Interpretable machine learning classifiers implicate GPC6 in Parkinson's disease from single-nuclei midbrain transcriptomes
Bibliographic record
Abstract
Parkinson's disease (PD) is a progressive and devastating neurodegenerative disease. An incomplete understanding of its genetic architecture remains a major barrier to the clinical translation of targeted therapeutics, necessitating novel approaches to uncover elusive genetic determinants. Single-cell and single-nuclear RNA sequencing (scnRNAseq) can help bridge this gap by profiling individual cells for disease-associated differential gene expression and nominating genes for targeted genomic analyses. Here, we introduce a machine learning framework to identify molecular features that characterize post-mortem brain cells from PD patients. We train classifiers to distinguish between PD and healthy cells, then decode the models to unravel the 'reasons' behind the classifications, revealing key genes expression signatures that characterize cells from the parkinsonian brain. Application of this framework to three publicly available snRNAseq datasets characterizing the post-mortem midbrain identified cell-type-specific gene sets that accurately classify PD cells across all datasets, demonstrating our approach's capacity to identify robust molecular markers of disease. Targeted genomic analyses of the key genes characterizing PD cells revealed a previously undescribed association between PD and rare variants in GPC6, a member of the heparan sulfate proteoglycan family, which have been implicated in the intracellular accumulation of alpha-synuclein preformed fibrils. We replicate this association in three separate case-control cohorts. Our method promises to enhance understanding of the genetic architecture in complex diseases like PD, representing a critical step toward targeted therapeutics. Our publicly available framework is readily applicable across diseases.
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 | 0.001 | 0.001 |
| 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 | 0.001 | 0.000 |
| Open science | 0.000 | 0.001 |
| Research integrity | 0.001 | 0.001 |
| Insufficient payload (model declined to judge) | 0.001 | 0.001 |
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".