Skip to content

Univariate MiXeR

Using software provided by the authors, I ran univariate MiXeR1 on the Johnston et al multsite pain summary statistics2.

As recommended by the MiXeR authors, I ran MiXeR 20 times using 20 different subsets of the reference panel of genetic variants. This serves as a form of bootstrapping.

The table below lists the key MiXeR output parameters:

Parameter Value
pi (mean) 0.003899
pi (std) 0.0001094
sig2_beta (mean) 8.855e-06
sig2_beta (std) 2.317e-07
sig2_zero (mean) 1.128
sig2_zero (std) 0.003001
h2 (mean) 0.07158
h2 (std) 0.0008682
nc@p9 (mean) 12430
nc@p9 (std) 349
AIC 84.84
BIC 94.47

A polygenicity score of \(\pi \approx 0.004\) indicates a very polygenic trait. A discoverability score of \(\sigma^2_\beta \approx 8.85\times 10^{-6}\) indicates very small individual genetic effects. Compared to ME/CFS, multisite pain appears to be determined by more genetic variants, each of which has a smaller genetic effect.

As befits a polygenic trait with weak effects, a very large sample size would be required to explain a significant proportion of SNP heritability. This is illustrated by the power plot:

pain_mixer_power

The QQ-plot suggests a good fit between the Mixer model and the multi-site pain GWAS data:

pain_qq

  1. Dominic Holland, Oleksandr Frei, Rahul Desikan, Chun-Chieh Fan, Alexey A Shadrin, Olav B Smeland, Vijay S Sundar, Paul Thompson, Ole A Andreassen, and Anders M Dale. Beyond snp heritability: polygenicity and discoverability of phenotypes estimated with a univariate gaussian mixture model. PLoS Genetics, 16(5):e1008612, 2020. URL: https://journals.plos.org/plosgenetics/article?id=10.1371/journal.pgen.1008612. ↩

  2. Keira JA Johnston, Mark J Adams, Barbara I Nicholl, Joey Ward, Rona J Strawbridge, Amy Ferguson, Andrew M McIntosh, Mark ES Bailey, and Daniel J Smith. Genome-wide association study of multisite chronic pain in UK Biobank. PLoS Genetics, 15(6):e1008164, 2019. URL: https://journals.plos.org/plosgenetics/article?id=10.1371/journal.pgen.1008164. ↩