As graph neural networks are becoming more and more powerful and useful in the field of drug discovery, many pharmaceutical companies are getting interested in utilizing these methods for their own in-house frameworks. This is especially compelling for tasks such as the prediction of molecular properties which is often one of the most crucial tasks in computer-aided drug discovery workflows. The immense hype surrounding these kinds of algorithms has led to the development of many different types of promising architectures and in this review we try to structure this highly dynamic field of AI-research by collecting and classifying 80 GNNs that have been used to predict more than 20 molecular properties using 48 different datasets.

Our study focuses on the reasons for market withdrawals of drugs and drug development project terminations in clinical phases I-III from 1992 to 2002. Over 90% of the market withdrawals were caused by drug toxicity. Hepatotoxicity and cardiovascular toxicity proved to be the major causes for two out of three market withdrawals in the respective time period. In clinical phases I-III 43% of drug development project terminations were due to insufficient efficacy of the investigated compound. The second important issue, which caused one third of the projects to be closed, was toxicity. ADME parameters and economic and other reasons played a minor role. The results of our study indicate that compared with previous studies on this subject, no major improvements have been achieved in the last decade.