The performance of AlphaMissense to identify genes influencing disease
Bibliographic record
Abstract
A novel algorithm, AlphaMissense, has been shown to have an improved ability to predict the pathogenicity of rare missense genetic variants. However, it is not known whether AlphaMissense improves the ability of gene-based testing to identify disease-influencing genes. Using whole-exome sequencing data from the UK Biobank, we compared gene-based association analysis strategies including sets of deleterious variants: predicted loss-of-function (pLoF) variants only, pLoF plus AlphaMissense pathogenic variants, pLoF with missense variants predicted to be deleterious by any of five commonly utilized annotation methods (Missense (1/5)) or only variants predicted to be deleterious by all five methods (Missense (5/5)). We measured performance to identify 519 previously identified positive control genes, which can lead to Mendelian diseases, or are the targets of successfully developed medicines. These strategies identified 0.85 million pLoF variants and 5 million deleterious missense variants, including 22,131 likely pathogenic missense variants identified exclusively by AlphaMissense. The gene-based association tests found 608 significant gene associations (at p < 1.25 × 10 −7 ) across 24 common traits and diseases. Compared with pLoFs plus Missense (5/5), tests using pLoFs and AlphaMissense variants found slightly more significant gene-disease and gene-trait associations, albeit with a marginally lower proportion of positive control genes. Nevertheless, their overall performance was similar. Merging AlphaMissense with Missense (5/5), whether through their intersection or union, did not yield any further enhancement in performance. In summary, employing AlphaMissense to select deleterious variants for gene-based testing did not improve the ability to identify genes that are known to influence disease.
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 distilled prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. Learned from the 10,348 direct Codex labels and 10,348 direct Gemma labels. Candidate is the union of thresholded teacher heads; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels or direct frontier model labels.
Codex and Gemma teacher scores by category
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.000 | 0.000 |
| Meta-epidemiology (narrow) | 0.000 | 0.000 |
| Meta-epidemiology (broad) | 0.000 | 0.000 |
| Bibliometrics | 0.000 | 0.000 |
| Science and technology studies | 0.000 | 0.000 |
| Scholarly communication | 0.000 | 0.000 |
| Open science | 0.000 | 0.000 |
| Research integrity | 0.000 | 0.000 |
| Insufficient payload (model declined to judge) | 0.000 | 0.000 |
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 teacher head, 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".