Supplementary Material for: Human Birth Weight and Reproductive Immunology: Testing for Interactions between Maternal and Offspring KIR and HLA-C Genes

<p><b><i>Background/Aims:</i></b> Maternal and offspring cell contact at the site of placentation presents a plausible setting for maternal-fetal genotype (MFG) interactions affecting fetal growth. We test hypotheses regarding killer cell immunoglobulin-like receptor <i>(KIR)</i> and <i>HLA-C</i> MFG effects on human birth weight by extending the quantitative MFG (QMFG) test. <b><i>Methods:</i></b> Until recently, association testing for MFG interactions had limited applications. To improve the ability to test for these interactions, we developed the extended QMFG test, a linear mixed-effect model that can use multi-locus genotype data from families. <b><i>Results:</i></b> We demonstrate the extended QMFG test's statistical properties. We also show that if an offspring-only model is fit when MFG effects exist, associations can be missed or misattributed. Furthermore, imprecisely modeling the effects of both <i>KIR</i> and <i>HLA-C</i> could result in a failure to replicate if these loci's allele frequencies differ among populations. To further illustrate the extended QMFG test's advantages, we apply the extended QMFG test to a UK cohort study and the Norwegian Mother and Child Cohort (MoBa) study. <b><i>Conclusion:</i></b> We find a significant <i>KIR</i>-<i>HLA-C</i> interaction effect on birth weight. More generally, the QMFG test can detect genetic associations that may be missed by standard genome-wide association studies for quantitative traits.</p>