Skripsi

Prediksi Penghambatan Senyawa Dalam Ricinus communis Terhadap NF-κB pada Sel Kanker Payudara MCF-7 dengan Pendekatan Machine Learning dan Kemoinformatik = Prediction of Inhibitory Compounds from Ricinus communis Against NF-κB in MCF-7 Breast Cancer Cells Using Machine Learning and Chemoinformatics Approaches.

Latar Belakang Kanker payudara merupakan kanker paling sering pada wanita. NF-κB berperan penting dalam proliferasi, metastasis, dan resistensi terapi. Ricinus communis diketahui mengandung senyawa bioaktif dengan potensi antikanker, namun aktivitas spesifik terhadap NF-κB masih belum banyak diteliti. Metode Penelitian analitik observasional ini dilakukan secara in silico menggunakan perangkat lunak DataWarrior. Struktur senyawa Ricinus communis diperoleh dari PubChem, sementara data inhibitor sel MCF-7 dan NF-κB berasal dari ChEMBL dan RSCB PDB. Model prediksi nilai IC50 divalidasi dengan metode machine learning (SVM Regression dan PLS) menggunakan deskriptor SkelSpheres dan FragFp, dengan parameter Q2 dan koefisien korelasi Pearson. Hasil Model terbaik untuk sel MCF-7 adalah SkelSpheres–PLS (Q2=0.903), sedangkan untuk NF-κB adalah SkelSpheres–SVM Regression (Q2=0.711). Sebanyak delapan senyawa teridentifikasi memiliki kategori aktivitas excellent hingga good pada kedua target biologis tersebut, yaitu: pyridine, ricinine, rutin, dehydromatricaria ester, fucosterol, kaempferol, voleneol, dan ricinitin. Kesimpulan Prediksi nilai IC50 berhasil menunjukkan bahwa Ricinus communis mengandung senyawa bioaktif potensial sebagai inhibitor NF-κB pada sel kanker payudara MCF-7 secara komputasional, sehingga dapat dikembangkan sebagai kandidat terapi alternatif kanker payudara.
Kata Kunci: In silico, Kanker Payudara, MCF-7, NF-κB, Ricinus communis


Introduction Breast cancer is the most common malignancy in women. NF-κB activation plays a critical role in proliferation, metastasis, and therapy resistance. Ricinus communis contains bioactive compounds with anticancer potential, but its specific activity against NF-κB remains underexplored. Method This observational analytic study was conducted in silico using DataWarrior. Ricinus communis compounds were retrieved from PubChem, while inhibitors of MCF-7 cells and NF-κB were collected from ChEMBL and RSCB PDB. IC50 prediction models were validated using machine learning (SVM Regression and PLS) with SkelSpheres and FragFp descriptors, evaluated by Q2 and Pearson’s correlation coefficient. Results The best model for MCF-7 was SkelSpheres–PLS (Q²=0.903, r=0.950), while for NF-κB it was SkelSpheres–SVM Regression (Q²=0.711, r=0.842). Eight compounds including pyridine, ricinine, rutin, dehydromatricaria ester, fucosterol, kaempferol, voleneol, and ricinitin showed excellent to good IC50 activity on both MCF-7 and NF-κB. Conclusion IC50 prediction successfully demonstrated that Ricinus communis contains potential bioactive compounds as NF-κB inhibitors in MCF-7 breast cancer cells computationally, indicating their potential development as alternative therapeutic candidates for breast cancer.
Keywords: Breast Cancer, In silico, MCF-7, NF-κB, Ricinus communis