![]() For comparison, a subset of the inactive compounds was docked in the same way. Compounds that inhibited growth were taken as active, and each active compound was computationally docked with each of 296 AlphaFold2-predicted E. coli essential protein structures. To define our chemical space of interest, we performed high-throughput screens of growth inhibition against wild-type E. coli. In addition to inspiring further studies that expand on the interactions discovered in this way, these experiments could be used to benchmark the performance of our modeling platform and reveal the prediction accuracy possible with AlphaFold2-enabled molecular docking simulations. As computational docking approaches are known to predict many false positives (Adeshina et al, 2020), the predicted protein-ligand interactions could be experimentally interrogated, in part, using biochemical assays that measure enzymatic activity, with binding interactions supported by enzymatic inhibition. We hypothesized that such an approach could enrich for true protein-ligand interactions from the large, combinatorial space of all possible interactions between antibacterial compounds and essential proteins. Here, we reasoned that the recent release of the AlphaFold2 database of protein structure predictions (Jumper et al, 2021 Varadi et al, 2022) could enable reverse docking approaches that span Escherichia coli's essential proteome, allowing for the extensive prediction of binding targets of antibacterial compounds (Fig 1A). Although versatile, reverse docking requires a priori knowledge of the protein structures of interest, and its application to drug-target identification has been limited by the number and quality of target protein structures (Chen & Zhi, 2001 Kharkar et al, 2014 Lee et al, 2016). While docking has been used to enrich for potential hit compounds that bind pre-specified proteins in “one target, many compounds” approaches, the process of “reverse docking,” in which a small molecule is docked across different potential protein targets, leverages docking to discover binding partners and drug mechanisms of action (Kharkar et al, 2014 Lee et al, 2016). In molecular docking, ligand binding poses within a targeted binding site of a protein are computationally modeled using scoring functions, and poses are optimized to provide structural information and activity predictions in the form of thermodynamic binding affinities. Molecular docking, in particular, has proven versatile for identifying protein-ligand interactions and drug mechanisms of action. Various approaches to identifying molecular drug targets have been developed, including those based on biochemical assays, genetic interactions, and molecular docking (Kitchen et al, 2004 Schenone et al, 2013). SynopsisĪ major challenge in drug discovery is the identification of drug-target interactions. This work indicates that advances in modeling protein-ligand interactions, particularly using machine learning-based approaches, are needed to better harness AlphaFold2 for drug discovery. ![]() We demonstrate that rescoring of docking poses using machine learning-based approaches improves model performance, resulting in average auROCs as large as 0.63, and that ensembles of rescoring functions improve prediction accuracy and the ratio of true-positive rate to false-positive rate. We confirm extensive promiscuity, but find that the average area under the receiver operating characteristic curve (auROC) is 0.48, indicating weak model performance. ![]() We benchmark model performance by measuring enzymatic activity for 12 essential proteins treated with each antibacterial compound. Here, we combine AlphaFold2 with molecular docking simulations to predict protein-ligand interactions between 296 proteins spanning Escherichia coli's essential proteome, and 218 active antibacterial compounds and 100 inactive compounds, respectively, pointing to widespread compound and protein promiscuity. Computational docking approaches have been widely used to predict drug binding targets yet, such approaches depend on existing protein structures, and accurate structural predictions have only recently become available from AlphaFold2. Efficient identification of drug mechanisms of action remains a challenge. ![]()
0 Comments
Leave a Reply. |
AuthorWrite something about yourself. No need to be fancy, just an overview. ArchivesCategories |