Integrating Novel Metabolomics Biomarkers with Multi‐modal Risk Factors in Predicting Mild Cognitive Impairment and Alzheimer's Disease
Bibliographic record
Abstract
BACKGROUND: Recent research approaches to an expanding collection of well-characterized databases have examined varying arrays of biomarkers and risk factors related to Alzheimer's disease (AD). Analytical techniques that simultaneously evaluate relative importance of numerous candidate factors can successfully identify leading predictors within multiple-variable, high-dimensional datasets. Our objective was to integrate new molecular biomarkers with established AD-related biomarkers and risk factors to evaluate their relative predictive contributions using a series of machine learning classifier models. METHOD: We assembled 92 AD-related indicators representing 10 key modalities of AD risk (demographic, biomarkers, lifestyle, sensory, vascular, imaging, functional, psychiatric, anthropometric, clinical health) for participants from three cohorts in the Canadian Consortium on Neurodegeneration in Aging database (aka COMPASS-ND): (1) Cognitively Unimpaired (CU), (2) Mild Cognitive Impairment (MCI), (3) AD. Each cohort was divided for separate 3-group training and validation analyses. Novelty: High-throughput metabolomics analyses were conducted on both serum and saliva, each analysis detecting over 7000 metabolite peaks. Leading molecular predictors for the two-group comparisons and two AD-related bile acids produced 20 novel predictors. Data-driven analyses (in Python 3.9) included three machine learning classifier algorithms (eg, sklearn RandomForestClassifier) processing 112 (92+20) predictors. RESULT: Three sets of analyses (same predictors) were conducted (Predicting AD (CU-AD); Predicting MCI (CU-MCI); Predicting dementia (MCI-AD)). All models performed successfully in the main metrics (eg, Accuracy, Precision; AUC range: 0.81 - 0.98). Key results displayed in Explainable AI Tree SHAP plots vividly depicted discriminating patterns of leading multi-modal predictors. Commonly observed predictors were from new metabolomics analyses (saliva, blood), imaging, vascular, AD biomarkers, sensory, clinical health and lifestyle. Figure 1 displays the overall models tested (left panel) and the results for Predicting Dementia (right panel). A striking result is that of 112 predictors, the top 20 included 8 newly discovered metabolites, 5 of which were from salivary analyses. CONCLUSION: New molecular biomarkers (from both blood and saliva) perform strongly when integrated into databases with established AD biomarkers and risk factors. Implications include (1) enhancing identification of early mechanistic pathways, (2) importance of conducting integrative and interactive biomarker analyses, especially data-driven, and (3) potential for non-invasive salivary biomarkers for promoting assessment accessibility across diverse communities.
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 | 0.004 |
| Meta-epidemiology (narrow) | 0.001 | 0.000 |
| Meta-epidemiology (broad) | 0.001 | 0.001 |
| Bibliometrics | 0.002 | 0.001 |
| Science and technology studies | 0.000 | 0.000 |
| Scholarly communication | 0.001 | 0.001 |
| Open science | 0.001 | 0.001 |
| Research integrity | 0.000 | 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".