De novo 3D models of SARS-CoV-2 RNA elements and small-molecule-binding RNAs to guide drug discovery

Ramya Rangan; Andrew M. Watkins; Wipapat Kladwang; Rhiju Das

Given these limitations in de novo modeling, we felt that it was important to provide analogous models of RNAs of known structure. Virtual screening approaches appear poised to make good use of computational models of RNA, but have so far, to our knowledge, not been used on de novo predicted models. To provide benchmark structural ensembles for such efforts, we have therefore used the same Rosetta-FARFAR2 modeling method and model selection procedure for a variety of RNA riboswitches with known small-molecule ligands. We hope this FARFAR2-Apo-riboswitch data set provides an opportunity for testing the accuracy of virtual screening approaches that use FARFAR2 model sets. With the current need for SARS-CoV-2 antiviral discovery, we believe this is . CC-BY 4.0 International license author/funder. It is made available under a The copyright holder for this preprint (which was not peer-reviewed) is the . https://doi.org/10.1101/2020.04.14.041962 doi: bioRxiv preprint an opportune time to explore and evaluate new approaches for virtual screening of small-molecule drug candidates that target structured RNA.