David Herreros

ReconSIREN is a deep learning framework that improves CryoEM map accuracy by addressing preferred orientation errors in heterogeneous samples.

This paper introduces a Sliced Wasserstein loss function to improve optimization robustness in cryo-EM, showing it outperforms MSE in handling noisy data and conformational heterogeneity.

The paper introduces FlexConsensus, an algorithm that merges conformational landscapes from different CryoEM heterogeneity methods into a shared consensus space to improve reliability and reproducibility.

The paper investigates using p-values to estimate the number of distinguishable classes in cryoEM data by testing if classifications differ from random partitions.

The paper presents a cryo-EM conformational heterogeneity challenge to evaluate and improve methods for analyzing molecular motions in heterogeneous samples.

HetSIREN is a deep learning method for CryoEM conformational state reconstruction and refinement, improving structural accuracy by decoupling pose and CTF information from conformational landscapes.

The paper analyzes the conformational dynamics of SARS-CoV-2 Spike prefusion structures using advanced cryo-EM data and reveals a continuous flexibility space rather than discrete states.

The paper improves a marker-free image alignment method for electron tomography to enhance 3D reconstruction quality by addressing deformations during acquisition.