Skip to main content

Table 4 Empirical power at 0.05 using collapsing and single-marker approaches at all 22 causal genes for long-term average phenotype

From: Whole genome sequence analysis of the simulated systolic blood pressure in Genetic Analysis Workshop 18 family data: long-term average and collapsing methods

 

Causal genes

Total SNPs

Causal SNPs1

% Variance explained2

Empirical Power

  

Common

Rare3

Common

Rare

Common

Rare

Collapse all SNPs

Collapse rare SNPs

Best rare4

1

MAP4

149

739

3

12

0.01456

0.06336

0.99

0.17

1.00

 

FLNB

321

628

1

5

0.00273

0.00007

0.69

0.03

0.20

 

ARHGEF3

899

1303

2

8

0.00001

0.00006

0.28

0.07

0.11

 

TUSC2

1

9

0

0

  

0.21

0.04

 
 

ABTB1

7

41

2

0

0.00132

 

0.18

0.02

 
 

NMNAT3

157

332

3

6

0.00014

0.00017

0.15

0.00

0.09

2

ZBTB38

175

408

5

4

0.00087

0.00003

0.09

0.15

0.14

 

DNASE1L3

57

79

4

4

0.00005

0.00026

0.01

0.34

0.08

 

LOC152217

3

4

1

0

0.00001

 

0.04

0.28

 
 

SCAP

31

174

0

2

 

0.00004

0.03

0.51

0.05

 

BTD

72

216

0

8

 

0.00011

0.03

0.08

0.07

 

FBLN2

230

452

1

3

0.00002

0.00006

0.02

0.06

0.04

3

CXCR6

5

18

0

1

 

<1.0E-5

0.12

0.03

0.28

 

PDCD6IP

145

318

2

3

0.00025

0.00003

0.04

0.14

0.15

 

SUMF1

236

502

1

2

0.00008

<1.0E-5

0.03

0.06

0.16

 

SENP5

130

275

0

5

 

0.00007

0.02

0.03

0.12

 

PTPLB

158

329

1

2

0.00002

0.00004

0.02

0.03

0.10

 

PPP2R3A

198

860

1

11

<1.0E-5

0.00010

0.02

0.01

0.12

4

SEMA3F

51

83

0

2

 

0.00001

0.05

0.03

0.05

 

TFDP2

363

852

0

5

 

0.00005

0.04

0.03

0.04

 

PAK2

294

516

0

0

  

0.04

0.02

 
 

B4GALT4

90

126

1

0

0.00002

 

0.01

0.00

 

Overall power

      

0.139

0.095

 

Type I error

      

0.071

0.057

 
  1. Bold type for % variance indicates 3 largest among common variants and 2 largest among rare variants; bold type for empirical power indicates the highest across three analysis options.
  2. 1Number of causal single-nucleotide polymorphisms (SNPs) is computed based on the used annotation file (not based on the answers).
  3. 2% Variance explained is computed by adding the percent variance explained by each SNP based on the used annotation file (not based on the answers).
  4. 3Rare SNPs are defined as SNPs with minor allele frequencies (MAFs) less than 0.5.
  5. 4Best rare is the best power across all rare variants in a gene.