Equivalent Distance Geometry Error for Molecular Conformation Comparison

TL;DR

This paper introduces the Equivalent Distance Geometry Error (EDGE), a novel loss function for molecular conformation generation that ensures fair and efficient optimization of geometric features like bond lengths, angles, and dihedrals.

Contribution

The paper proposes EDGE, a new loss function that guarantees equivalence in geometric optimization and improves efficiency in conformation generation models.

Findings

EDGE outperforms existing loss functions in accuracy.

EDGE reduces computational time in conformation optimization.

Experimental results validate EDGE's effectiveness across tasks.

Abstract

Straight-forward conformation generation models, which generate 3-D structures directly from input molecular graphs, play an important role in various molecular tasks with machine learning, such as 3D-QSAR and virtual screening in drug design. However, existing loss functions in these models either cost overmuch time or fail to guarantee the equivalence during optimization, which means treating different items unfairly, resulting in poor local geometry in generated conformation. So, we propose Equivalent Distance Geometry Error (EDGE) to calculate the differential discrepancy between conformations where the essential factors of three kinds in conformation geometry (i.e. bond lengths, bond angles and dihedral angles) are equivalently optimized with certain weights. And in the improved version of our method, the optimization features minimizing linear transformations of atom-pair distances within 3-hop. Extensive experiments show that, compared with existing loss functions, EDGE performs effectively and efficiently in two tasks under the same backbones.