QC Score

Per-residue interaction energy decomposition for protein-ligand complexes. Know which atoms drive binding, what physics is responsible, and where your next modification should go.

Know Which Atoms Drive Binding and Why

QC Score computes per-residue interaction energies within the full electronic environment of the binding site. InteractionMap then decomposes each contact into electrostatics, dispersion, exchange-repulsion, and induction using symmetry-adapted perturbation theory. Together, they give you the resolution to see that a PEG linker fragment in a PROTAC contributes more dispersion energy than the warhead itself, or that a specific amide region is the sole electrostatic anchor at a protein interface.

Performance at a glance

5-15

minutes per typical drug-like ligand

1000

ligands in parallel

<1

hour for 1000 ligands

100%

parameter-free

Accuracy that holds up against standard benchmarks

Competitive R-squared values across JAK1, CHK1, CDK2, and other standard benchmark datasets. Fully ab initio, completed in minutes, with no parameterization required.

QC Score is a fully first-principles tool benchmarked against industry-standard methods and datasets.

Dataset QC Score
(R2)
FEP+
(R2)
QC Score
(Tb)
FEP+
(Tb)
QC Score
(Tw)
FEP+
(Tw)
QC Score
(p)
FEP+
(p)
JAK1 0.72 0.69 0.82 0.46 0.97 0.66 0.93 0.61
CHK1 0.74 0.62 0.66 0.61 0.87 0.83 0.8 0.76
CDK2 0.71 0.42 0.67 0.44 0.86 0.69 0.82 0.62
T4 Lysozyme 0.37 0.43 0.33 0.36 0.68 0.67 0.51 0.55
LigA 0.85 0.93 0.6 0.78 0.9 0.97 0.74 0.92
JAK2(1) 0.67 0.76 0.59 0.77 0.75 0.87 0.72 0.9
1Ross, G.A., Lu, C., Scarabelli, G. et al. The maximal and current accuracy of rigorous protein-ligand binding free energy calculations. Commun Chem 6, 222 (2023). https://doi.org/10.1038/s42004-023-01019-9

GPU-Accelerated Quantum Chemistry

QC Score calculations complete in minutes per drug-like ligand on Promethium's GPU engine. Run scoring campaigns in parallel across multiple GPU instances, so your team can iterate on scoring cycles weekly instead of monthly.

First-Principles Accuracy, Zero Parameters

Fully ab initio with PCM solvation modeling. QC Score handles charged species, metal-containing active sites, PROTACs, molecular glues, macrocyclic inhibitors, antibody-drug conjugates (ADCs), and covalent ligands with the same first-principles treatment it applies to standard drug-like molecules.

Use with InteractionMap

Export from QC Score into InteractionMap for four-component energy decomposition based on symmetry-adapted perturbation theory. QC Score identifies which residues dominate binding in context. InteractionMap reveals whether each contact is electrostatic, dispersive, or inductive. When the two methods diverge on a residue, that divergence reveals environmental dependence worth investigating.

Built for PROTACs, Molecular Glues, and Macrocycles

PROTACs, molecular glues, and macrocyclic inhibitors operate through ternary cooperativity, neomorphic interfaces, and bifunctional energy distribution. QC Score resolves these at the functional-group level, giving you the same atom-level insight on a 2,000-atom ternary complex that you get on a traditional small-molecule binding pocket.

How QC Score compares

Traditional Methods
Complex setup and configuration
Requires extensive parameterization
Days for high-throughput screening
Limited accuracy for charged species
Poor scalability for large studies
QC Score + InteractionMap
Minimal setup, intuitive interface
No experimental data required
Hours for 1000s of ligands
Fully ab initio and parameter-free
Excellent for charged ligands
Massive parallel processing capability

Platform Capabilities

  • Proprietary fragmentation-based quantum chemistry
  • Advanced PCM solvation modeling
  • Automated binding site cutout definition (3-10 Å)
  • One-click export to InteractionMap powered by F-SAPT for functional-group analysis
  • Cloud-native scalability for high-throughput

See the Physics Behind Binding

Computational chemistry teams at biotech and pharmaceutical companies use QC Score and InteractionMap for their lead optimization decisions.