When screening for phenotypes there are many methods that are used. Usin’Life uses artificial intelligence methods to analyze the image data sets but there have other laboratories that use other interesting methods. Dr Lu’s lab published a paper in Nature methods (Nat. Methods 9, 977–980, 2012) that uses microfluidic technology to determine the phenotypic characteristics of the C.elegans and also separate the worms that are of interest. It is not one of the first microfluidic based sorters for larger multi-cellular organisms since there is significant precedence for similar devices that the authors acknowledge and also mention their previous work in their references.
Rohde, C. B., et al. Microfluidic system for on-chip high-throughput whole-animal sorting and screening at subcellular resolution. PNAS104, 13891-13895, (2007).
Doitsidou, M., et al. Automated screening for mutants affecting dopaminergic-neuron specification in C. elegans. Nat Meth5, 869-872, (2008).
Chung, K., et al. Automated on-chip rapid microscopy, phenotyping and sorting of C. elegans. Nat Meth5, 637-643, (2008).
Crane, M. M., et al. Computer-enhanced high-throughput genetic screens of C. elegans in a microfluidic system. Lab on a Chip9, 38-40, (2009).
Pardo-Martin, C.et al.High-throughput in vivo vertebrate screening. Nat Meth7, 634-636, (2010).
However, what is neat about this approach is their method of using Support Vector Machines (SVM) based methods to separate the worms of interest. This system is not fast – it takes about 14 seconds to separate one worm but it is certainly many folds faster than the manually used methods that involve tedious separation by hand.
This does validate the possibility of using fast algorithms to do image based screening in real time.