About this Research Topic
Nowadays, Machine Learning tools are routinely used for generating/designing molecules for ideation and funneled through an array of ADME/T endpoints before the chemist decides to make any compounds. We have seen a shift in the industry of using ML models to guide us on which compounds have the best chance of success rather than to inform us which one of our ideas has a certain predicted property. One emerging challenge as we get more models that we rely on is how best to combine these diverse endpoints into a score that can be used to rank or prioritize compounds.
This Research Topic is dedicated to the broad topic of how to best utilize ADME/T models with an emphasis on:
1. Developing the “best” Machine Learning model for a specific endpoint e.g., permeability, transporters, hERG liability, or hepatotoxicity.
2. Interpreting Machine Learning predictions e.g., what functional group can be replaced with improved permeability, reduced hERG liability, or hepatotoxicity.
3. Estimating how reliable a prediction is e.g., using the applicability domain or estimating predicted errors and uncertainties.
4. Assessing ADME/T using a multi-parameter object score e.g., how to calculate the score for multiple ADME/T endpoints.
We accept different article types including Original Research, Reviews, and Perspectives. A full list of accepted article types, including descriptions, can be found at this link.
Keywords: Machine Learning, Drug Discovery, ADME, Tox, New Compounds Discovery, Cheminformatics, Pharmacokinetics Modeling
Important Note: All contributions to this Research Topic must be within the scope of the section and journal to which they are submitted, as defined in their mission statements. Frontiers reserves the right to guide an out-of-scope manuscript to a more suitable section or journal at any stage of peer review.