Linear scaling approaches to quantum macromolecular similarity: evaluating the similarity function

Area of Science:

• Computational Chemistry
• Structural Biology
• Bioinformatics

Background:

• Electron density-based similarity functions traditionally scale quadratically with molecular size.
• This quadratic scaling limits computational efficiency for large molecular systems.

Purpose of the Study:

• To develop an improved algorithm for evaluating the Quantum Molecular Similarity (QMS) function.
• To achieve linear scaling for faster similarity evaluations, particularly for large molecules.

Main Methods:

• An improved algorithm identifies and computes only non-negligible interatomic contributions.
• A minimalist dynamic electron density model using approximate, single shell densities is introduced.
• The algorithm avoids computing unnecessary interatomic squared distances.

Main Results:

• The improved algorithm achieves linear scaling for electron density-based similarity evaluation.
• The method effectively reduces computational cost for large molecular systems.
• Fast electron density-based alignments on macromolecules are facilitated.

Conclusions:

• The proposed algorithm significantly enhances the efficiency of QMS calculations.
• This approach enables rapid and accurate analysis of large biomolecules.
• The method is suitable for fast electron density-based alignments on macromolecules.