An accurate genetic colocalization method for the HLA locus
Bibliographic record
Abstract
Abstract Genetic colocalization analyses are frequently conducted to determine if causal signals at a genetic locus are shared between two phenotypes. However, colocalization is rarely undertaken at the HLA locus, due to its complex linkage disequilibrium (LD) and high polymorphism density. This lack of genetic causal inference method limits our ability to translate HLA associations into therapeutic targets. Here we present a method that uses HLA alleles, instead of nucleotide variants, to perform genetic colocalization of two traits at HLA genes. The method, which we call HLA-colocalization, works by controlling for LD using a Bayesian variable selection algorithm (here implemented with SuSiE), then performing Bayesian regression on the resulting posterior inclusion probabilities. We first show through simulation that the method correctly identifies truly colocalizing genes. We then test the method in two positive control scenarios, showing colocalization between hepatitis B and liver disease at HLA-DPB1 , and between Epstein-Barr virus and multiple sclerosis at HLA-DRB1 and HLA-DQB1 . Lastly, we perform a large colocalization scan between multiple viruses and auto-immune diseases, demonstrating that the method is well calibrated, and uncovering multiple biologically plausible novel causal associations, such as cytomegalovirus and ulcerative colitis. To our knowledge, HLA-colocalization is the first accurate genetic colocalization method for the HLA locus (github: https://github.com/DrGBL/hlacoloc ).
Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.
How this classification was reachedexpand
Full frame machine prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. The Gemma side is a direct model label for every work in the frame, read from the title-only record. The Codex side is a classifier learned from the 10,348 direct Codex labels and calibrated to design-weighted sample rates; fields without enough sample support carry no Codex call. Candidate is the union of the two sides; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels.
Distilled classifier scores by category (both heads)
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.003 | 0.010 |
| Meta-epidemiology (narrow) | 0.001 | 0.000 |
| Meta-epidemiology (broad) | 0.001 | 0.001 |
| Bibliometrics | 0.002 | 0.001 |
| Science and technology studies | 0.001 | 0.001 |
| Scholarly communication | 0.001 | 0.001 |
| Open science | 0.001 | 0.002 |
| Research integrity | 0.001 | 0.001 |
| Insufficient payload (model declined to judge) | 0.006 | 0.002 |
Machine scores (provisional)
The two teacher heads of the student model, read on this work. A score orders the frame for review; it never asserts a category, and the validation status ships verbatim with every row.
Baseline scores from an immature model (maturity gate not passed, 7 training rounds). Scores rank; they never assert a category.
score_only:v0-immature-baseline · verbatim from the scoring run: score_only means the number may rank works, and no category label ships from itClassification
machine, unvalidatedMachine predicted; a candidate call from one source (direct Gemma or distilled Codex), not a consensus.
How this classification was reached, model by model and score by score, is at the end of the page under "How this classification was reached".