Schaid, Daniel J.

Abstract Linkage disequilibrium, the nonrandom association of alleles from different loci, can provide valuable information on the structure of haplotypes in the human genome and is often the basis for evaluating the association of genomic variation with human traits among unrelated subjects. But, linkage phase of genetic markers measured on unrelated subjects is typically unknown, and so measurement of linkage disequilibrium, and testing whether it differs significantly from the null value of zero, requires statistical methods that can account for the ambiguity of unobserved haplotypes. A common method to test whether linkage disequilibrium differs significantly from zero is the likelihood-ratio statistic, which assumes Hardy-Weinberg equilibrium of the marker phenotype proportions. We show, by simulations, that this approach can be grossly biased, with either extremely conservative or liberal type I error rates. In contrast, we use simulations to show that a composite statistic, proposed by Weir and Cockerham, maintains the correct type I error rates, and, when comparisons are appropriate, has similar power as the likelihood-ratio statistic. We extend the composite statistic to allow for more than two alleles per locus, providing a global composite statistic, which is a strong competitor to the usual likelihood-ratio statistic.

Rheumatoid arthritis (RA) is a heterogeneous syndrome of which a subset of patients develops vascular inflammation. The genetic determinants that confer risk for rheumatoid vasculitis are not known, but patients with vascular complications are known to have an expansion of CD4+CD28null T cells, a cell population potentially involved in endothelial damage. CD4+CD28null T cell clones isolated from RA patients with vasculitis were found to express killer cell immunoglobulin–like receptors (KIRs) with the stimulatory KIR2DS2 often present in the absence of opposing inhibitory receptors with related specificities. To test the hypothesis that the KIR2DS2 gene is involved in the development of vasculitis, association studies were performed. The KIR2DS2 gene was significantly enriched among patients with rheumatoid vasculitis compared with normal individuals (odds ratio 5.56, P = 0.001) and patients with RA but no vasculitis (odds ratio 7.96, P = 0.001). Also, the distribution of human histocompatibility leukocyte antigen (HLA)-C, the putative ligand for KIRs, was significantly different in patients with rheumatoid vasculitis in comparison with the control populations. These data suggest that HLA class I–recognizing receptors and HLA class I genes are genetic risk determinants that modulate the pattern of RA expression. Specifically, KIR2DS2 in conjunction with the appropriate HLA-C ligand may have a role in vascular damage by regulating CD4+CD28null T cells.

Human cancer cells with a mutated p53 tumor-suppressor gene have a selective growth advantage and may exhibit resistance to ionizing radiation and certain chemotherapeutic agents. To examine the prognostic value of mutations in the p53 gene, a cohort of 90 Midwestern Caucasian breast cancer patients were analyzed with methodology that detects virtually 100% of all mutations. The presence of a p53 gene mutation was by far the single most predictive indicator for recurrence and death (relative risks of 4.7 and 23.2, respectively). Direct detection of p53 mutations had substantially greater prognostic value than immunohistochemical detection of p53 overexpression. Analysis of p53 gene mutations may permit identification of a subset of breast cancer patients who, despite lack of conventional indicators of poor prognosis, are at high risk of early recurrence and death.