Decoding Surface Fingerprints for Protein-Ligand Interactions

Abstract

Small molecules have been the preferred modality for drug development and therapeutic interventions. This molecular format presents a number of advantages, e.g. long half-lives and cell permeability, making it possible to access a wide range of therapeutic targets. However, finding small molecules that engage “hard-to-drug” protein targets specifically and potently remains an arduous process, requiring experimental screening of extensive compound libraries to identify candidate leads. The search continues with further optimization of compound leads to meet the required potency and toxicity thresholds for clinical applications. Here, we propose a new computational workflow for high-throughput fragment-based screening and binding affinity prediction where we leverage the available protein-ligand complex structures using a state-of-the-art protein surface embedding framework (dMaSIF). We developed a tool capable of finding suitable ligands and fragments for a given protein pocket solely based on protein surface descriptors, that capture chemical and geometric features of the target pocket. The identified fragments can be further combined into novel ligands. Using the structural data, our ligand discovery pipeline learns the signatures of interactions between surface patches and small pharmacophores. On a query target pocket, the algorithm matches known target pockets and returns either potential ligands or identifies multiple ligand fragments in the binding site. Our binding affinity predictor is capable of predicting the affinity of a given protein-ligand pair, requiring only limited information about the ligand pose. This enables screening without the costly step of first docking candidate molecules. Our framework will facilitate the design of ligands based on the target’s surface information. It may significantly reduce the experimental screening load and ultimately reveal novel chemical compounds for targeting challenging proteins.

Protein–protein interfaces in molecular glue-induced ternary complexes: classification, characterization, and prediction

“…Experimentally, this is achieved by performing high throughput screens and hit validation followed by lead optimization. Computational approaches such as binding pocket identification [127][128][129][130][131][132] , protein-ligand docking [133][134][135][136][137][138] , virtual screening [139][140][141] , and ligand binding free energy calculations 94,95,142,143 can also be applied and have been shown to be helpful in this area. With a combination of computational and experimental approaches, rational discovery of molecular glues may indeed be feasible, even though the best and the most effective solution to this task is yet to be determined.…”

Single-cell atlas of a non-human primate reveals new pathogenic mechanisms of COVID-19

“…AlphaFold [223], AlphaFold2 [128], RosettaFold [9], RosettaFold2 [10], RFAA [145], EigenFold [126], RFdiffusion [267], Chroma [110], ESMFold [157], HelixFold-Single [60] Protein Co-Design Generative Chroma [110], RFdiffusion [267], PROTSEED [226] Pretraining Mixed ProtTrans [57], ProtGPT2 [63], GearNet [303], PromptProtein [264], ProFSA [71], DrugCLIP [72], ESM-1b [208], ESM2 [157], Guo et al [88] Mol + Mol Linker Design Generative DiffLinker [107], DeLinker [108], 3DLinker [103] Chemical Reaction Generative OA-ReactDiff [49], TSNet [112] Mol + Protein Ligand Binding Affinity Prediction Non-Generative TargetDiff [86], MaSIF [68], GET [143], ProtNet [257], HGIN [304], BindNet [62] Protein-Ligand Docking Mixed EquiBind [232], DiffDock [41], TankBind [170], DESERT [169], FABind [193] Pocket-Based Mol Sampling Mixed Pocket2Mol [194], TargetDiff [86], SBDD [172], FLAG [302] Protein + Protein Protein Interface Prediction Non-Generative DeepInteract <...…”

“…Mol + Mol ZINC [234] 727K Linker Design 3DLinker [103] CASF [235] 0.28K Linker Design DeLinker [108] GEOM [8] 450K Linker Design DiffLinker [107] SN2-TS [112] 0.11K Chemical Reaction TSNet [112] Transition1x [217] 9.6M Chemical Reaction OA-ReactDiff [49] Mol + Protein…”

Exploring Chemical Reaction Space with Machine Learning Models: Representation and Feature Perspective

“…116,117 Besides, graph edits are also applied for the single-step retrosynthesis task, where target molecules are transformed into reactants by applying an iterative refining procedure. For example, Igashov et al 118 generate the reactants for the target molecule by applying diffusion models on the graph level. Transition kernels that bridge the distribution between product structures and reactant structures alter the connectivity and atom types directly on the product structure.…”