De novo drug design is a computational approach that generates novel molecular structures from atomic building blocks with no a priori relationships. Conventional methods include structure-based and ligand-based design, which depend on the properties of the active site of a biological target or its known active binders, respectively. Artificial intelligence, including ma-chine learning, is an emerging field that has positively impacted the drug discovery process. Deep reinforcement learning is a subdivision of machine learning that combines artificial neural networks with reinforcement-learning architectures. This method has successfully been em-ployed to develop novel de novo drug design approaches using a variety of artificial networks including recurrent neural networks, convolutional neural networks, generative adversarial networks, and autoencoders. This review article summarizes advances in de novo drug design, from conventional growth algorithms to advanced machine-learning methodologies and high-lights hot topics for further development.

Significance: The catalytic mechanisms of interfacial enzymes acting directly on the interfacial surface of the membrane are notoriously difficult to study experimentally with X-ray crystallography and other biophysical methods. This scientific study is, to our knowledge, the first to highlight similarities and differences in the extraction and binding of a phospholipid molecule into the substrate binding pocket of two human phospholipases A ₂ (PLA ₂ s): the cytosolic Group IVA cPLA ₂ and the calcium-independent Group VIA iPLA ₂ . Molecular dynamics simulations, guided by deuterium exchange experiments, are used to show that pathways to the active sites of these PLA ₂ s are opened upon allosteric interaction with the membrane to facilitate entry of the substrate lipid. These enzymes are involved in various diseases, and understanding their mechanisms will aid in the discovery of therapeutics.

We demonstrate that lipidomics coupled with molecular dynamics reveals unique phospholipase A2 specificity toward membrane phospholipid substrates. We discovered unexpected head-group and acyl-chain specificity for three-major human phospholipases A2. These differences between each enzyme’s specificity coupled with molecular dynamicsbased structural and binding studies revealed unique active site and interfacial surface binding moieties for each enzyme that explains the observed specificity at a hitherto inaccessible structural level. Surprisingly, we discovered that a unique hydrophobic binding site for the cleaved fatty acid dominates each enzyme’s specificity, rather than its catalytic residues and polar head-group binding site. Molecular dynamics simulations revealed the optimal phospholipid binding mode leading to a detailed understanding of the preference of cytosolic phospholipase A2 for cleavage of pro-inflammatory arachidonic acid, calcium-independent phospholipase A2 which is involved in membrane remodeling for cleavage of linoleic acid, and for anti-bacterial secreted phospholipase A2 favoring linoleate, saturated fatty acids, and phosphatidylglycerol.