Table 1 Proportion of phenotypic variance predicted in stable emotional problems, and in 12 anxiety and depression measures, by polygenic scores for Major Depressive Disorder and for General Anxiety Disorder

From: Extracting stability increases the SNP heritability of emotional problems in young people

 

R 2

p-value

Empirical p-value

 

PGC MDD

UKBB ANX

PGC MDD

UKBB ANX

PGC MDD

UKBB ANX

Stable CFA score

0.0048

0.0041

<0.0001

<0.0001

0.0001*

0.0001*

Stable IRT score

0.0042

0.0035

<0.0001

<0.0001

0.0001*

0.0001*

Parent-rated SDQ at 7

0.0007

0.0012

0.0481

0.0115

0.1305

0.038*

Teacher-rated SDQ at 7

0.0031

0.003

0.0002

0.0003

0.0013*

0.0010*

Self-rated MFQ at 12

0.0041

0.0008

<0.0001

0.0544

0.0002*

0.1623

Self-rated SDQ at 12

0.001

0.0022

0.0362

0.0014

0.1041

0.0046*

Parent-rated SDQ at 12

0.0017

0.0015

0.0054

0.0087

0.0186*

0.0313*

Parent-rated MFQ at 12

0.0015

0.0025

0.0086

0.0021

0.0277*

0.0030*

Teacher-rated SDQ at 12

0.0008

0.0012

0.0791

0.0019

0.2087

0.034*

Self-rated SDQ at 16

0.0015

0.0031

0.0106

0.0002

0.0325*

0.0005*

Self-rated CASI at 16

0.003

0.0024

0.0006

0.0013

0.0022*

0.0054*

Self-rated MFQ at 16

0.0018

0.0013

0.0075

0.0241

0.0217*

0.0771*

Parent-rated MFQ at 16

0.003

0.0016

0.0007

0.0135

0.0023*

0.0135*

Parent-rated ARBQ at 16

0.003

0.0017

0.0004

0.0089

0.0017*

0.0300*

  1. Note: R2 indicates proportion of variance predicted by each of the two polygenic scores in the 2 stable and 12 individual phenotypes; PGC MDD polygenic scores for Depression (from the PGC without 23&Me data); UKBB ANX polygenic scores for Anxiety (from UK Biobank); Empirical p-values were obtained from 10,000 permutations to correct for multiple testing of p-value thresholds, and ‘*’ indicates target TEDS phenotypes that were significantly predicted by polygenic scores at the more stringent empirical p-value thresholds; Supplementary Figures 5, 6 show that prediction is reliable across multiple thresholds; Supplementary Table 8 shows that similar results were obtained using summary stats for 10,000 SNPs from a larger sample including 23&Me data
Back to article page