Methoden und Algorithmen der Kopplungsanalyse bei quantitativen Phänotypen

Motivation: Krankheiten beim Menschen werden zu einem großen Teil durch geneti- sche Varianten beeinflusst oder verursacht. Um den Krankheitsmechanismus zu ver- stehen und um Patienten ursächlich behandeln zu können, ist ein erster Schritt, die genetische Variante im menschlichen Genom zu lokali Methoden um ein mächtiges Verfahren zur genetischen Kartierung quantitativer Phänotypen erweitert. Da genehunter-qmod auf dem Lander-Green-Algorithmus basiert, können viele Marker gleichzeitig in die Kopplungsanalyse einbezogen werden. Darum eignet sich genehunter-qmod gut für die Anwendung in Genkartierungs- projekten mit diallelischen SNP-Markern, die weniger informativ sind als Mikrosatelliten und daher in größerer Zahl in die Analyse eingehen müssen. genehunter-qmod ist nicht kommerzielle Software und im Internet unter http://www.helmholtz-muenchen.de/genepi/downloads frei erhältlich.
Objective: To a large degree human diseases are influenced or caused by genetic variants. In order to understand the mechanism of the disease and to treat patients in a causative way, a first step is to locate the genetic variants in the human genome. An important tool for this goal is linkage analysis. In this work, a parametric method for linkage analysis of quantitative phenotypes is presented. The method provides a test for linkage as well as an estimate of different parameters of the genotype-phenotype relation.We have implemented our new method in the program genehunter-qmod and performed simulations to compare its power and type I error to existing methods, i.e. variance components analysis (VCA) and Haseman-Elston regression. Methods: The phenotype is modeled as a normally distributed variable, with a separate distribution for each genotype. Estimates of the genotype-specific expectation values and standard deviations are obtained by maximizing the LOD score over these parameters with a gradient-based optimization called pgrad method. That way, genehunter-qmod can both locate the putative disease locus and provide specific information about the genotype-phenotype relation. Results: genehunter-qmod has lower power to detect linkage than VCA in ca- se of a normal distribution and with sib pairs. However, it outperforms VCA and Haseman-Elston regression for larger pedigrees, non-randomly ascertained data or non-normally distributed phenotypes. Here, the higher power even goes along with conservativeness, while VCA has an inflated type I error. Parameter estimation tends to underestimate residual variances, but performs better for expectation values of the genotype-specific phenotype distributions. Conclusion: With genehunter-qmod, a powerful new tool is provided to ex- plicitly model quantitative phenotypes in the context of linkage analysis. Because genehunter-qmod is based on the Lander-Green algorithm, it can simultaneously use many markers in the analysis, which makes it applicable to gene-mapping projects based on SNP arrays. The program is freely available at http://www.helmholtz-muenchen.de/genepi/downloads.