RESUMO
Accurate prediction of protein-ligand binding affinity (PLA) is important for drug discovery. Recent advances in applying graph neural networks have shown great potential for PLA prediction. However, existing methods usually neglect the geometric information (i.e. bond angles), leading to difficulties in accurately distinguishing different molecular structures. In addition, these methods also pose limitations in representing the binding process of protein-ligand complexes. To address these issues, we propose a novel geometry-enhanced mid-fusion network, named GEMF, to learn comprehensive molecular geometry and interaction patterns. Specifically, the GEMF consists of a graph embedding layer, a message passing phase, and a multi-scale fusion module. GEMF can effectively represent protein-ligand complexes as graphs, with graph embeddings based on physicochemical and geometric properties. Moreover, our dual-stream message passing framework models both covalent and non-covalent interactions. In particular, the edge-update mechanism, which is based on line graphs, can fuse both distance and angle information in the covalent branch. In addition, the communication branch consisting of multiple heterogeneous interaction modules is developed to learn intricate interaction patterns. Finally, we fuse the multi-scale features from the covalent, non-covalent, and heterogeneous interaction branches. The extensive experimental results on several benchmarks demonstrate the superiority of GEMF compared with other state-of-the-art methods.
Assuntos
Redes Neurais de Computação , Ligação Proteica , Proteínas , Proteínas/química , Proteínas/metabolismo , Ligantes , Algoritmos , Biologia Computacional/métodos , Descoberta de Drogas/métodosRESUMO
Glioma is the leading cancer of the central nervous system (CNS). The efficacy of glioma treatment is significantly hindered by the presence of the blood-brain barrier (BBB) and blood-brain tumour barrier (BBTB), which prevent most drugs from entering the brain and tumours. Hence, we established a novel drug delivery nanosystem of brain tumour-targeting that could self-assemble the method using an amphiphilic Zein protein isolated from corn. Zein's amphiphilicity prompted it to self-assembled into NPs, efficiently containing TMZ. This allowed us to investigate temozolomide (TMZ) for glioblastoma (GBM) treatment. To construct TMZ-encapsulated NPs (TMZ@RVG-Zein NPs), the NPs' Zein was clocked to rabies virus glycoprotein 29 (RVG29). To verify that the NPs could penetrate the BBB and precisely target and kill the GBM cancer cell line, in vitro studies were performed. The process of NPs penetrating cancer cell membranes was investigated using enzyme-linked immunosorbent assays (ELISAs) to measure the expressions of nicotinic acetylcholine receptors (nAChRs) on the U87 cell line. Therefore, effective targeted brain cancer treatment is possible by forming NP clocks, a cell-penetrating natural Zein protein with an RVG29. These NPs can penetrate the blood-brain barrier (BBB) and enter the glioblastoma (U87) cell line to release TMZ. These NPs have a distinct cocktail of biocompatibility and brain-targeting abilities, making them ideal for involving brain diseases.