Alignment of density maps in Wasserstein distance.

Area of Science:

• Structural biology
• Computational imaging
• Biophysics

Background:

• Accurate alignment of 3D structures is crucial in cryo-electron microscopy (cryo-EM) for determining molecular shapes.
• Existing alignment algorithms may face challenges with complex density landscapes.

Purpose of the Study:

• To develop a novel algorithm for aligning three-dimensional objects represented as density maps.
• To improve the accuracy and computational efficiency of structural alignment in cryo-EM.

Main Methods:

• Minimizing the 1-Wasserstein distance between density maps under rigid transformations.
• Utilizing Bayesian optimization for efficient computation of the loss function.
• Testing the algorithm on real protein molecule data.

Main Results:

• The proposed algorithm demonstrates improved accuracy and efficiency compared to existing methods for protein molecule alignment.
• The 1-Wasserstein distance provides a more favorable loss landscape for optimization.
• Identified a need for new distance functions when aligning heterogeneous molecular pairs.

Conclusions:

• The 1-Wasserstein distance-based algorithm offers a robust and efficient solution for 3D density map alignment in cryo-EM.
• This method enhances the analysis of protein structures.
• Further research into novel distance metrics is warranted for heterogeneous structural comparisons.