Automated classification of crystallisation experiments

Julie C. Wilson, Naomi Chayen (Editor)

Research output: Chapter in Book/Report/Conference proceedingChapter


With robotic systems capable of performing tens of thousands of experiments a day in high-throughput mode, visual inspection of the results is becoming increasingly impractical and the development of tools to automate analysis is essential. Many laboratories now have imaging systems to record the results of crystallisation trials at regular intervals and the development of software to automatically analyse the images is discussed. Whilst the detection of crystals is the fundamental aim, reliable classification of other outcomes provides information for subsequent trials and allows experiments to be ordered so that only the highest scoring images need be checked by eye. Pattern recognition procedures require characteristic features to be quantified for classification and various methods for the extraction of features from crystallisation images are discussed. The different classes and most commonly used classification methods are also described.
Original languageEnglish
Title of host publicationProtein crystallisation strategies for Structural Genomics
PublisherInternational University Line
Number of pages20
ISBN (Print)978-0-9720774-3-9
Publication statusPublished - 2006

Cite this