# Feature functional theory - binding predictor (FFT-BP) for the blind prediction of binding free energies

## TL;DR

FFT-BP is a novel machine learning-based method that predicts protein-ligand binding free energies using microscopic features and similarity assumptions, showing high accuracy across benchmark datasets.

## Contribution

The paper introduces FFT-BP, a new feature functional theory-based predictor that combines physical models and machine learning for accurate binding affinity prediction.

## Key findings

- Achieves RMSEs around 2 kcal/mol on benchmark sets.
- Shows Pearson correlation coefficients above 0.75.
- Demonstrates robustness and accuracy in blind predictions.

## Abstract

We present a feature functional theory - binding predictor (FFT-BP) for the protein-ligand binding affinity prediction. The underpinning assumptions of FFT-BP are as follows: i) representability: there exists a microscopic feature vector that can uniquely characterize and distinguish one protein-ligand complex from another; ii) feature-function relationship: the macroscopic features, including binding free energy, of a complex is a functional of microscopic feature vectors; and iii) similarity: molecules with similar microscopic features have similar macroscopic features, such as binding affinity. Physical models, such as implicit solvent models and quantum theory, are utilized to extract microscopic features, while machine learning algorithms are employed to rank the similarity among protein-ligand complexes. A large variety of numerical validations and tests confirms the accuracy and robustness of the proposed FFT-BP model. The root mean square errors (RMSEs) of FFT-BP blind predictions of a benchmark set of 100 complexes, the PDBBind v2007 core set of 195 complexes and the PDBBind v2015 core set of 195 complexes are 1.99, 2.02 and 1.92 kcal/mol, respectively. Their corresponding Pearson correlation coefficients are 0.75, 0.80, and 0.78, respectively.