Diagnostic yield in NGS genome diagnostics is constraint by the high fraction of variants of uncertain significance (VUS), in large part due to insufficient interpretability of missense variation. Existing pathogenicity predictors offer strong performance, but often produce an unexplainable score lacking mechanistic insight. Here, we present the Digital Approximation of Variant Effects (MOLGENIS DAVE), an explainable missense variant predictor built on 12 biophysically grounded features spanning stability, hydrophobicity, electrostatics, and molecular interactions. Trained on curated Dutch diagnostic data, DAVE reliably classifies and breaks down predictions into interpretable feature contributions. With a focus on explainability, this framework aims to alleviate the VUS burden, advances clinically actionable variant interpretation and enables mechanistic follow-up.
- Published in F1000 and currently awaiting peer-review.
- Freeze 6 dataset (Gzipped CSV file)
- 23417 rows, each a DNA/protein variant (9345 LB, 2703 LP, 11221 VUS and 148 CF)
- 81 columns, metadata and features
- Modeling and prediction on freeze 6 (R script)
- Random Forest prediction model (RData file)
This repository contains both software code and data.
- All source code in this repository is licensed under the GNU General Public License v3.0 (GPL-3.0).
- All datasets, documentation data, annotations, and other non-code content in this repository are licensed under the Creative Commons Attribution 4.0 International License (CC BY 4.0).
Unless otherwise stated in a specific file or directory, these licenses apply respectively to code and data.
SPDX identifiers:
- Code: LGPL-3.0
- Data: CC-BY-4.0