The SBGrid Train-the-Trainer Program provides hands-on continuing education for structural biology faculty, principal investigators, senior researchers, cryo-EM facility staff, and other mentors who are positioned to bring advanced data-processing expertise back to their home institutions. Program development and delivery are informed by ongoing participant feedback and iterative curriculum refinement.

Funded through an NIH R25 award to Piotr Sliz at Harvard Medical School and Jamaine Davis at Belmont University, the program was created to expand access to practical, mentor-centered training in modern structural biology computation. The program has grown from its original focus on single-particle cryo-EM into a broader training platform that now includes cryo-electron tomography, or cryo-ET, data processing.

All workshops are delivered using SBCloud, the SBGrid cloud-based training environment. SBCloud gives participants access to a consistent, ready-to-run computational platform with curated software, CPU and GPU resources, high-performance storage, and visualization tools. This allows participants to focus on scientific workflows rather than local software installation, hardware limitations, or version conflicts.

Program goals

• Train faculty and senior researchers who can teach others.
• Expand access to modern cryo-EM and cryo-ET data-processing workflows.
• Provide reusable training materials, datasets, and community support.
• Reduce computing and software barriers through SBCloud.
• Build a national community of structural biology mentors.

• Contact the SBGrid training team
• Learn more about SBCloud

Single-Particle CryoEM Data Processing

The CryoEM Train-the-Trainer workshop was developed to provide practical instruction in single-particle cryo-electron microscopy data processing using modern cloud-based computational infrastructure. Single-particle cryo-EM has become a central structural biology technique for determining high-resolution macromolecular structures, but the field continues to face challenges related to computational complexity, large-scale data handling, and access to standardized training materials.

The two-day Single-Particle CryoEM Data Processing curriculum guides participants through a complete image-processing workflow, from raw movie data to refined 3D reconstruction and interpretation of results. The curriculum uses the EMPIAR-11422 Titan Krios dataset, which resulted in the cryoEM structure of full-length dimeric ClbP (PDB 7UL6; EMD-26593), originally published by Velilla et al., 2022. Participants work with authentic experimental data while learning reproducible processing workflows on cloud-based computational resources.

Topics include:

• Cryo-EM workflow overview
• Data import and organization
• Motion correction and CTF estimation
• Particle picking and extraction
• 2D classification and particle selection
• Initial model generation
• 3D classification and refinement
• Local resolution estimation and filtering
• Visualization and interpretation of cryoEM density maps

CryoET Data Processing

The CryoET Train-the-Trainer workshop was developed in response to growing community demand for practical training in cryo-electron tomography. Cryo-ET is rapidly emerging as a powerful approach for studying macromolecular organization in native cellular environments, but the field remains limited by complex multi-tool workflows and a shortage of standardized training resources.

The two-day CryoET Data Processing curriculum guides participants through a complete tomography workflow, from raw data to subtomogram averaging and presentation of results. The curriculum uses a cryo-ET training dataset from milled Saccharomyces cerevisiae cells, made publicly available through the SBGrid Data Bank under doi:10.15785/sbgrid/1255. The dataset includes tilt-series of milled yeast cells showing mitochondria, nuclear envelopes, lipid bodies, smooth endoplasmic reticulum, and other intracellular components.

Topics include:

• Cryo-ET workflow overview
• Tilt-series preprocessing and alignment
• Tomogram reconstruction
• Denoising and missing-wedge correction
• Segmentation and particle identification
• Particle export
• Subtomogram averaging
• Figure generation and final participant presentations

• Access the CryoET training dataset at data.sbgrid.org/dataset/1255/
• View data access instructions
• Watch the CryoET training workflow webinar

SBGrid workshops are designed around a Train-the-Trainer model. Participants learn workflows in a hands-on setting and receive materials that can be reused or adapted with their own students, trainees, laboratory members, or facility users.

Each workshop includes:

• Instructor-led teaching sessions
• Hands-on tutorials in SBCloud
• Use of shared training datasets
• Guided and independent data-processing exercises
• Discussion of best practices and common pitfalls
• Final participant presentations
• Access to training materials after the workshop
• Continued support through office hours and alumni channels

This structure helps participants build both technical proficiency and the confidence to teach or support these workflows in their own research communities.

Since the start of the NIH R25 award, SBGrid has delivered a series of cryo-EM workshops for faculty and mentors from institutions across the United States. The current workshops page notes that the program has trained faculty participants from 55 U.S. institutions and that all workshops use SBCloud to provide a uniform processing environment.

The program has now expanded to include cryo-ET data processing. During the most recent reporting period, SBGrid developed, piloted, refined, and launched the CryoET Train-the-Trainer curriculum through pilot, faculty-directed, and faculty-only workshop formats. These workshops trained research fellows, faculty, facility staff, and senior researchers from a growing set of institutions, while participant feedback was used to refine the curriculum, pacing, and support materials.

In the next phase of the program, SBGrid will continue offering in-person CryoEM and CryoET workshops while also testing online formats. The online format will preserve core course content, including lectures, hands-on tutorials, guided data processing, and final presentations, while distributing the same total instructional time across a longer period, such as two weeks. This approach is intended to broaden participation for investigators who cannot easily travel or commit to a two-day intensive in-person course.

The Train-the-Trainer program provides multiple resources for participants, alumni, and the broader structural biology community.

Public training dataset

• CryoET Training data from Milled Saccharomyces cerevisiae cells
• SBGrid Data Bank ID: 1255
• doi:10.15785/sbgrid/1255
• Dataset type: CryoET training
• Dataset size: approximately 37 GB
• Data source: Raunser Laboratory, Max Planck Institute of Molecular Physiology

• Open dataset page

Training materials

Workshop training materials are shared with participants through Google Workspace. These materials are designed to support reuse by workshop alumni at their home institutions.

• Request access or more information

Recorded educational content

SBGrid webinars and selected training-related presentations are made available through YouTube when possible, allowing participants and community members to revisit material or access it asynchronously.

• SBGrid YouTube channel
• SBGrid Webinar archive

Community updates

Training opportunities and updates are shared through SBGrid communications, including the monthly SBGrid newsletter and community channels. The SBGrid newsletter is distributed monthly to current consortium members.

• Follow the SBGrid Community News on Medium
• Read the SBGrid newsletter
• Subscribe or request to be added

Workshop participation does not end when the course concludes. The program provides continued support to help alumni apply, adapt, and disseminate what they learned.

Office hours

Monthly virtual office hours provide ongoing support for workshop alumni and members of their groups. These sessions support troubleshooting, workflow implementation, curriculum adaptation, and discussion of cloud-based or local computing environments. Currently, Open Office Hours are held at 3pm on the last Thursday of every month. If this time does not suit you, you may request an alternative time using the pre-registration form below.

Please note, our open office hours are not intended to be general structural biology or project support. Our office hours are targeted at supporting alumni to integrate their workshop learnings into their home institutions or teams.

• Submit a question for office hours

Train-the-Trainer dissemination

Participants are encouraged to reuse workshop materials with students, postdoctoral fellows, laboratory members, collaborators, and facility users. The goal is to extend the impact of each workshop beyond direct attendance and support a growing national community of structural biology mentors.

The program is intended for researchers and educators who are likely to disseminate training or support others after the workshop.

Appropriate applicants include:

• Faculty and principal investigators
• Senior researchers and staff scientists
• Cryo-EM and cryo-ET facility staff
• Educators developing structural biology training programs
• Advanced trainees who support data processing or training in their groups

Applicants do not need to be expert computational scientists, but should have a strong interest in structural biology workflows and in sharing knowledge with others.

In-person workshops

In-person workshops provide intensive hands-on training over two days. These workshops include direct interaction with instructors, teaching assistants, and other participants.

Online workshops

Beginning in the next phase of the program, SBGrid will test online workshop formats. Online courses will preserve the same instructional content and total training hours as the in-person format, but may distribute the material over a longer period. This format is designed to increase access for participants who cannot travel or step away from local responsibilities for a two-day intensive course.

Future curriculum development

SBGrid also plans to add dedicated cryo-EM/cryo-ET curriculum development workstations at Harvard Medical School and Belmont University. These systems will support rapid development, testing, and troubleshooting of new training modules before they are migrated into SBCloud for broader workshop delivery.

The SBGrid Train-the-Trainer Program is led by:

Piotr Sliz
Harvard Medical School / SBGrid Consortium

Jamaine Davis
Belmont University

The SBGrid Train-the-Trainer Program is supported by our wonderful teaching team:

Shaun Rawson
Harvard Medical School / SBGrid Consortium

Peter Meyer
Harvard Medical School / SBGrid Consortium

Alice-Roza Eruera
Harvard Medical School / SBGrid Consortium

Estrella Fernandez-Gimenez
SBGrid Consortium

The program is supported by SBGrid scientists, software curators, cloud engineers, instructors, teaching assistants, and collaborators from the broader structural biology community.

• Contact the program team