Better estimation of SNP heritability from summary statistics

AIAS Fellow Doug Speed, together with David Balding, University of Melbourne, have created SumHer, a new software for analyzing results from genetic association studies. The results are published this week in ‘Nature Genetics’.

2018.12.06 | AIAS

The two main aims of SumHer are to estimate confounding bias (measure how well an association study was performed) and to estimate SNP heritability (calculate how much of the variation in a trait is due to common genetic variants).

By applying SumHer to results for a wide variety of traits, Speed and Balding have shown that previous studies have tended to over-correct for confounding, causing them to find about 25% less causal variants that they would have with more appropriate quality control. They have also found that for each trait, common variants explain at least twice as much heritability as previously reported. The results from SumHer enable us to more efficiently analyze future data. SumHer is made freely available through the software package LDAK (

Find the article 'SumHer better estimates the SNP heritability of complex traits from summary statistics' in Nature Genetics, 3 December 2018:


Doug Speed, AIAS Fellow, Postdoc
Phone:  +45 8715 3262

Aarhus Institute of Advanced Studies (AIAS) and 
Bioinformatics Research Centre (BiRC)
Aarhus University 

Research news