Prediction & validation
Backbone dihedral prediction
Predict backbone φ and ψ angles from chemical shifts, with per-residue reliability.
Chemical shifts encode local backbone geometry well enough that φ and ψ can be predicted directly from them, which is why shift-derived dihedral restraints are standard input to structure calculation.
The prediction is only as good as its honesty about reliability. A confident prediction in a region where the shifts are ambiguous produces a restraint that actively pulls a structure calculation wrong. CHORD, our own predictor, is trained on public data and reports calibrated per-residue confidence, so restraints can be filtered on something meaningful rather than accepted wholesale.
Predictions come back as φ/ψ with uncertainties, plotted on a Ramachandran surface, and export directly as restraints.
What it needs
- Assigned backbone chemical shifts
- Protein sequence
What you get
- φ and ψ per residue with confidence
- Ramachandran plot
- Dihedral restraints for structure calculation
What does the work
CHORD
Built in-houseOur dihedral predictor with calibrated confidence
Related in Prediction & validation
- Chemical shift predictionPredict chemical shifts from a structure — to validate a model, to check an assignment, or to seed one.
- Secondary structure predictionDetermine secondary structure from chemical shifts or from sequence alone.
- Structure validation against shiftsScore how well a structure explains the chemical shifts measured for it — a check that geometry-only validation cannot make.