RESUMEN
A technique for automatically evaluating microbatch (400 nl) protein-crystallization trials is described. This method addresses analysis problems introduced at the sub-microlitre scale, including non-uniform lighting and irregular droplet boundaries. The droplet is segmented from the well using a loopy probabilistic graphical model with a two-layered grid topology. A vector of 23 features is extracted from the droplet image using the Radon transform for straight-edge features and a bank of correlation filters for microcrystalline features. Image classification is achieved by linear discriminant analysis of its feature vector. The results of the automatic method are compared with those of a human expert on 32 1536-well plates. Using the human-labeled images as ground truth, this method classifies images with 85% accuracy and a ROC score of 0.84. This result compares well with the experimental repeatability rate, assessed at 87%. Images falsely classified as crystal-positive variously contain speckled precipitate resembling microcrystals, skin effects or genuine crystals falsely labeled by the human expert. Many images falsely classified as crystal-negative variously contain very fine crystal features or dendrites lacking straight edges. Characterization of these misclassifications suggests directions for improving the method.
Asunto(s)
Cristalización/instrumentación , Procesamiento de Imagen Asistido por Computador/clasificación , Microquímica/métodos , Robótica/métodos , Isomerasas Aldosa-Cetosa/química , Inteligencia Artificial , Cristalización/métodos , Microquímica/instrumentación , Nanotecnología , Reproducibilidad de los ResultadosRESUMEN
A method to rationally predict crystallization conditions for a previously uncrystallized macromolecule has not yet been developed. One way around this problem is to determine initial crystallization conditions by casting a wide net, surveying a large number of chemical and physical conditions to locate crystallization leads. A facility that executes the rapid survey of crystallization lead conditions is described in detail. Results and guidelines for the initial screening of crystallization conditions, applicable to both manual and robotic setups, are discussed.