ProSim: A Method for Prioritizing Disease Genes Based on Protein Proximity and Disease Similarity

Area of Science:

• Bioinformatics
• Computational Biology
• Genetics

Background:

• Predicting disease genes is a significant challenge in bioinformatics.
• Network-based approaches, specifically utilizing protein-protein interaction (PPI) networks, offer a promising avenue for tackling this challenge.
• Improving accuracy requires integrating disease similarity with protein proximity to known disease genes within PPI networks.

Purpose of the Study:

• To propose a novel algorithm, ProSim (proximity disease similarity algorithm), for prioritizing disease genes.
• To enhance disease gene prediction accuracy by incorporating both disease similarity and protein proximity within PPI networks.
• To evaluate the performance of ProSim across six diverse case studies, including major cancers and neurodegenerative diseases.

Main Methods:

• Development of the ProSim algorithm, which integrates disease similarity and protein proximity metrics.
• Application of ProSim to six disease case studies: prostate cancer, Alzheimer's disease, diabetes mellitus type 2, breast cancer, colorectal cancer, and lung cancer.
• Rigorous evaluation using leave-one-out cross-validation, mean enrichment, tenfold cross-validation, and ROC curves, comparing ProSim against established methods like PRINCE, RWR, and DADA.

Main Results:

• The ProSim algorithm demonstrated superior performance in prioritizing disease genes compared to existing methods.
• Quantitative evaluations using cross-validation and ROC curves confirmed the effectiveness of ProSim.
• Case studies across various diseases validated the algorithm's robustness and accuracy.

Conclusions:

• The ProSim algorithm represents a significant advancement in disease gene prediction.
• Integrating disease similarity and protein proximity is crucial for improving the accuracy of network-based disease gene prioritization.
• ProSim offers a more effective tool for identifying potential disease genes, aiding in genetic research and therapeutic development.