new PDB depositions vs. their blind AlphaFold predictions — a running test of "is folding solved?"

30JE_C

Complement C3 · P01024 · RCSB 30JE · AF-P01024-F1 (v6)

0.5775
TM-score
0.7893
lDDT
10.147
Cα-RMSD Å
76.96
mean pLDDT
0.4556
FRAUD score
100.0%
novelty (max id)

Per-domain breakdown

SourceDomainRangeResiduesTMRMSD Åmean Cα ΔName
PAEPAE:1-16631-1663 2970.3786 10.1479.299

All metrics

Global fold agreement

TM-score (norm. experiment)0.5775
TM-score (norm. model)0.1222
TM-score (norm. shorter)0.5775
TM-score (norm. longer)0.1222
Cα-RMSD (Å)10.147
backbone-RMSD (Å)10.142
all-atom-RMSD (Å)10.578
core-RMSD (Å)1.79
core fraction0.236
GDT_TS12.63
GDT_HA3.54
MaxSub0.8754
structural overlap (3.5 Å)0.0909

Local, superposition-free

lDDT0.7893
contact-map Jaccard0.7633
contact precision0.8487
contact recall0.8835
distance-matrix mean Δ (Å)5.162
CAD-score (approx)0.8181

Backbone & secondary structure

SS agreement Q3 (%)82.15
mean Δφ (°)23.5
mean Δψ (°)24.6
torsion within 30° (frac)0.6162
Rg experiment (Å)24.23
Rg model (Å)26.6
ΔRg (Å)2.37

Confidence calibration

mean pLDDT76.96
pLDDT↔lDDT Pearson0.6889
pLDDT↔lDDT Spearman0.7306
PAE↔observed Pearson0.5257
PAE overconfident frac0.0285
mean PAE (Å)13.802
mean observed error (Å)5.145

Context & headline

coverage of model0.1786
coverage of experiment1.0
seq identity aligned (%)100.0
confidently-wrong residue frac0.6397
FRAUD score0.4556

Deposited 2026-04-29 · released 2026-05-06 · EM · 3.5 Å · closest pre-cutoff chain: 2A73_2