RFT Research
Proprietary Physics Engine

Next-Generation
Molecular Physics

A unified geometric engine powering hardware-accelerated drug discovery at LatticeZero.

Not Originally Designed for Biology

The LatticeZero compute engine is powered by a unified geometric framework originally developed to model physical systems across vastly different scales. To apply it to molecular systems, we transposed the same continuous lattice mathematics into the angstrom regime. The result is a molecular docking engine that bypasses explicit pairwise interaction summation and expensive quantum-mechanical approximations entirely.
“From first principles to sub-angstrom steric precision.
One unified mathematical framework.

First-Principles Physics, Not Fitted Heuristics

While much of the industry still relies on black-box machine learning or legacy force-field tuning, the RFT engine is built on first-principles physics. Core interaction terms and constants are derived from geometry instead of learned from labeled poses. The result is HTVS-class throughput on a single consumer GPU, with internal RTX 3090 benchmarks exceeding 10,000 ligands per second, without training data or empirical curve fitting.

The RFT Architecture

Three foundational advances that enable physics-accurate molecular scoring at speeds previously impossible.

Pillar 01

Advanced Geometric Discretization

Most docking pipelines use fixed Cartesian sampling and spend substantial compute in low-value pocket volume. RFT uses a unified geometric framework with proprietary isotropic spatial packing mathematics to evaluate continuous protein pockets with higher volumetric efficiency. This reduces dead-space computation and enables sub-angstrom conformational searches.

Pillar 02

Unified Energy Decomposition

Rather than patching together disconnected empirical approximations, the RFT engine evaluates 14 distinct physical interactions, including precise coordination geometries and second-shell hydrophobic effects, from a unified mathematical foundation.

Pillar 03

Bare-Metal GPU Parallelization

The RFT engine was designed for matrix-native GPU parallelization, allowing complex energy tensor calculations to run client-side via WebGPU compute shaders at HTVS-class throughput. No server queues are required. Internal RTX 3090 benchmarks exceed 10,000 ligands per second, and internal mobile WebGPU tests on recent iPhone-class hardware reach multi-thousand ligands per second, including runs around 5,000 ligands per second. Workflows that once required cloud farms can now be achieved on a device in your pocket.

Proven on Standard Datasets

Validated against DUD-E and DEKOIS2 benchmarks with property-matched decoys. Every result includes holdout cross-validation and Y-scramble null controls.

0.980
Best AUC
100+
Validated Targets
10K+
Ligands / Second
0
Training Data
Target Family Validated AUC
HMGR Reductase 0.980
PDE5 Phosphodiesterase 0.883
Neuraminidase Glycosidase 0.875
MCL1 Apoptosis Regulator 0.868
ESR1 Nuclear Receptor 0.858
ADRB2 GPCR 0.834
CATL Cysteine Protease 0.829

Validated via 5-fold cross-validation with Y-scramble null controls. All AUC values are holdout-validated on DEKOIS 2.0 with property-matched decoys.

Quantum-Accurate Interaction Energies

Beyond ranking poses, the same engine computes quantum-mechanical interaction energies directly from geometry, with zero fitted parameters and no training data. On gold-standard quantum-chemistry benchmarks it reaches the accuracy tier of machine-learning potentials trained on millions of reference calculations. Because the physics is computed rather than learned, accuracy holds across chemistry, not only the molecules it was built on.

0.92
PLA15 Correlation (r)
21.7
MAE (kcal/mol)
0
Fitted Parameters
4.2M
Systems / Second
Method Correlation (r) MAE (kcal/mol) Trained Parameters
eSEN0.99614.4millions
GFN2-xTB0.99210.6semi-empirical
UMA-S0.99116.3millions
AIMNet20.98535.8millions
RFT Engine0.9221.7none
ANI-2x0.7173.0millions
MACE0.78112.2millions

PLA15 protein-ligand interaction-energy benchmark (Kříž & Řezáž, J. Chem. Inf. Model. 2020), reference CCSD(T)/CBS. Comparator predictions from Rowan Scientific's published pla15-benchmarks (MIT-licensed); correlation and mean absolute error computed on the identical 15 structures. Every method at or above the RFT engine was trained on millions of quantum-chemistry calculations. The RFT engine uses none.

Cross-checked on standard physics benchmarks: S66 dimers ρ = 0.92 (vs SAPT2+), A24 ρ = 0.84 (vs near-full-CI), S66×8 / 528 points ρ = 0.90.

Read the full method note

Experience the Engine

The RFT engine is available exclusively through the LatticeZero platform.

Explore LatticeZero