Fully-Automated Spot Recognition and Matching Algorithms for 2-D Gel Electrophoretogram of Genomic DNA.

摘要: We have developed the fully-automated algorithms for processing 2-D gel electrophoretograms based on RLGS (restriction landmark genomic scanning) method; one spot recognition from electrophoretogram and another pairwise matching of spots found such electrophoretograms. Without any human interaction, several thousands a electrophoretogram, including hidden at shoulder large spots, can be identified correctly by applying our algorithm, except only few true-negative false-positive spots. Once locations intensities are recognized automatically, algorithm reliably rapidly identifies equivalent pairs nonlinearly distorted in fully-automatic way, i.e., boring annoying landmarking process is unnecessary. At beginning process, most suitable pair corresponding searched then other identified. With powerful image algorithms, it possible to detect DNA molecular changes as deletions, additions, amplifications or methylations occurring near restriction enzyme cleavage sites means comparing amount electrophoretograms, without visual inspection interaction.