HLA typing of WES data from Alzheimer's disease cases and healthy controls

We have recently undertaken association testing between HLA haplotype and AD risk using higher resolution whole-exome sequencing (WES) data, a warranted approach given how similar HLA alleles are to one another. Previous studies have fine-mapped the HLA region in AD but this is incredibly difficult in regions of high linkage disequilibrium and requires large sample sets. We have utilized a dataset of >2200 AD cases and controls with available WES data and employed an algorithm to bioinformatically HLA type our dataset validated against PCR based HLA typing service for 11 main HLA genes. We now wish to validate our findings in a replication cohort.
We will HLA type the ADSP discovery cohort of 5,096 AD cases and 4,965 controls to further explore the HLA association with AD.
We will use the algorithm HLA-HD (PMID: 28419628) to HLA type all individuals.
Using sex as a confounding factor we will test for an association between HLA type and AD using logistic regression modelling, testing for statistical significance using permutation testing.

Alzheimer’s disease (AD) is the most common form of dementia, and it is known that the immune system plays a major role in the development of disease. An individual's risk for developing AD is partially determined by changes in their genes. Recent studies comparing the genetic information from healthy individuals and those with AD have identified that small changes in genes alter risk of developing disease. The genes encoding the Human Leukocyte Antigens (HLA) are some of these disease influencing genes. The HLA genes encode for proteins that tell the difference between your own cells and foreign cells. HLA genes are very difficult to study as they come in many different forms. In the brain HLA genes are mainly found on microglia (the immune cells of the brain). We have already used a dataset of >2000 participants living with AD and elderly well controls and identified novel changes in HLA genes that alter risk of developing AD. We would now like to replicate our findings in another dataset and request access to the ADSP discovery cohort.