Analytical methods for disease association studies with immunogenetic data

Abstract

Disease association studies involving highly polymorphic immunogenetic data may involve analyses at one or many units of analysis, including amino acid, allele, genotype and haplotype levels, as well as consideration of gene-gene or gene-environment interactions. The selection of the appropriate statistical tests is critical and will be dependent on the nature of the dataset (e.g., case-control vs. family data) as well as the specific research hypotheses being tested. This paper describes the various study and analysis categories used for such analyses, including the advantages and limitations of such techniques.

Fig. 1

The four family-based ascertainment schemes are selected in the presence of at least: (a) one affected child (S), (b) one affected child and one affected parent (multiplex parent child (MPC)), (c) two affected sibs (multiplex sib pairs (MSP)), and (d) two affected sibs and one affected parent (multiplex parent sibs (MPS)). As noted in the text, additional family members (parent or sib) may be affected within each of these schemes; they occur with frequencies expected under the disease model and are retained in analyses. For the MSP and MPS ascertainment schemes, using the ordered notation of Thomson (55) the first affected sib ascertained is always denoted by genotype AC, with the four possible genotypes for the second affected sib listed.