Recovering drug-induced apoptosis subnetwork from Connectivity Map data

Area of Science:

• Computational Biology
• Cancer Research
• Systems Biology

Background:

• The Connectivity Map (CMAP) project aims to link anticancer compounds to genes and treatments.
• Understanding drug-induced apoptosis is crucial for cancer therapy and tumor treatment.
• Apoptosis pathways are central to the efficacy of many anticancer drugs.

Purpose of the Study:

• To identify drug-induced cellular pathways in breast cancer using CMAP data.
• To reconstruct the apoptosis network and identify key genes involved in drug response.
• To elucidate the roles of specific genes in intrinsic and extrinsic apoptosis pathways.

Main Methods:

• Applied Gaussian Bayesian network modeling to CMAP microarray data from MCF7 breast cancer cells.
• Focused on 13 differentially expressed apoptotic genes across drug-perturbed samples.
• Validated inferred network interactions against existing literature.

Main Results:

• Identified apoptosis as a major drug-induced cellular pathway.
• Reconstructed a high-confidence apoptosis subnetwork with 9 validated interactions.
• Identified BCL2L11 and PMAIP1 as key players in the intrinsic pathway, and TAXBP1 and TNFAIP3 in the extrinsic pathway.
• Highlighted BCL2L11 and TNFAIP3 as potential "gateway" genes in drug-induced apoptosis.

Conclusions:

• Gaussian Bayesian networks can effectively model drug-induced cellular pathways.
• The inferred apoptosis network provides insights into cancer cell response to chemical stimuli.
• The study identifies critical genes for intrinsic and extrinsic apoptosis, aiding in understanding cancer treatment mechanisms.