Machine Learning-driven Prediction of Cervical Cancer Cell Viability After Treatment With Thymoquinone, Curcumin, and...

Area of Science:

• Oncology
• Pharmacology
• Computational Biology

Background:

• Cervical cancer persists globally, characterized by uncontrolled cell proliferation and apoptosis evasion.
• Phytochemicals like Thymoquinone (TQ) and Curcumin (CUR) are recognized for their safety and therapeutic potential.
• Investigating novel treatments for cervical cancer is crucial due to its prevalence.

Purpose of the Study:

• To evaluate the cytotoxic, anti-proliferative, and apoptotic effects of TQ and CUR in cervical cancer cells.
• To compare the efficacy of TQ and CUR with 5-Fluorouracil (5-FU).
• To develop and validate machine-learning models for predicting cell viability based on phytochemical dose-response data.

Main Methods:

• HeLa cells were treated with varying concentrations of TQ, CUR, and 5-FU.
• Cell viability was assessed using MTT assays, and apoptosis was analyzed via AO/EtBr staining.
• Machine learning models (ANN, SVM, GPR, etc.) were trained on experimental data, with sensitivity analysis performed.

Main Results:

• Higher concentrations of TQ and CUR reduced HeLa cell viability in a dose-dependent manner.
• TQ induced both early and late apoptosis, CUR triggered early apoptosis, and 5-FU caused extensive late apoptosis.
• Artificial Neural Network (ANN) achieved the highest predictive accuracy, identifying IC₅₀ as a key predictor.

Conclusions:

• TQ and CUR exhibit significant anticancer activity against cervical cancer cells.
• Machine learning models, especially ANN, offer a robust framework for predicting phytochemical efficacy.
• This study highlights the potential of integrating AI with experimental data for advancing cervical cancer therapeutics.