Making the Most of Case-Mother/Control-Mother Studies

Biostatistics Branch, National Institute of Environmental Health Sciences, Research Triangle Park, NC 27709, USA.
American journal of epidemiology (Impact Factor: 5.23). 09/2008; 168(5):541-7. DOI: 10.1093/aje/kwn149
Source: PubMed


The prenatal environment plays an important role in many conditions, particularly those with onset early in life, such as childhood cancers and birth defects. Because both maternal and fetal genotypes can influence risk, investigators sometimes use a case-mother/control-mother design, with mother-offspring pairs as the unit of analysis, to study genetic factors. Risk models should account for both the maternal genotype and the correlated fetal genotype to avoid confounding. The usual logistic regression analysis, however, fails to fully exploit the fact that these are mothers and offspring. Consider an autosomal, diallelic locus, which could be related to disease susceptibility either directly or through linkage with a polymorphic causal locus. Three nested levels of assumptions are often natural and plausible. The first level simply assumes Mendelian inheritance. The second further assumes parental mating symmetry for the studied locus in the source population. The third additionally assumes parental allelic exchangeability. Those assumptions imply certain nonlinear constraints; the authors enforce those constraints by using Poisson regression together with the expectation-maximization algorithm. Calculations reveal that improvements in efficiency over the usual logistic analysis can be substantial, even if only the Mendelian assumption is honored. Benefits are even more marked if, as is typical, information on genotype is missing for some individuals.

  • Source
    • "Usually, standard logistic regression is used in the case-mother and control-mother design. However such analysis is inefficient here because it does not take into account the natural family-based constraints present in the parent-child relationship (Shi, Umbach, Vermeulen, and Weinberg 2008). "

    Preview · Article · Jan 2015 · Journal of statistical software
  • Source
    • "The expectation-maximization algorithm was applied to fully utilize families with missing parental genotypes [26]. We also performed a haplotype analysis for all the six SNPs in the LNPEP gene using TRIMM [27]. The test statistics was constructed based on the vector of genotype differences between the mother and the father. "
    [Show abstract] [Hide abstract]
    ABSTRACT: Preterm birth (PTB) is a complex disorder associated with significant neonatal mortality and morbidity and long-term adverse health consequences. Multiple lines of evidence suggest that genetic factors play an important role in its etiology. This study was designed to identify genetic variation associated with PTB in oxytocin pathway genes whose role in parturition is well known. To identify common genetic variants predisposing to PTB, we genotyped 16 single nucleotide polymorphisms (SNPs) in the oxytocin (OXT), oxytocin receptor (OXTR), and leucyl/cystinyl aminopeptidase (LNPEP) genes in 651 case infants from the U.S. and one or both of their parents. In addition, we examined the role of rare genetic variation in susceptibility to PTB by conducting direct sequence analysis of OXTR in 1394 cases and 1112 controls from the U.S., Argentina, Denmark, and Finland. This study was further extended to maternal triads (maternal grandparents-mother of a case infant, N=309). We also performed in vitro analysis of selected rare OXTR missense variants to evaluate their functional importance. Maternal genetic effect analysis of the SNP genotype data revealed four SNPs in LNPEP that show significant association with prematurity. In our case--control sequence analysis, we detected fourteen coding variants in exon 3 of OXTR, all but four of which were found in cases only. Of the fourteen variants, three were previously unreported novel rare variants. When the sequence data from the maternal triads were analyzed using the transmission disequilibrium test, two common missense SNPs (rs4686302 and rs237902) in OXTR showed suggestive association for three gestational age subgroups. In vitro functional assays showed a significant difference in ligand binding between wild-type and two mutant receptors. Our study suggests an association between maternal common polymorphisms in LNPEP and susceptibility to PTB. Maternal OXTR missense SNPs rs4686302 and rs237902 may have gestational age-dependent effects on prematurity. Most of the OXTR rare variants identified do not appear to significantly contribute to the risk of PTB, but those shown to affect receptor function in our in vitro study warrant further investigation. Future studies with larger sample sizes are needed to confirm the findings of this study.
    Full-text · Article · Jul 2013 · BMC Medical Genetics
  • Source
    • "In contrast, CLL, which assumes the disease being rare, has much larger biases, even when the disease is indeed rare, since rare disease is a necessary but not a sufficient condition for CLL to be valid. Finally, mating symmetry is commonly assumed for many imprinting and/or maternal effects detection methods [Ainsworth et al. (2011), Shi et al. (2008), Weinberg, Wilcox and Lie (1998), Weinberg (1999), Zhou et al. (2009)]. However, when this assumption is violated, there can be large biases and greatly inflated type I error rates, whereas LIME is not affected at all by departure from such assumptions (Figures 1 and 2). "
    [Show abstract] [Hide abstract]
    ABSTRACT: Genomic imprinting and maternal effects are two epigenetic factors that have been increasingly explored for their roles in the etiology of complex diseases. This is part of a concerted effort to find the "missing heritability." Accordingly, statistical methods have been proposed to detect imprinting and maternal effects simultaneously based on either a case-parent triads design or a case-mother/control-mother pairs design. However, existing methods are full-likelihood based and have to make strong assumptions concerning mating type probabilities (nuisance parameters) to avoid overparametrization. In this paper we propose to augment the two popular study designs by combining them and including control-parent triads, so that our sample may contain a mixture of case-parent/control-parent triads and case-mother/control-mother pairs. By matching the case families with control families of the same structure and stratifying according to the familial genotypes, we are able to derive a partial likelihood that is free of the nuisance parameters. This renders unnecessary any unrealistic assumptions and leads to a robust procedure without sacrificing power. Our simulation study demonstrates that our partial likelihood method has correct type I error rate, little bias and reasonable power under a variety of settings.
    Preview · Article · Apr 2013 · The Annals of Applied Statistics
Show more