Speaker
Description
Cytochrome P450-mediated mechanisms play a significant role in drug metabolism, with estimates indicating that 75% of commercially available drugs are metabolized by only 6 of the 57 human CYP enzymes [1]. These heme-containing cytochromes P450 display a broad spectrum of ligand specificity [2], a property enabled in part by the flexible protein region adjacent to the binding site. Consequently, this flexibility challenges docking methodologies that rely on a rigid protein approximation, resulting in reduced reliability [2],[3]. This work [4] evaluates open-source docking engines in a high-throughput manner on a dataset of 128 ligands. We employ four notably different engines: RosettaFold-AllAtoms [5] (rfaa), GalaxyDock2 HEME [6] (gdock), AutoDock VINA [7,8] and GNINA [9]. Redocking and crossdocking simulations were employed to assess the docking protocols. In redocking, the ligand is docked into a protein that possesses the optimal binding conformation, whereas in crossdocking, the ligand is docked into a folded protein lacking the binding-specific conformational information, thereby necessitating adjustments. Consequently, crossdocking offers a more realistic representation of practical applications. To compare the engines, we introduced system-specific metrics focused on the heme iron atom and evaluated model performance using the mean absolute error. We report significant improvement for flexible rfaa full sequence prediction and during the presentation, we will outline our simulation workflow, elaborate on our system-specific metrics, benchmark results, and conclude with a discussion on the current challenges.
[1] Brändén, G., Sjögren, T., Schnecke, V., & Xue, Y. (2014). Structure-based ligand design to overcome CYP inhibition in drug discovery projects. Drug Discovery Today, 19(7).
[2] Lokwani, D.K.; Sarkate, A.P.; Karnik, K.S.; Nikalje, A.P.G.; Seijas, J.A. Structure-Based Site of Metabolism (SOM) Prediction of Ligand for CYP3A4 Enzyme: Comparison of Glide XP and Induced Fit Docking (IFD). Molecules 2020, 25, 1622.
[3] Matthew R. Masters, Amr H. Mahmoud, Yao Wei, and Markus A. Lill, Deep Learning Model for Efficient Protein–Ligand Docking with Implicit Side-Chain Flexibility, Journal of Chemical Information and Modeling 2023 63 (6), 1695-1707
[4] D. Suwała and E. Hruška, “The wins and failures of current docking methods tested on the flexible active site of cytochromes P450,” Nov. 29, 2024.
[5] Rohith Krishna et al. Generalized biomolecular modeling and design with RoseTTAFold All-Atom, Science384, eadl2528 (2024).
[6] C. Lee, J. Yang, S. Kwon, C. Seok. GalaxyDock2-HEME: Protein–ligand docking for heme proteins J. Comput. Chem. 2023, 44(14), 1369
[7] Eberhardt, J., Santos-Martins, D., Tillack, A.F., Forli, S. (2021). AutoDock Vina 1.2.0: New Docking Methods, Expanded Force Field, and Python Bindings. Journal of Chemical Information and Modeling.
[8] Trott, O., & Olson, A. J. (2010). AutoDock Vina: improving the speed and accuracy of docking with a new scoring function, efficient optimization, and multithreading. Journal of computational chemistry, 31(2), 455-461.
[9] McNutt, A.T., Francoeur, P., Aggarwal, R. et al. GNINA 1.0: molecular docking with deep learning. J Cheminform 13, 43 (2021).
