Table 2. Comparison of s-LDSC and i-LDSC estimates of phenotypic variance explained (PVE) by genetic effects for 25 complex traits in the UK Biobank.

Here, we use stratified LD score regression (s-LDSC) to partition heritability across different genomic elements (Finucane et al., 2015). We used 97 functional annotations from Gazal et al. to estimate heritability in 25 traits. We then appended cis-interaction LD scores as an additional annotation to obtain heritability estimates (this method is referred to as s+i-LDSC in the table). p-values for the s+i-LDSC model detailing the contributions of tagged non-additive genetic effects for 25 traits are provided in the last column. Note that, while not statistically significant in this stratified analysis with the additional annotations, the non-additive component still makes nonzero contributions to the PVE estimation for all 25 traits.

Trait	UKB PVE (s-LDSC)	UKB PVE (s+i-LDSC)	s+i-LDSC p-value
Basophil	0.0363	0.0375	0.4728
BMI	0.2100	0.2482	0.8126
Cholesterol	0.1042	0.1358	0.6202
CRP	0.0452	0.0524	0.6483
DBP	0.1228	0.1441	0.6125
EGFR	0.1826	0.2105	0.8507
Eosinophil	0.1403	0.1578	0.1867
HBA1C	0.1040	0.1275	0.6917
HDL	0.1820	0.2373	0.5754
Height	0.4315	0.4726	0.5224
Hematocrit	0.1416	0.1646	0.3956
Hemoglobin	0.1504	0.1795	0.2299
LDL	0.0858	0.1131	0.8812
Lymphocyte	0.0545	0.0651	0.1453
MCH	0.1497	0.1545	0.0968
MCHC	0.0450	0.0496	0.3728
MCV	0.1814	0.1930	0.1530
Monocyte	0.1085	0.1431	0.5421
Neutrophil	0.1320	0.1599	0.2499
Platelet	0.2317	0.2628	0.7371
RBC	0.1933	0.2223	0.3197
SBP	0.1206	0.1419	0.1100
Triglycerides	0.1335	0.1621	0.5301
Urate	0.1530	0.1736	0.1177
WBC	0.1221	0.1482	0.5155