APOGEE 2: multi-layer machine-learning model for the interpretable prediction of mitochondrial missense variants
Problem
Interpreting the pathogenicity of missense variants in mitochondrial DNA is difficult due to highly variable clinical manifestations.
Why it matters
Misclassified mitochondrial variants can delay or misdirect diagnosis of mitochondrial disorders.
What the AI did
APOGEE 2, an improved ensemble machine-learning method with a curated training set, predicts pathogenicity of mitochondrial missense variants and provides region-wise genome fragility assessments.
Results
'Enhanced performance metrics' claimed; no concrete numbers stated
Impact
Precompiled APOGEE 2 scores and probabilities made available for clinical and research use.
Record
| Venue | Nature Communications |
|---|---|
| Published | 2023 |
| Work type | article |
| Authors | Salvatore Daniele Bianco, Luca Parca, Francesco Petrizzelli, Tommaso Biagini, Agnese Giovannetti, Niccolò Liorni, Alessandro Napoli, Massimo Carella et al. |
| Citations | 42 (OpenAlex, at import time) |
| References | 54 |
| Model or method | APOGEE 2 (ensemble machine-learning method) |
| Organization | not stated in source |
| Primary source | https://doi.org/10.1038/s41467-023-40797-7 |
| Open access | gold |
| Keywords | mitochondrial DNA, missense variants, pathogenicity, machine learning, ensemble |
Importance rating
real but narrower advance
Variant interpretation for mitochondrial disorders is a real clinical need, but abstract gives no scale or prevalence; a useful diagnostic tool for a specific genetic domain.