Microarrays (microplates etc) are plastic rectangles with a grid of “wells” containing biological materials. When another biological or chemical substance is added to these cells, the reaction is captured in digital images. For example, various concentrations of a chemical or a drug are added to the wells containing biological cells. The cells then start to divide faster, or slower, or simply die. The result affects the color of the substance in each cell. The image analysis automatically captures this data and draws conclusions. For example, you can pinpoint exactly at what concentration the drug becomes toxic. It’s like hundreds experiments in one! Appropriately, this is also called high throughput screening.

I’ve been working on a related project for one of our clients and I would like to present a modified version of Pixcavator. First it captures all the wells in the form of a list with all the data about them – in the usual way. Then it displays the gray level (intensity) for each well – according to its position in the microarray. Of course, instead of intensity you can display other characteristics of these objects: the average intensity, or the standard deviation, or the average color (for color images), etc.

The point of the post is this: the hard part of collecting the data about the objects is taken care of by Pixcavator - the rest is a easy exercise with the Pixcavator SDK.