title: “Toxicology Prediction” excerpt_separator: “”
A machine learning model trained on Toxicity data [1] for over 8,000 molecules in attempt to find a relationship between molecular structure and relative toxicity. This project was then deployed along with a Streamlit web app for user interaction.
💡 Inspiration
Drug discovery remains one of the most resource‑intensive challenges in modern biomedical research. According to the American Society for Biochemistry and Molecular Biology (ASBMB) [2], approximately 90% of drug candidates fail during clinical testing, with molecular toxicity accounting for nearly one‑third of these failures. These setbacks translate into billions of dollars of substantial financial losses and slow development of urgently needed therapeutics.
Understanding and predicting toxicity earlier in the drug development pipeline could dramatically reduce these costs. Traditional experimental approaches, while essential, are slow and expensive. Computational methods offer a promising complement, but many existing models struggle to generalize across diverse chemical structures or provide interpretable insights into why a compound may be harmful.
Our project addresses this challenge by:
- Developing a machine learning model that predicts toxicity directly from molecular structure
- Using Structure-Activity Relationship (SAR) principles to justify predictions based on similarity to known toxic compounds
- Building an interactive Streamlit app to make the model accessible for exploration and teaching
⚛️ Innovation
By integrating machine learning with chemical domain knowledge, this project explores how computational tools can support safer, more efficient drug development and help reduce the high failure rates that currently burden the pharmaceutical industry.
📚 References
[1] National Center for Advancing Translational Sciences, “Toxicology in the 21st Century (Tox21)”, [Online]. Available: https://tox21.gov. Accessed: Feb. 20, 2025.
[2] D. Sun, “90% of drugs fail clinical trials”, ASBMB Today, Mar. 12, 2022. [Online]. Available: https://www.asbmb.org/asbmb-today/opinions/031222/90-of-drugs-fail-clinical-trials. Accessed: Feb. 20, 2025.