Prospecting for antifungal molecules: Marine bacteria produce a plethora of natural products that often have unusual chemical structures and corresponding reactivity, which sometimes translate into a valuable biological function. Zhang et al. used a metabolomic screen to zero in on microbial strains from the microbiome of a sea squirt that produces a high diversity of chemical structures. They then screened these molecules for inhibition of fungi (see the Perspective by Cowen). A polycyclic molecule dubbed turbinmicin possessed potent antifungal activity against the multidrug-resistant fungal pathogens Candida auris and Aspergillus fumigatus . Preliminary mechanism-of-action and mouse toxicity studies suggest that this molecule works though a fungus-specific pathway and is well tolerated at therapeutic doses. Science , this issue p. 974 ; see also p. 906

Virtual (computational) high-throughput screening provides a strategy for prioritizing compounds for experimental screens, but the choice of virtual screening algorithm depends on the dataset and evaluation strategy. We consider a wide range of ligand-based machine learning and docking-based approaches for virtual screening on two protein-protein interactions, PriA-SSB and RMI-FANCM, and present a strategy for choosing which algorithm is best for prospective compound prioritization. Our workflow identifies a random forest as the best algorithm for these targets over more sophisticated neural network-based models. The top 250 predictions from our selected random forest recover 37 of the 54 active compounds from a library of 22,434 new molecules assayed on PriA-SSB. We show that virtual screening methods that perform well on public datasets and synthetic benchmarks, like multi-task neural networks, may not always translate to prospective screening performance on a specific assay of interest.