Skip to main content

Table 1 Top 10 association rules regarding their p-values in the replication dataset (TGEN + BoMa)

From: Detecting significant genotype–phenotype association rules in bipolar disorder: market research meets complex genetics

PID

Groups

Statistics

Adjusted p-value

GP

Gp

gP

gp

p_chisq

Odds ratio (0.95 CI)

Bonferroni

FDR

12978

25

105

107

1598

3.576e−08

3.566 [2.169–5.681]

0.00075

0.00075

6221

26

84

162

1563

1.780e−06

2.995 [1.841–4.730]

0.03717

0.01859

12681

33

103

187

1512

4.648e−06

2.596 [1.682–3.917]

0.09706

0.02771

12981

25

129

107

1574

5.720e−06

2.860 [1.751–4.520]

0.11944

0.02771

6225

25

84

163

1563

6.635e−06

2.862 [1.747–4.545]

0.13855

0.02771

6228

26

93

162

1554

1.585e−05

2.690 [1.661–4.225]

0.33102

0.05517

4428

31

88

212

1504

2.021e−05

2.505 [1.600–3.830]

0.42198

0.06028

6111

21

109

111

1594

4.096e−05

2.779 [1.636–4.529]

0.85530

0.10654

6183

20

66

168

1581

4.592e−05

2.864 [1.652–4.765]

0.95887

0.10654

6178

20

68

168

1579

7.577e−05

2.777 [1.604–4.611]

1

0.15823

  1. Listed are the rule identifier (PID), the counts per group of the contingency tables, the p-values based on the Chi squared test along with the odds ratios including confidence intervals (CI), and results from two multiple correction methods (based on 20,882 tests). The coding of the groups is as follow: G, if genotype pattern is present, g if not; P, if phenotype pattern is present, p if not. FDR false discovery rate