Integrating polygenic risk scores, modifiable dementia risk factors, and sex as predictors of cohort membership in the Alzheimer’s and Lewy body disease spectra: A COMPASS‐ND Study
Bibliographic record
Abstract
BACKGROUND: Polygenic risk scores (PRSs) summarize genetic risk across single nucleotide polymorphisms (SNPs). Alzheimer's disease (AD) and Lewy body disease (LBD) PRSs contribute to risk detection via both multi-mechanism (SNPs from multiple mechanistic pathways) and mechanism-specific (SNPs from separate mechanistic pathways) versions. Machine learning techniques compared the relative importance of 14 PRSs and 14 AD/LBD risk factors in discriminating cohorts of AD and LBD spectra, disaggregated by sex. METHODS: Data from the Comprehensive Assessment of Neurodegeneration and Dementia database of the Canadian Consortium on Neurodegeneration in Aging included 7 cohorts (n=899; M age=72; 49%F). The common benchmark was the cognitively unimpaired (CU) cohort. AD spectrum cohorts: subjective cognitive impairment (SCI), mild cognitive impairment (MCI), and AD. LBD spectrum cohorts: Parkinson's disease (PD), PD-MCI, and LBD. PRSice-2 calculated two multi-mechanism PRSs (AD-related, 387 SNPs; LBD-related, 379 SNPs) and 12 mechanism-specific PRSs (5-17 SNPs/mechanism). AD-specific PRSs: amyloid-beta metabolism, tau metabolism, lipid metabolism, mitochondrial function, immune response, nervous function, and basic cellular processes. LBD-specific PRSs: alpha-synuclein metabolism, lysosomal degradation, ceramide metabolism, dopamine metabolism, and oxidative stress response. We assessed APOE+/- PRSs. All comparisons included 14 risk factors: demographic (e.g., age), mobility (e.g., gait), lifestyle (e.g., physical activity), biomarkers (e.g., homocysteine), vascular (e.g., pulse pressure), and functional (grip strength). Random forest classifier models tested predictions across spectra, disaggregated by sex. SELECTED RESULTS: Overall, we observed in each spectrum (a) strong discrimination across cohorts, (b) varying leading predictors of same-spectra cohorts, and (c) different prediction patterns by sex. Example for AD spectrum: four leading predictors of SCI/AD in females were multi-mechanism AD-related (APOE) PRS, gait, age, and education (AUC=.81), whereas four leading predictors in males were amyloid-beta metabolism (APOE) PRS, lipid metabolism (APOE) PRS, LDL cholesterol, and vitamin B12 (AUC=.92). Example for LBD spectrum: four leading predictors of PD-MCI/LBD in females were multi-mechanism LBD-related PRS, ceramide metabolism PRS, education, and homocysteine (AUC=.77), whereas five leading predictors in males were multi-mechanism LBD-related (APOE) PRS, lysosomal degradation PRS, age, nutrition, and total bilirubin (AUC=.84). CONCLUSION: Integration of neurodegenerative disease-related PRSs and risk factors promotes precision identification of sex-specific cohort discrimination in the AD and LBD spectra.
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 | 0.007 | 0.008 |
| Meta-epidemiology (narrow) | 0.001 | 0.001 |
| Meta-epidemiology (broad) | 0.001 | 0.003 |
| Bibliometrics | 0.001 | 0.001 |
| Science and technology studies | 0.001 | 0.001 |
| Scholarly communication | 0.001 | 0.001 |
| Open science | 0.001 | 0.001 |
| Research integrity | 0.001 | 0.001 |
| Insufficient payload (model declined to judge) | 0.002 | 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 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".