Filtering of 3D volumes for heterogenity estimation
Title: Filtering of 3D volumes for heterogenity estimation
DNr: LiU-compute-2025-29
Project Type: LiU Compute
Principal Investigator: Björn Forsberg <bjorn.forsberg@liu.se>
Affiliation: Linköpings universitet
Duration: 2025-06-11 – 2026-01-01
Classification: 10203
Keywords:

Abstract

Cryo-Electron Microscopy (cryo-EM) has emerged as a pivotal tool for structural biology, enabling the detailed visualization of biological macromolecules in near-native states. The increasing complexity of cryo-EM data, including high-resolution reconstructions, has driven the need for robust computational resources. While GPUs substantially accelerate computational tasks, such as particle alignment, 3D reconstruction, and noise reduction, far from all computational tasks demand or can even efficiently use GPUs. We are currently undertaking a systematic assessment and further development and testing of our tools for cryo-EM reconstruction that are not GPU accelerated.