High-throughput screening technologies continues to produce large amounts of multiomics data from different populations and cell types for various diseases, such as cancer. However, analysis of such data encounters difficulties due to cancer heterogeneity, further exacerbated by human biological complexity and genomic variability. There is a need to redefine the drug discovery development pipeline, bringing an Artificial Intelligence (AI)-powered informational view that integrates relevant biological information and explores new ways to develop effective anticancer approaches. Here, we show SynPred, an interdisciplinary approach that leverages specifically designed ensembles of AI-algorithms, links omics and biophysical traits to predict synergistic anticancer drug synergy. SynPred exhibits state-of-the-art performance metrics: accuracy – 0.85, precision – 0.77, recall – 0.75, AUROC – 0.82, and F1-score - 0.76 in an independent test set. Moreover, data interpretability was achieved by deploying the most current and robust feature importance approaches. A simple web-based application available online at http://www.moreiralab.com/resources/synpred/ was constructed to predict synergistic anticancer drug combinations requiring only the upload of the two drug SMILES to be tested, allowing easy access by non-expert researchers.

Background: Selectively targeting dopamine receptors has been a persistent challenge in the last years for the development of new treatments to combat the large variety of diseases evolving these receptors. Although, several drugs have been successfully brought to market, the subtype-specific binding mode on a molecular basis has not been fully elucidated. Methods: Homology modeling and molecular dynamics were applied to construct robust conformational models of all dopamine receptor subtypes (D1-like and D2-like receptors). Fifteen structurally diverse ligands were docked to these models. Contacts at the binding pocket were fully described in order to reveal new structural findings responsible for DR sub-type specificity. Results: We showed that the number of conformations for a receptor:ligand complex was associated to unspecific interactions > 2.5 Å and hydrophobic contacts, while the decoys binding energy was influenced by specific electrostatic interactions. Known residues such as 3.32Asp, the serine microdomain and the aromatic microdomain were found interacting in a variety of modes (HB, SB, π-stacking). Purposed TM2-TM3-TM7 microdomain was found to form a hydrophobic network involving Orthosteric Binding Pocket (OBP) and Secondary Binding Pocket (SBP). T-stacking interactions revealed as especially relevant for some large ligands such as apomorphine, risperidone or aripiprazole. Conclusions: This in silico approach was successful in showing known receptor-ligand interactions as well as in determining unique combinations of interactions, key for the design of more specific ligands.