• Corpus ID: 224802880

Wasserstein K-Means for Clustering Tomographic Projections

  title={Wasserstein K-Means for Clustering Tomographic Projections},
  author={R. Bharat Rao and Amit Moscovich and Amit Singer},
Motivated by the 2D class averaging problem in single-particle cryo-electron microscopy (cryo-EM), we present a k-means algorithm based on a rotationally-invariant Wasserstein metric for images. Unlike existing methods that are based on Euclidean ($L_2$) distances, we prove that the Wasserstein metric better accommodates for the out-of-plane angular differences between different particle views. We demonstrate on a synthetic dataset that our method gives superior results compared to an $L_2… 

Figures from this paper

On the robustness of certain norms

A family of norms defined for functions on an interval obtained by taking the p-norm of the Volterra operator applied to the function are studied, and it is shown that they are robust to additive noise and bounded by the size of the difference in projection directions.

AlignOT: An optimal transport based algorithm for fast 3D alignment with applications to cryogenic electron microscopy density maps

A new procedure, called AlignOT, which relies on the use of computational optimal transport (OT) to align EM maps in 3D space, by embedding a fast estimation of OT maps within a stochastic gradient descent algorithm.

Probing Structural Perturbation of Biomolecules by Extracting Cryo-EM Data Heterogeneity

An overview is provided that briefly describes the workflow of single-particle cryo-EM, including imaging and data processing, and new methods developed for analyzing the data heterogeneity to understand the structural variability of biomolecules.

Manifold Learning with Arbitrary Norms

This paper determines the limiting differential operator for graph Laplacians constructed using any norm, and demonstrates the potential benefits of non-Euclidean norms in manifold learning in the task of mapping the motion of large molecules with continuous variability.

CryoGAN: A New Reconstruction Paradigm for Single-Particle Cryo-EM via Deep Adversarial Learning

CryoGAN is an unsupervised algorithm that only demands projection images and an estimate of the contrast transfer function parameters that can provide reconstructions in a matter of hours on a high-end GPU and opens the door to a family of novel likelihood-free algorithms for cryo-EM reconstruction.



Fast Computation of Wasserstein Barycenters

The Wasserstein distance is proposed to be smoothed with an entropic regularizer and recover in doing so a strictly convex objective whose gradients can be computed for a considerably cheaper computational cost using matrix scaling algorithms.

Earthmover-Based Manifold Learning for Analyzing Molecular Conformation Spaces

A novel approach for manifold learning that combines the Earthmover's distance (EMD) with the diffusion maps method for dimensionality reduction is proposed and the potential benefits of this approach for learning shape spaces of proteins and other flexible macromolecules are demonstrated.

Fast maximum-likelihood refinement of electron microscopy images

Application of this reduced-search approach to a cryo-EM dataset yielded practically identical results as the original approach, but in approximately one day instead of one week of CPU.

Reconstructing continuous distributions of 3D protein structure from cryo-EM images

The proposed method, termed cryoDRGN, is the first neural network-based approach for cryo-EM reconstruction and the first end-to-end method for directly reconstructing continuous ensembles of protein structures from cryo -EM images.

Rotationally Invariant Image Representation for Viewing Direction Classification in Cryo-EM

Cryo-EM reconstruction of continuous heterogeneity by Laplacian spectral volumes

A new method for the reconstruction of macromolecules exhibiting continuous heterogeneity is presented, using projection images from multiple viewing directions to construct a graph Laplacian through which the manifold of three-dimensional conformations is analyzed.

Unsupervised particle sorting for high-resolution single-particle cryo-EM

It is shown that particles can be successfully sorted based on a simple statistical model for the distribution of scores assigned during refinement, an important step towards the development of automated workflows for protein structure determination using single-particle cryo-EM.

Computational Optimal Transport: With Applications to Data Science

Computational Optimal Transport presents an overview of the main theoretical insights that support the practical effectiveness of OT before explaining how to turn these insights into fast computational schemes.