Eurofins DiscoveryAI, Eurofins Panlabs, Inc., Saint Charles, MO 63304, USA.
Future Med Chem. 2024 Apr;16(7):587-599. doi: 10.4155/fmc-2024-0007. Epub 2024 Feb 19.
To prioritize compounds with a higher likelihood of success, artificial intelligence models can be used to predict absorption, distribution, metabolism, excretion and toxicity (ADMET) properties of molecules quickly and efficiently. Models were trained with BioPrint database proprietary data along with public datasets to predict various ADMET end points for the SAFIRE platform. SAFIRE models performed at or above 75% accuracy and 0.4 Matthew's correlation coefficient with validation sets. Training with both proprietary and public data improved model performance and expanded the chemical space on which the models were trained. The platform features scoring functionality to guide user decision-making. High-quality datasets along with chemical space considerations yielded ADMET models performing favorably with utility in the drug discovery process.
为了优先选择成功率更高的化合物，可以使用人工智能模型快速有效地预测分子的吸收、分布、代谢、排泄和毒性（ADMET）性质。模型是使用 BioPrint 数据库的专有数据以及公共数据集进行训练的，以预测 SAFIRE 平台的各种 ADMET 终点。SAFIRE 模型在验证集上的准确率达到或超过 75%，马修相关系数为 0.4。使用专有数据和公共数据进行训练可以提高模型性能，并扩展模型训练所涉及的化学空间。该平台具有评分功能，可指导用户决策。高质量的数据集以及对化学空间的考虑，使得 ADMET 模型具有良好的性能，可用于药物发现过程。
