ProModEv

Welcome to ProModEv (Protein Model Evaluation Portal). This help page explains how to use the portal, how to interpret benchmarking metrics, what the limitations are, and where to find downloadable resources.

ProModEv presents curated benchmarking results for AlphaFold2, AlphaFold3 and ESMFold across monomeric and dimeric targets (PDB releases Jan 1, 2022 – Dec 31, 2024). For each target the portal displays prediction quality metrics, comparative visualisations and downloadable coordinates

Main pages and functionality

1. Comparison View: Click on the Monomer and Dimer header for comparisonal view
2. Each tool Report: Various classification and metrics were displayed

The above pages are displayed in the table. The users are allowed to sort, query and search. Each PDB ID hyperlinked to detailed visualization page

---

Metrics

pLDDT (predicted LDDT; 0–100):Per-residue confidence predicted by AlphaFold/ESMFold. Higher is better. Use as a relative indicator — but do not treat it as proof of correctness, especially for interfaces. See portal-calibration plots for tool-specific calibration

LDDT (0–1): Local Distance Difference Test computed against the experimental structure. LDDT ≥ 0.70 is used here as the threshold for “correct” backbone/local geometry for monomers

RMSD (Å): Global superposition RMSD of Cα atoms. RMSD ≤ 3.0 Å + LDDT ≥ 0.70 → “Accurate Prediction” in our classification (monomers).

DockQ (0–1): Composite docking score for protein–protein interfaces. DockQ ≥ 0.80 = high quality, 0.49–0.80 = medium, 0.23–0.49 = acceptable, <0.23 = incorrect

---

Interpretation

• Start with the per-target page — check pLDDT, LDDT/RMSD (if experimental structure available), DockQ (for dimers)


• If pLDDT ≥ 90 and LDDT ≥ 0.9 → model is highly reliable for both local and global geometry.


• If pLDDT 70–90 and LDDT ≥ 0.7 → reliable backbone prediction; verify functional sites or loops if important


• If pLDDT ≥ 70 but LDDT < 0.7 or DockQ < 0.23 (for dimers) → exercise caution: pLDDT can be overconfident for interfaces and NMR ensembles. Inspect interface area, evolutionary support (TQ) and MSA depth.


• If TQ = 0 (no homologs) → treat prediction as a de novo model; cross-check multiple tools and, if possible, use orthogonal data (crosslinking, SAXS, mutagenesis).


• If interface area < ~1000–1200 Ų and DockQ < 0.23 → interface may be crystallographic / not biological. Consider PISA or biological-assembly annotations

**Always inspect models visually for context-specific decisions