Assessing models for genetic prediction of complex traits: a comparison of visualization and quantitative methods

BMC Genomics

Table 3 Descriptive statistics for the various genetic prediction models from Gagliano et al. (2014) to be used as examples here

Phenotype-specific analyses		N	Minimum	25 % percentile	Median	Mean	75 % percentile	Maximum	Standard deviation	N outliers*
Brain-related	Hits	144	0.40	0.42	0.51	0.51	0.57	0.77	0.09	3
	Non-hits	32723	0.40	0.40	0.46	0.48	0.53	0.79	0.07	61
Autoimmune	Hits	234	0.29	0.45	0.55	0.55	0.66	0.86	0.14	0
	Non-hits	33266	0.29	0.30	0.44	0.45	0.55	0.93	0.13	0
All phenotype analyses
p < 5E-8	Hits	1292	0.32	0.44	0.54	0.54	0.62	0.92	0.13	4
	Non-hits	30135	0.32	0.35	0.44	0.46	0.55	0.91	0.12	7
all GWAS Catalogue	Hits	3405	0.44	0.45	0.50	0.51	0.54	0.81	0.06	144
	Non-hits	30039	0.44	0.44	0.48	0.49	0.52	0.80	0.05	336

*Outliers are defined as data points outside 1.5x interquartile range (interquartile range = 75 % percentile - 25 % percentile)

ISSN: 1471-2164