Development and implementation of linkage-analysis methods for the genetic mapping of complex traits
Final Report Abstract
The aim of this DFG-funded project was to further develop, implement, and assess methods of linkage analysis for the genetic mapping of complex traits. We had previously developed the GENEHUNTER-MODSCORE software, the first implementation of MOD-score analysis, in which the parametric LOD score is maximized with respect to the parameters of the trait model, i.e., the penetrances and disease allele frequency. A MOD-score analysis is especially useful if these parameters cannot be specified prior to the analysis, which is the case for most complex diseases. In this project, we have performed an extensive simulation study to compare the power to detect linkage of the MOD-score approach with various other parametric and nonparametric linkage methods. These simulations have shown that MOD- score analysis is indeed more powerful than other methods for larger pedigrees, or for mixtures of pedigrees of various types, which represents a common situation in practice. Because of the high computational demands of a MOD-score analysis, we have developed a completely new algorithm for the calculation of the disease-locus likelihood, which is the most computationally intensive step. Our new algorithm uses an algebraic representation of the disease-locus likelihood and collapses inheritance vectors, i.e., patterns of allele-sharing within a pedigree, into classes with identical likelihood contribution. The speed-up of our new algorithm, implemented in a new version of GENEHUNTER-MODSCORE, ranges from 1.9 to 11.5, with higher speed-ups obtained for larger pedigrees. Using the new software, we have undertaken a simulation study to investigate to what degree and accuracy the parameters of the disease model can be estimated by the MOD-score approach. In a further simulation study, we have investigated how the confounding between genomic imprinting and sex differences in recombination fractions can best be controlled when testing for imprinting in a linkage analysis, in order to avoid an increased type I error rate. Our results show that assuming sex-averaged recombination fractions in the analysis leads to a much more powerful imprinting test than assuming the true sex-specific values, provided that an appropriate adjustment of the critical value is applied to control the type I error rate. Also the effect of the confounding can be reduced by performing multi-marker analysis. In addition, we have developed a new method of parametric MOD-score analysis for quantitative traits and implemented it in the GENEHUNTER-QMOD software. The phenotype is modeled by a likelihood that assumes genotype-specific normal distributions. The method has higher power than classical methods, i.e., variance components analysis or Haseman- Elston regression, (i) for larger sibships, (ii) for samples that have been selected for certain phenotypes or (iii) if the phenotypes do not (!) follow a normal distribution. Finally, based on our previous implementation of linkage analysis with two trait loci, we have devised an extension that allows researchers to perform a two-trait-locus linkage analysis in a genome-wide fashion, and implemented it in a new version of GENEHUNTER- TWOLOCUS. It is able to perform a two-trait-locus analysis systematically for all pairwise combinations of chromosomes at almost no additional computational cost, compared to an analysis of only a single pair of chromosomes with one trait locus modeled on each chromosome. In addition, our new implementation can perform an analysis with two trait loci on the same chromosome, which has not been possible with the previous version. We have applied our new methodological developments to datasets for several complex diseases and were able to map loci that are likely to harbor genes responsible for bipolar disorder, congenital heart defects, and house dust mite allergy.
Publications
- A New Susceptibility Locus for Bipolar Affective Disorder in PAR1 on Xp22.3/Yp11.3. American Journal of Medical Genetics Part B 2010;153B:1110-1114
Antònia Flaquer, Rami Abou Jamra, Karolin Etterer, Guillermo Orozco Díaz, Fabio Rivas, Marcella Rietschel, Sven Cichon, Markus M. Nöthen, Konstantin Strauch
- A comparison of different linkage statistics in small to moderate sized pedigrees with complex diseases. BMC Research Notes 2012;5:11
Antònia Flaquer und Konstantin Strauch
- Parameter estimation and quantitative parametric linkage analysis with GENEHUNTER-QMOD. Human Heredity 2012;73:208-219
Thomas Künzel und Konstantin Strauch
- Genome-wide linkage analysis of congenital heart defects using MOD score analysis identifies two novel loci. BMC Genetics 2013;14:44
Antònia Flaquer, Clemens Baumbach, Estefania Piñero, Fernando García Algas, María Angeles de la Fuente Sanchez, Jordi Rosell, Jorge Toquero, Luis Alonso-Pulpon, Pablo Garcia-Pavia, Konstantin Strauch, Damian Heine-Suñer
(See online at https://doi.org/10.1186/1471-2156-14-44) - Fast linkage analysis with MOD scores using algebraic calculation. Human Heredity 2014;78:179-194
Markus Brugger und Konstantin Strauch
(See online at https://doi.org/10.1159/000369065)
