Cryo-EM processing cluster

• Subject Area: Basic Research in Biology and Medicine
• Term: since 2026
• Project identifier: Deutsche Forschungsgemeinschaft (DFG) - Project number 580866234

Over the past years cryo-electron microscopy (cryo-EM), including single-particle analysis and cryo-electron tomography, has revolutionized the field of structural biology. It has established itself as an indispensable technique in life science, offering several advantages compared to other methods. Cryo-EM enables the high-resolution visualization of biological macro-molecules in vitro and in situ, providing crucial insights into their molecular structures, their mechanisms of action, function and regulation, as well as into their interaction with other proteins and their surroundings. Despite its pivotal role, cryo-EM still faces several challenges, such as the high purchase and maintenance costs of microscopes, the limited access time and the need for local high-level expertise, all of which are covered at the host institution. Another major obstacle that is often underestimated is the exceptionally high demand for dedicated computational resources. During single-particle analysis, millions of particles are iteratively identified, aligned, classified, and refined in three-dimensional space, making it particularly GPU-intensive. As a result, the reconstruction of a high-resolution cryo-EM map originates from the processing of several tens of terabytes of data over several months (up to years), composed of a plethora of different jobs, each of which might require days of uninterrupted GPU computation. This demand further increases when analyzing small membrane proteins or when analyzing complexes displaying high conformational or compositional heterogeneity - the type of samples addressed in the research projects. To accommodate our increasing computational needs as well as the full transition from a decentralized to a centralized solution, this proposal seeks to expand our existing centralized high-performance computer cluster (HPC), ensuring a sustainable, scalable and efficient solution for our cryo-EM research.

• DFG Programme: Major Research Instrumentation
• Major Instrumentation: Kryo-EM Rechencluster
• Instrumentation Group: 7000 Datenverarbeitungsanlagen, zentrale Rechenanlagen
• Applicant Institution: Ruprecht-Karls-Universität Heidelberg
• Leader: Professorin Dr. Cristina Paulino