Context-specific functional module based drug efficacy prediction.

Area of Science:

• Computational biology
• Pharmacogenomics
• Systems biology

Background:

• Personalized medicine requires accurate drug efficacy evaluation.
• Clinical trials are costly and time-consuming.
• Current computational models lack tissue-specific interaction data.

Purpose of the Study:

• To develop a cell line-specific computational model for predicting drug efficacy.
• To identify functional modules associated with drug sensitivity.
• To improve drug repositioning and overcome drug resistance.

Main Methods:

• Developed cell line-specific functional modules (gene clusters with similar biological functions).
• Utilized enriched scores of these modules as cell line-specific features.
• Employed linear regression for drug efficacy prediction and assessed performance using leave-one-out cross-validation (LOOCV).

Main Results:

• Achieved accurate drug efficacy prediction using cell line-specific functional modules.
• Outperformed the elastic net model in prediction accuracy.
• Identified drug sensitivity-associated functions for lapatinib, erlotinib, raloxifene, tamoxifen, and gefitinib.

Conclusions:

• The developed model enables cell line-specific drug efficacy prediction.
• Identified functions can guide drug repositioning strategies.
• Provides insights for developing secondary drugs to combat drug resistance.