Combining RNAi Imaging Data With Genomic Data For Gene Interaction Network Construction

A computer program product for constructing a gene interaction network, the computer program product comprising a computer readable storage device having program code embodied therewith, the program code executable by a processor to:

• for each pair of genes in a first set of gene expression data, compute a first gene interaction measure between a pair of genes, the computation based upon gene expression data, and cluster genes from the first set based on the expression data;
• for each pair of genes in a second set of a multi-dimensional image stack having image data depicting groups of cells or intra-cellular structures associated with a second set of genes, compute a gene relationship measure based upon the cellular images using a morphological similarity measure between corresponding groups of cells;
• construct an initial graph by combining the first set of gene expression data with the image data of the second set, each node in the graph associated with one of the genes belonging to one of the first and second sets;
• and draw edges between nodes, the edges based on membership of gene clusters or morphological distance for a corresponding gene knockdown image stack;
• expand connectivity among the nodes in the graph using network propagation, the expanded connectivity forming a final graph;
• and
• verify the expanded connectivity by sampling biological experiments eliciting relationships between the corresponding genes as represented by the nodes and edges in the final graph.

A system comprising:

• a processor in communication with memory and data storage;
• a functional unit in communication with the memory, the function unit to construct a gene interaction network using gene expression analysis, the function unit comprising:
• a first gene manager to compute a first gene interaction measure between a pair of genes, the computation based on gene expression data, and the first gene manager to cluster the genes based on the expression data;
• for each pair of genes in a second set, a second gene manager to compute a gene relationship measure based upon image data using morphological similarity measure between corresponding groups of cells;
• a construction manager in communication with the first and second gene managers, the construction manager to construct an initial graph by combining the gene expression data and the gene image data, each node in the graph associated with one of the genes belonging to one of the first and second sets, and to draw edges between nodes, the edges based on membership of gene clusters or morphological distance for a corresponding gene knockdown image stack;
• the construction manager to expand connectivity among the nodes in the graph using network propagation, the expanded connectivity forming a final graph;
• and
• a director in communication with the construction manager, the director to verify the expanded connectivity by sampling designed biological experiments eliciting the relationships between the corresponding genes as represented by the nodes and edges in the final graph.