Skip to content

CryoFAST – Fast Acquisition for Single Particle & Tomography

“ CryoFAST provides cutting edge AI-driven solutions for automating data collection. By lowering the requirement for human intervention, CryoFAST can help provide a more streamlined, higher throughput screening of Cryo-EM grids and increase the rate of data output….”
Vamseedhar Rayaprolu, PhD, Cryo-EM Microscopist 2, PNCC

CryoFAST and the Cryo-EM workflow

workflow1b

User Challenges with Cryo-EM image acquisition

ch1

Sample Screening requires manual selection of squares & holes on the grid using complex microscope control software

Screening can take up to 30 minutes per sample​ of detailed manual work

ch2

Data Collection can take up to 16 hours of overnight automated collection

Over 70% of data collected can be “junk” data​ and is wasted for processing

Today’s Cryo-EM image acquisition is​ Complex, Costly, and Failure-Prone​

painpoints2b

HTI’s Automated Cryo-EM dataflow solution

Deep learning models provide automated, unbiased 2D/3D data collection & data processing​

Trained using proprietary and public datasets, Patent Pending1​

Interfaces with Cryo-EM instrument control solutions​

​Enterprise workbench for lab operation (supports multiple users, samples/grids, & microscopes

1Patent Pending US No. 63/370,066

HTI’s CryoFAST AI automation software

pp1

Uses AI/ML to automate Sample Screening & ​Data Collection​

  • Requires little to no user involvement with “push-button” automation​
  • Transfers screening insights to​ data collection
pp2

Includes an Easy-to-Use GUI and data management features

  • Reduces complexity removing the need for advanced user training​
  • Enterprise workbench supporting multiple users, grids/samples, scopes
pp3

Improves Productivity​ and Accuracy​​

  • Increases screening throughput by 2X​
  • Optimizes data collection reducing overnight scope time by 50%​
  • Improves data quality reducing wasted data to less than 30%​

A portion of this research was supported by NIH grant U24GM129547 and performed at the PNCC at OHSU and accessed through EMSL (grid.436923.9), a DOE Office of Science User Facility sponsored by the Office of Biological and Environmental Research