EGBIOIMAGE: A Software Tool for Gel Images Analysis and Hierarchical Clustering

This paper presents a novel software called EGBioImage that is implemented for analyzing gel Electrophoresis images, computing the molecular weights of the bands of the unknown size, detecting the types of the bands, grouping the bands according to their molecular weights and labeled each band with its group number in the gel image and generating the corresponding phylogenetic tree using the data extracted from the analyzed gel image. Students and Researchers working in molecular biology and genetics would benefit greatly from EGBioImage. This software is designed and developed under windows operating system using a C-sharp programming language. It uses Emgu CV platform to extract contours objects from an image that uses the green theorem to detect the bands in each detected lane. It uses polynomial regression to calculate the molecular weights of the bands of unknown size. It uses K-means clustering algorithm to cluster bands according to their molecular weights. It uses each of upgma algorithm and the matching bands amongst lanes and to construct the phylogenetic tree. EGBioImage provides a very friendly Graphical User Interface that guides the user from the step of uploading a gel image toward getting the molecular weights of the bands of unknown size and generating the Phylogenetic Tree. Moreover, EGBioImage provides the user with the ability of processing the gel image using a completely manual processing and automatic processing with the ability of manual correction of lanes and bands "Semi-automatic process" and it is more accurate than comparable software in some respects and the only software that divides the bands into groups and labeled them with numbers in the gel image based on their group by implementing k-means algorithm and these claims are supported with experiments. EGBioImage is suitable for students and researchers who do not have access to commercial software.