MTR3D

MTR3D computes three-dimensional spatial intolerance scores for missense variation to identify protein regions under purifying selection by integrating population variation with protein structural data.

Key Features:

• 3D Spatial Intolerance (MTR3D): MTR3D computes spatial intolerance scores using spherical neighborhoods with radii of 5–8 Å, providing improved identification of regions under purifying selection compared to sequence-based methods.
• Reduced regional bias: The method reduces bias in regions with limited data, enhancing reliability in pinpointing critical areas within protein structures.
• Integration with sequence-based scores: MTR3D combines spatial-based scores with the Missense Tolerance Ratio (MTR) to produce a consensus score (MTRX) that distinguishes pathogenic from benign variants with an AUC of 0.85.
• Visualization and data coverage: Visualizes population variation, MTR, MTR3D, and MTRX across genes and structures and includes data for over 17,000 human genes and more than 42,000 alternative transcripts with Ensembl and RefSeq annotations.
• Enrichment of pathogenic variants: Intolerant regions identified by MTR3D are highly enriched for ClinVar pathogenic and COSMIC somatic missense variants (Mann-Whitney U test P < 2.2 × 10^-16).

Scientific Applications:

• Identification of disease-causal variants: Identifying disease-causal missense variants within protein structures by highlighting spatially intolerant regions.
• Variant interpretation: Distinguishing pathogenic from benign missense variants by integrating spatial and sequence intolerance measures.
• Genetic disease and personalized medicine: Prioritizing variants for studies of genetic diseases and for personalized medicine applications.

Methodology:

Population variation data are mapped to structural models and evaluated using sequence-based (MTR) and spatial-based (MTR3D) intolerance scores, which are integrated into the consensus metric MTRX.