This indicates that assessing the sensitivity of results of GWAS (or GWAS secondary analyses, especially those based on genome-wide statistics) on clinically-diagnosed cases will become increasingly important as the proportion of proxy and biobank cases included in the GWAS increases.

Perhaps 😉
I am sure that this controversy over APOE4 did not last long. However, it's true that he insisted, despite the odds being stacked against him, that it was TOMM40, not APOE, that was ultimately responsible for the APOE locus signal.

Why do you say that the research community was sceptical? In fact, it was one of the least controversial genetic results obtained through a gene candidate approach. It was very quickly accepted by the research community because it was systematically replicated.

Cette université est trop occupée à sauver la Science en accueillant des chercheurs américains

5. The PGS/PRS associations mainly capture genetic information related to AD because they weakened as the diagnosis was broadened. The quality of the clinical diagnosis can interfere with the measurement of the association between the PGS/PRS and the AD risk in a given population.

4. However, this is not the case when the PRS includes the APOE region. This indicates that the APOE region appears to contain additional multi-ancestry genetic variability.

3. In contrast to other multifactorial diseases, a cross-ancestry polygenic risk score (PRS) did not systematically outperform the simple PGS when the APOE locus was excluded. A high proportion of AD genetic risk could be already accounted for by the European-ancestry GWAS-defined loci.

2. This simple PGS appears to be enough to detect an AD genetic risk in most ancestry populations, suggesting that most of the various ancestry populations are likely to be affected by shared pathophysiological processes that are driven in part by genetic risk factors.

Our work produced several important findings:

1. The associations between a simple polygenic score (PGS) based on the European GWAS-defined loci and AD risk in European populations is slightly influenced by the APOE genotype, suggesting existence of independent genetic entities for sporadic AD

This work would not have been possible without this global and strongly collaborative effort (particularly Richard Sherva and Mark Logue of the Million Veteran Programme in the US, and Yoontae Kim and Jungsoo Gim of the GARD study in South Korea).