Acta Cryst. (2008). D64, 1187-1195 [ doi:10.1107/S090744490802982X ]
Abstract: The ability of computers to learn from and annotate large databases of crystallization-trial images provides not only the ability to reduce the workload of crystallization studies, but also an opportunity to annotate crystallization trials as part of a framework for improving screening methods. Here, a system is presented that scores sets of images based on the likelihood of containing crystalline material as perceived by a machine-learning algorithm. The system can be incorporated into existing crystallization-analysis pipelines, whereby specialists examine images as they normally would with the exception that the images appear in rank order according to a simple real-valued score. Promising results are shown for 319 112 images associated with 150 structures solved by the Joint Center for Structural Genomics pipeline during the 2006-2007 year. Overall, the algorithm achieves a mean receiver operating characteristic score of 0.919 and a 78% reduction in human effort per set when considering an absolute score cutoff for screening images, while incurring a loss of five out of 150 structures.
Keywords: image analysis; machine learning; structural genomics; feature extraction.
 |   Portable Document Format (PDF) file (188.5 kbytes) |
| Portable Document Format (PDF) file (13706.1 kbytes) |
| Portable Document Format (PDF) file (3193.0 kbytes) |
To open or display or play some files, you may need to set your browser up to use the appropriate software. See the full list of file types for an explanation of the different file types and their related mime types and, where available links to sites from where the appropriate software may be obtained.
The download button will force most browsers to prompt for a file name to store the data on your hard disk.
Where possible, images are represented by thumbnails.
Copyright © International Union of Crystallography
IUCr Webmaster