|
|
 |
|
| Bernhard Rupp - | |
|
|
TB Structural Genomics Consortium, U. of California - LLNL, Livermore, CA 94551
Substantial funding of public and commercial structural proteomics efforts has led to
accelerated technical development and availability of high throughput robotics for crystallography.
Automation deployed in protein crystallography extends from parallel, small scale protein expression
screening to fully automated protein crystallization systems, and finally automated sample mounting,
data collection, and structure solution and refinement. Successful implementation of laboratory
automation requires careful planning with a balance of expectations and resources versus long-term
costs and return on investment. It is important to recognize that not every research environment
requires the same type or degree of automation, and operations research and process analysis are
necessary to select the robotic automation layout most suitable to achieve maximal overall efficiency.
Based on process outlines and a review of critical steps we provide examples for cost-effective,
modular designs for crystallography laboratory automation, emphasizing the importance of automated
process scheduling, data capture, and adequate data base support. We will also discuss basics of data
analysis and machine learning towards predictive models for protein crystallization.
Lawrence Livermore National Laboratory is operated by the University of California for the United
States Department of Energy under contract no. W-7405-ENG-48. Work funded by NIH P50 GM62410 (TB
Structural Genomics) center grant.
|
|
| |
|
|
