Machine Learning Deciphered Molecular Mechanistics with Accurate Kinetic and Thermodynamic Prediction.

Time-lagged independent component analysis (tICA) and the Markov state model (MSM) have been extensively employed for extracting conformational dynamics and kinetic community networks from unbiased trajectory ensembles. However, these techniques may not be the optimal choice for elucidating transition mechanisms within low-dimensional representations, especially for intricate biosystems. Unraveling the association mechanism in such complex systems always necessitates permutations of several essential independent components or collective variables, a process that is inherently obscure and may require empirical knowledge for selection. To address these challenges, we have implemented an integrated unsupervised dimension reduction model: uniform manifold approximation and projection (UMAP) with hierarchy density-based spatial clustering of applications with noise (HDBSCAN). This approach effectively generates low-dimensional configurational embeddings. The hierarchical application of this architecture, in conjunction with MSM, reveals global kinetic connectivity while identifying local conformational states. Consequently, our methodology establishes a multiscale mechanistic elucidation framework. Leveraging the benefits of the uniform sample distribution and a denoising approach, our model demonstrates robustness in preserving global and local data structures compared to traditional dimension reduction methods in the field of MD analysis area. The interpretability of hyperparameter selection and compatibility with downstream tasks are cross-validated across various simulation data sets, utilizing both computational evaluation metrics and experimental kinetic observables. Furthermore, the predicted Mcl1-BH3 association kinetics (0.76 s-1) is in close agreement with surface plasmon resonance experiments (0.12 s-1), affirming the plausibility of the identified pathway composed of representative conformations. We anticipate that the devised workflow will serve as a foundational framework for studying recognition patterns in complex biological systems. Its contributions extend to the exploration of protein functional dynamics and rational drug design, offering a potent avenue for advancing research in these domains.

Computational drug development for membrane protein targets.

The application of computational biology in drug development for membrane protein targets has experienced a boost from recent developments in deep learning-driven structure prediction, increased speed and resolution of structure elucidation, machine learning structure-based design and the evaluation of big data. Recent protein structure predictions based on machine learning tools have delivered surprisingly reliable results for water-soluble and membrane proteins but have limitations for development of drugs that target membrane proteins. Structural transitions of membrane proteins have a central role during transmembrane signaling and are often influenced by therapeutic compounds. Resolving the structural and functional basis of dynamic transmembrane signaling networks, especially within the native membrane or cellular environment, remains a central challenge for drug development. Tackling this challenge will require an interplay between experimental and computational tools, such as super-resolution optical microscopy for quantification of the molecular interactions of cellular signaling networks and their modulation by potential drugs, cryo-electron microscopy for determination of the structural transitions of proteins in native cell membranes and entire cells, and computational tools for data analysis and prediction of the structure and function of cellular signaling networks, as well as generation of promising drug candidates.

The structure and function of olfactory receptors.

Olfactory receptors (ORs) form the most important chemosensory receptor family responsible for our sense of smell in the nasal olfactory epithelium. This receptor family belongs to the class A G protein-coupled receptors (GPCRs). Recent research has indicated that ORs are involved in many nonolfactory physiological processes in extranasal tissue, such as the brain, pancreas, and testes, and implies the possible role of their dysregulation in various diseases. The recently released structures of OR51E2 and consensus OR52 have also unveiled the uniqueness of ORs from other class A GPCR members. In this review, we discuss these recent developments and computational modeling efforts toward understanding the structural properties of unresolved ORs, which could guide potential future OR-targeted drug discovery.

Machine learning for data-driven design of high-safety lithium metal anode.

Here, we present a protocol for developing an inorganic-organic hybrid interphase layer using the self-assembled monolayers technique to enhance the surface of the lithium metal anode. We describe steps for extracting organic molecules from open-sourced databases and calculating their microscopic properties. We then detail procedures for developing a machine learning model for predicting the ionic diffusion barrier and preparing the inputs for prediction. This protocol enables a cost-effective workflow to identify promising self-assembled monolayers with exceptional performance. For complete details on the use and execution of this protocol, please refer to Zhang et al. (2023).1.

Molecular basis of ligand selectivity for melatonin receptors.

Sleep disorders in adults are related to adverse health effects such as reduced quality of life and increased mortality. About 30-40% of adults are suffering from different sleep disorders. The human melatonin receptors (MT1 and MT2) are family A G protein-coupled receptors that respond to the neurohormone melatonin MEL which regulates circadian rhythm and sleep. Many efforts have been made to develop drugs targeting melatonin receptors to treat insomnia, circadian rhythm disorders, and even cancer. However, designing subtype-selective melatonergic drugs remains challenging due to their high similarities in both sequences and structures. MEL (a function-selective compound with a bulky ß-naphthyl group) behaves as an MT2-selective antagonist, whereas it is an agonist of MT1. Here, molecular dynamics simulations were used to investigate the ligand selectivity of MT receptors at the atomic level. We found that the binding conformation of MEL differs in different melatonin receptors. In MT1, the naphthalene ring of MEL forms a structure perpendicular to the membrane surface. In contrast, there is a 130° angle between the naphthalene ring of MEL and the membrane surface in MT2. Because of this conformational difference, the MEL leads to a constant water channel in MT1 which activates the receptor. However, MEL hinders the formation of continuous water channels, resulting in an inactive state of MT2. Furthermore, we found that A1173.29 in MT2 is a crucial amino acid capable of hindering the conformational flip of the MEL molecule. These results, coupled with previous functional data, reveal that although MT1 and MT2 share highly similar orthosteric ligand-binding pockets, they also display distinctive features that could be used to design selective compounds. Our findings provide new insights into functionally selective melatonergic drug development for sleep disorders.

Clinical HDAC Inhibitors Are Effective Drugs to Prevent the Entry of SARS-CoV2.

The outbreak of COVID-19 by the end of 2019 has posed serious health threats to humanity and jeopardized the global economy. However, no effective drugs are available to treat COVID-19 currently and there is a great demand to fight against it. Here, we combined computational screening and an efficient cellular pseudotyped virus system, confirming that clinical HDAC inhibitors can efficiently prevent SARS-CoV-2 and potentially be used to fight against COVID-19.

Multicenter phase II study of apatinib treatment for metastatic gastric cancer after failure of second-line chemotherapy.

Apatinib is a tyrosine kinase inhibitor and vascular endothelial growth factor receptor 2 (VEGFR-2) targeted drug. A phase I clinical trial showed that this agent has antitumor activity in Chinese patients with metastatic gastric cancer (mGC). The aim of this study was to investigate the safety and efficacy of apatinib treatment in patients with mGC. This was an open-label, multicenter, single-arm study involving four institutions in China. We enrolled 42 patients from March 2015 to October 2015 who experienced tumor progression after second-line chemotherapy and had no other treatment options that clearly conferred a survival benefit. Oral apatinib (850 mg daily) was administered within 30 min of eating breakfast, lunch, or dinner on days 1 through 28 of each 4-week cycle. The median progression-free survival (PFS) time and median overall survival (OS) time were 4.0 months (95% CI, 2.85-5.15) and 4.50 months (95% CI, 4.03-4.97), respectively. The disease control rate (DCR) and objective response rate (ORR) were, respectively, 78.57% and 9.52% after 2 cycles and 57.14% and 19.05% after 4 cycles. The main adverse events (AEs) were secondary hypertension, elevated aminotransferase, and hand-foot syndrome, with incidences of 35.71%, 45.24%, and 40.48%, respectively. The most common grade 3 to 4 AEs were secondary hypertension and elevated aminotransferase, with incidences of 7.14% each. Apatinib is effective and safe in heavily pretreated patients with mGC who fail to respond to two or more prior chemotherapy regimens. Toxicities were tolerable or could be clinically managed.