Table 4 The impact of misdiagnosis in estimating genetic parameter from genome-wide genotypes

From: Impact of diagnostic misclassification on estimation of genetic correlations using genome-wide genotypes

       

Proportion of variance explained by SNPs

 

M TA

M TB

‘True’ (T) or diagnosed (D) disorders

Estimated from data or calculated a

σ uA 2

σ uB 2

σ 2 uA,uB

Disorder A

Disorder B

r g

 0

0

T

Estimated

0.096

0.112

0.002

0.391 (0.089)

0.470 (0.093)

0.023 (0.155)

0.1

0

D

Estimated

0.096

0.092

0.013

0.387 (0.029)

0.393 (0.023)

0.139 (0.055)

  

D

Calculated

0.096

0.095

0.010

0.385

0.396

0.109

0.1

0.1

D

Estimated

0.075

0.093

0.024

0.304 (0.034)

0.388 (0.035)

0.296 (0.092)

  

D

Calculated

0.079

0.093

0.021

0.316

0.383

0.244

  1. Calculated using equations in text based on the estimates from the true disorders and misclassification rates.
  2. MTA proportion of disorder A cases labelled as disorder B cases; MTB proportion of disorder B cases labelled as disorder A cases. Values in parentheses are the standard errors for the parameters estimated when MTA=MTB=0, but otherwise are the standard deviations over 100 replicates.