VirtualFlow Ants—Ultra-Large Virtual Screenings with Artificial Intelligence Driven Docking Algorithm Based on Ant Colony Optimization

Abstract

The docking program PLANTS, which is based on ant colony optimization (ACO) algorithm, has many advanced features for molecular docking. Among them are multiple scoring functions, the possibility to model explicit displaceable water molecules, and the inclusion of experimental constraints. Here, we add support of PLANTS to VirtualFlow (VirtualFlow Ants), which adds a valuable method for primary virtual screenings and rescoring procedures. Furthermore, we have added support of ligand libraries in the MOL2 format, as well as on the fly conversion of ligand libraries which are in the PDBQT format to the MOL2 format to endow VirtualFlow Ants with an increased flexibility regarding the ligand libraries. The on the fly conversion is carried out with Open Babel and the program SPORES. We applied VirtualFlow Ants to a test system involving KEAP1 on the Google Cloud up to 128,000 CPUs, and the observed scaling behavior is approximately linear. Furthermore, we have adjusted several central docking parameters of PLANTS (such as the speed parameter or the number of ants) and screened 10 million compounds for each of the 10 resulting docking scenarios. We analyzed their docking scores and average docking times, which are key factors in virtual screenings. The possibility of carrying out ultra-large virtual screening with PLANTS via VirtualFlow Ants opens new avenues in computational drug discovery.

Efficient Exploration of Chemical Space with Docking and Deep Learning

“…The first such screen was performed by OpenEye Scientific on their Orion cloud computing platform, leveraging vast quantities of CPUs to explicitly predict the pose and score of each compound (detailed experimental outcomes were not provided). Several billion compound screens have been performed with VirtualFlow, a workflow that relies on less-expensive, and arguably less-accurate, methods to triage compounds. ,, Docking an ultralarge library of one billion compounds on 1000 CPUs would take ∼11 days with DOCK3.7 (at 1 s/lig), 300 days with Glide SP (at 30 s/lig), and 173 days with VirtualFlow (at 15 s/lig). This does not account for the cost to prepare the library for screening, which typically includes the generation of low-energy ionization and tautomeric states, as well as stereochemical states for unspecified centers.…”

Recent Developments in Structure-Based Virtual Screening Approaches

“…Software. Software platforms, such as VirtualFlow (3,19,20), that can screen ultra-large ligand libraries using the above-mentioned computational resources, including cloud computing infrastructure, became freely available.…”

“…One of the features of VFVS is that it supports a variety of external docking programs, such as AutoDock Vina (50), Smina (51), QuickVina 2 (52), QuickVina-W (53), Vina-Carb (54), or VinaXB (55). Later GWO Vina and PLANTS were added (19,20). Most of these docking programs have special features, such as blind-docking capabilities (QuickVina-W), or enhanced accuracy for docking carbohydrates (Vina-Carb).…”

Large Scale Docking in the Cloud

“…Since 2015, the number of commercially accessible compounds has grown by over three orders of magnitude, leading to new opportunities to discover new chemistry for new biology. Ultra large scale docking has now been applied to over a dozen targets [1][2][3][4][5][6][7][8][9][10][11] and has discovered new compounds with activities often in the nM and occasionally in the sub-nM range 1,3,6,12 , often leading to molecules with interesting in vivo activities 3,6,12,13 . A challenge to making this approach widely accessible has been the very size of the new libraries, where over 40 billion tangible molecules have been enumerated, and over 4.5 billion of these have structures calculated that are suitable for docking.…”