De novo design of dual-target JAK2, SMO inhibitors based on deep reinforcement learning, molecular docking and...

Area of Science:

• Oncology
• Medicinal Chemistry
• Computational Biology

Background:

• Triple-negative breast cancer (TNBC) and HER2-positive breast cancer are aggressive subtypes with limited therapeutic options.
• Existing therapies for HER2-positive breast cancer often face resistance, while TNBC lacks effective treatments.
• JAK2 and SMO signaling pathways are implicated in the progression of these breast cancer types.

Purpose of the Study:

• To identify novel small molecule compounds that can inhibit both JAK2 and SMO.
• To leverage deep reinforcement learning and molecular docking for drug discovery.
• To develop potential dual-target inhibitors for aggressive breast cancers.

Main Methods:

• Deep reinforcement learning was employed to analyze existing JAK2 and SMO inhibitors.
• A novel library of small molecule compounds was generated.
• Molecular docking, molecular dynamics simulations, and binding free energy calculations were used for screening and validation.

Main Results:

• Seven compounds were identified as potential dual inhibitors of JAK2 and SMO.
• The top three compounds demonstrated stable binding to both target proteins.
• Compounds A8976 and A10625 exhibited favorable binding properties, suggesting potential as dual-target inhibitors.

Conclusions:

• Deep reinforcement learning and computational screening can effectively identify novel dual-target inhibitors.
• Compounds A8976 and A10625 represent promising candidates for further development against TNBC and HER2-positive drug-resistant breast cancer.
• Targeting both JAK2 and SMO pathways offers a potential therapeutic strategy for aggressive breast cancers.