Automatic Microarray Image Segmentation

The"Amip" GUI is intended to provide code examples of the methods proposed in:
"Unsupervised image segmentation for microarray spots with irregular contours and inner holes"
authors: Bogdan Belean, Monica Borda, Jörg Ackermann, Ina Koch and Ovidiu Balacescu
Journal: BMC Bioinformatics, 2015, 16:412
https://bmcbioinformatics.biomedcentral.com/articles/10.1186/s12859-015-0842-3

PDE were used for grid alignment, whereas autocorrelation driven PDE and k-means clustering
were used for spot segmentation. The proposed segmentation procedure yields the spots median
intensity values for the microarray images recorded from Cy3 and Cy5 fluorescent dyes. Based on
the fold change factor, the differentially expressed genes were determined.

Both images with single spot groups and multiple spot groups (372 and 324 spots/group) can be
processed. Example of such images were taken from the GEO and SMD databases. e.g., GSM102718 - GEO experiment for single spot group.