Search bioRxiv⌕ Search

Biology subjects

Kuechenhoff, H.

Publications and source records attributed to Kuechenhoff, H..

2 recordsLinked to original sources
bioRxiv2019

Genome-wide association meta-analysis for early age-related macular degeneration highlights novel loci and insights for advanced disease

BackgroundAdvanced age-related macular degeneration (AMD) is a leading cause of blindness. While around half of the genetic contribution to advanced AMD has been uncovered, little is known about the genetic architecture of the preceding early stages of the diseases. MethodsTo identify genetic factors for early AMD, we conducted a genome-wide association meta-analysis with 14,034 early AMD cases and 91,214 controls from 11 sources of data including data from the International AMD Genomics Consortium (IAMDGC) and the UK Biobank (UKBB). We ascertained early AMD via color fundus photographs by manual grading for 10 sources and by using an automated machine learning approach for >170,000 images from UKBB. We searched for significant genetic loci in a genome-wide association screen (P<5x10-8) based on the meta-analysis of the 11 sources and via a candidate approach based on 13 suggestive early AMD variants from Holliday et al 2013 (P<0.05/13, additional 3,432 early AMD cases and 11,235 controls). For the novel AMD regions, we conducted in-silico follow-up analysis to prioritize causal genes and pathway analyses. ResultsWe identified 11 loci for early AMD, 9 novel and 2 known for early AMD. Most of these 11 loci overlapped with known advanced AMD loci (near ARMS2/HTRA1, CFH, APOE, C2, C3, CETP, PVRL2, TNFRSF10A, VEGFA), except two that were completely novel to any AMD. Among the 17 genes within the two novel loci, in-silico functional annotation suggested CD46 and TYR as the most likely responsible genes. We found the presence or absence of an early AMD effect to distinguish known pathways of advanced AMD genetics (complement/lipid pathways or extracellular matrix metabolism, respectively). ConclusionsOur data on early AMD genetics provides a resource comparable to the existing data on advanced AMD genetics, which enables a joint view. Our large GWAS on early AMD identified novel loci, highlighted shared and distinct genetics between early and advanced AMD and provides insights into AMD etiology. The ability of early AMD effects to differentiate the major pathways for advanced AMD underscores the biological relevance of a joint view on early and advanced AMD genetics.

genetics↗
bioRxiv2019

Chances and challenges of machine learning based disease classification in genetic association studies illustrated on age-related macular degeneration

Imaging technology and machine learning algorithms for disease classification set the stage for high-throughput phenotyping and promising new avenues for genome-wide association studies (GWAS). Despite emerging algorithms, there has been no successful application in GWAS so far. We established machine learning based disease classification in genetic association analysis as a misclassification problem. To evaluate chances and challenges, we performed a GWAS based on automated classification of age-related macular degeneration (AMD) in UK Biobank (images from 135,500 eyes; 68,400 persons). We quantified misclassification of automatically derived AMD in internal validation data (images from 4,001 eyes; 2,013 persons) and developed a maximum likelihood approach (MLA) to account for it when estimating genetic association. We demonstrate that our MLA guards against bias and artefacts in simulation studies. By combining a GWAS on automatically derived AMD classification and our MLA in UK Biobank data, we were able to dissect true association (ARMS2/HTRA1, CFH) from artefacts (near HERC2) and to identify eye color as relevant source of misclassification. On this example of AMD, we are able to provide a proof-of-concept that a GWAS using machine learning derived disease classification yields relevant results and that misclassification needs to be considered in the analysis. These findings generalize to other phenotypes and also emphasize the utility of genetic data for understanding misclassification structure of machine learning algorithms.

genetics↗