Highlights d A deep learning model is trained to predict antibiotics based on structure d Halicin is predicted as an antibacterial molecule from the Drug Repurposing Hub d Halicin shows broad-spectrum antibiotic activities in mice d More antibiotics with distinct structures are predicted from the ZINC15 database

This is a PDF file of a peer-reviewed paper that has been accepted for publication. Although unedited, the content has been subjected to preliminary formatting. Nature is providing this early version of the typeset paper as a service to our authors and readers. The text and figures will undergo copyediting and a proof review before the paper is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers apply.

Historically, multicellular bacterial communities, known as biofilms, have been thought to be held together solely by a self-produced extracellular matrix. Our study identified a novel mechanism maintaining Bacillus subtilis and Mycobacterium smegmatis biofilms—active production of calcite minerals. We studied, for the first time, the effects of mutants defective in biomineralization and calcite formation on biofilm development, resilience and morphology. We demonstrated that an intrinsic rise in carbon dioxide levels within the biofilm is a strong trigger for the initiation of calcite-dependent patterning. The calcite-dependent patterns provide resistance to environmental insults and increase the overall fitness of the microbial community. Our results suggest that it is highly feasible that the formation of mineral scaffolds plays a cardinal and conserved role in bacterial multicellularity.

Changes in expression patterns may occur when organisms are presented with new environmental challenges, for example following migration or genetic changes. To elucidate the mechanisms by which the translational machinery adapts to such changes, we perturbed the tRNA pool of Saccharomyces cerevisiae by tRNA gene deletion. We then evolved the deletion strain and observed that the genetic adaptation was recurrently based on a strategic mutation that changed the anticodon of other tRNA genes to match that of the deleted one. Strikingly, a systematic search in hundreds of genomes revealed that anticodon mutations occur throughout the tree of life. We further show that the evolution of the tRNA pool also depends on the need to properly couple translation to protein folding. Together, our observations shed light on the evolution of the tRNA pool, demonstrating that mutation in the anticodons of tRNA genes is a common adaptive mechanism when meeting new translational demands.DOI: http://dx.doi.org/10.7554/eLife.01339.001