.4	60	46.7	58.3	204	77.0	75.5	156	92.2	83.3	48	40.7	50.0
Mismatches alllowed
CREB1 (0 mm)	199	122	138	137	58.8	85.4	137	95.9	85.4	0	-	-	160	63.3	78.8	160	91.3	78.8	0	-	-
CREB1 (1 m m)^e	200	125	141	190	73.0	76.8	134	96.8	90.3	56	33.3	44.6	203	76.0	74.9	160	93.6	82.5	43	33.9	46.5
CREB1 (2 mm)	199	124	137	245	80.4	65.3	133	97.6	91.0	112	52.0	34.8	251	81.4	64.5	159	96.4	83.0	92	48.4	32.6
CREB1 (3 mm)	213	123	143	301	79.2	53.2	132	98.4	90.9	169	50.0	23.7	313	81.7	52.7	161	96.4	83.9	152	47.6	19.7
Minimum reads/allele
CREB1 (2 reads)	301	178	199	486	73.4	45.5	187	97.2	92.5	299	39.0	16.1	515	75.4	44.1	228	95.0	82.9	287	37.3	13.2
CREB1 (3 reads)	261	156	173	267	70.1	68.8	162	94.9	91.4	105	33.3	33.3	289	72.8	65.7	191	92.5	83.8	98	34.1	30.6
CREB1 (4 reads)	230	142	159	218	71.4	76.0	148	95.1	91.2	70	33.7	42.6	235	74.8	73.2	175	92.5	84.0	60	35.2	41.7
CREB1 (5 reads)^e	200	125	141	190	73.0	76.8	134	96.8	90.3	56	33.3	44.6	203	76.0	74.9	160	93.6	82.5	43	33.9	46.5
CREB1 (6 reads)	198	122	136	174	70.7	80.5	130	96.7	90.8	44	28.9	50.0	188	73.7	77.7	153	93.4	83.0	35	30.6	54.3
CREB1 (7 reads)	173	109	123	154	72.8	81.8	116	97.2	91.4	38	31.2	52.6	167	75.7	78.4	138	92.7	82.6	29	34.0	58.6
CREB1 (8 reads)	157	100	111	141	72.0	80.1	107	97.0	90.7	34	28.1	47.1	148	75.2	79.7	124	92.8	83.1	24	32.6	62.5
CREB1 (9 reads)	144	91	101	130	72.2	80.0	98	96.7	89.8	32	30.2	50.0	140	75.7	77.9	115	93.1	81.7	25	34.9	60.0
CREB1 (10 reads)	124	80	88	117	74.2	78.6	88	96.2	87.5	29	34.1	51.7	125	76.6	76.0	102	92.0	79.4	23	38.9	60.9
CREB1 (15 reads)	88	60	66	82	77.3	82.9	66	96.7	87.9	16	35.7	62.5	88	80.7	80.7	76	92.4	80.3	12	45.5	83.3
CREB1 (20 reads)	63	47	52	64	84.1	82.8	53	97.9	86.8	11	43.8	63.6	67	88.9	83.6	60	96.2	83.3	7	54.5	85.7
Imputation Rsq threshold
CREB1 (Rsq > .3) ^e	200	125	-	190	73.0	76.8	134	96.8	90.3	56	33.3	44.6	-	-	-	-	-	-	-	-	-
CREB1 (Rsq > .4)	200	122	-	190	72.5	76.3	133	97.5	89.5	57	33.3	45.6	-	-	-	-	-	-	-	-	-
CREB1 (Rsq > .5)	200	121	-	187	72.5	77.5	129	98.3	92.2	58	32.9	44.8	-	-	-	-	-	-	-	-	-
CREB1 (Rsq > .6)	200	118	-	186	72.5	78.0	124	98.3	93.5	62	35.4	46.8	-	-	-	-	-	-	-	-	-
CREB1 (Rsq > .7)	200	117	-	185	72.0	77.8	123	98.3	93.5	62	34.9	46.8	-	-	-	-	-	-	-	-	-
CREB1 (Rsq > .8)	200	104	-	182	70.5	77.5	111	99.0	92.8	71	39.6	53.5	-	-	-	-	-	-	-	-	-
CREB1 (Rsq > .9)	200	96	-	176	69.5	79.0	99	99.0	96.0	77	42.3	57.1	-	-	-	-	-	-	-	-	-

^aComplete genotype alignments use sequencing-based genotypes ^bPartial genotype alignments use array-based genotypes and imputation ^cNo genotypes alignments use common variants (MAF > .05) from 1000 Genomes EUR ^dImbalances called after a second alignment using refined genotypes; known variants are variants included in the first alignment ^eCondition used by default by AA-ALIGNER; N_t total imbalance count, N_imp imbalances at heterozygous sites identified by imputation, N_com imbalances at common variants, Sens, percent sensitivity, Prec, percent precision

Back to article page

ISSN: 1755-8794

Contact us

Submission enquiries: bmcmedicalgenomics@biomedcentral.com
General enquiries: ORSupport@springernature.com

BMC Medical Genomics

Contact us