Advancements in text mining and deep learning to improve toxicity prediction
Assessing toxicity is an important step in drug development. Pharmaceutical companies are required to adhere to strict guidelines set forth by regulatory agencies to ensure the safety of the final product. The increasing knowledge available through scientific literature, databases and advances in computational chemistry have enabled the use of in silico tools to predict the genotoxicity potential of compounds. However, their current utilization remains considerably restricted.
Thanks to the vast amount of data being accumulated and advances in computational paradigms, Natural Language Processing (NLP) and machine learning (more recently deep learning) are driving the principled extraction and consolidation of scientific knowledge and using them for driving toxicity prediction. In this talk we will discuss the state of the art of using NLP and AI for genotoxicity prediction and highlight our work in this area.