Charbonneau Lab Using A.I. to Recognize Crystals in Protein Crystallization Experiments

Conceptual representation of a convolutional neural network

The Charbonneau lab is bringing together researchers from industry and academia to develop a tool for automatizing protein crystal recognition.  Scientists came together to collect half a million images of protein crystallization experiments into a database that can specify which of these protein cocktails led to crystallization, based on human evaluation. The team then worked with a group led by Vincent Vanhoucke from Google Brain to apply the latest in artificial intelligence to help identify crystals in the images. Read more on the Duke Research Blog and in Charbonneau's recent PLOS One article.