Software ‘learns’ to automate cell imaging and experiments

Researchers at the European Molecular Biology Laboratory Heidelberg have developed new software that quickly learns what researchers are looking for and automatically performs microscopy experiments, as published in Nature Methods.

Feb 10th, 2011

Researchers at theEuropean Molecular Biology Laboratory (EMBL) Heidelberg (Germany) have developed new software that quickly learns what researchers are looking for and automatically performs microscopy experiments, as published in Nature Methods.

Called Micropilot, theimaging software combines microscopy with machine learning. It analyzes low-resolution images taken by a microscope and, once it has identified a cell or structure the scientists are interested in, automatically instructs the microscope to begin the experiment, which may range in complexity from recording time-lapse videos to using lasers with fluorescently tagged proteins.

Jan Ellenberg and Rainer Pepperkok, whose teams at EMBL designed the intelligent imagingsoftware, have used the program to initiate several microscopy experiments, investigating aspects of cell division. They determined when structures known as endoplasmic reticulum exit sites form, and uncovered the roles of two proteins, CBX1 and CENP-E, in condensing genetic material into tightly wound chromosomes and in forming the spindle that helps align those chromosomes. This software will be a key tool for the European systems biology projects Mitosys and SystemsMicroscopy, for which Ellenberg and Pepperkok are developing technology.

The Micropilot software is available asopen source code.

SOURCE: C. Conrad et al., “Micropilot: automation of fluorescence microscopy– based imaging for systems biology,”Nature Methods, advance online publication (Jan. 23, 2011); doi: 10.1038/NMETH.1558.

-- Posted byVision Systems Design

More in Boards & Software