Wiley Interdisciplinary Reviews: Computational Molecular Science最新文献

Computational prediction and design of antimicrobial peptides (AMPs) has progressed quickly, moving from physicochemical descriptors paired with classical machine learning to transformer architectures and generative models. Each new architecture claims improved performance, yet basic questions remain unanswered. Does added complexity produce proportionally better biological predictions, and what do different architectures actually learn? Recent panoramic surveys have cataloged progress; the present review takes a different stance, arguing that the field has favored architectural novelty over biological understanding and has widened the gap between computational metrics and therapeutic relevance. Three findings emerge from a re-examination of the evidence. First, deep learning models do not consistently outperform well-designed shallow approaches, and both encode statistically similar chemical information. Second, reported in vitro validation rates of 59% to 100% for heavily pre-screened designs cluster within a narrow band irrespective of algorithm class, which suggests that pre-screening pipelines, rather than raw model accuracy, drive these numbers. Third, generative models such as HydrAMP, Multi-CGAN, and latent-diffusion frameworks sample within the physicochemically constrained envelope defined by their training data, and because antimicrobial activity can change sharply with a single substitution, descriptor-space novelty does not by itself establish functional novelty. To support such a critique with original evidence, the validation-rate analysis is extended across 18 representative studies. Two community instruments are then proposed: a Biology-First Algorithm Selection (BFAS) decision framework that operationalizes a four-axis selection protocol over task, data, interpretability, and compute budget; and AMP-CASP, an AMP-equivalent of the CASP blind-benchmark that turns an earlier call for a CASP-like community assessment into an operational protocol, with held-out activity, toxicity, and stability labels, homology-aware splitting, organism-resolved endpoints, and explicit assay-condition metadata. Progress now depends on an honest assessment of what algorithms can and cannot deliver in moving from computational prediction to clinical translation.

Equivariant graph neural networks (EGNNs) are becoming the geometric infrastructure of 3D molecular generation for AI-aided drug discovery. By enforcing E(n), E(3), or SE(3) equivariance, they separate physical molecular structure from arbitrary coordinate-frame conventions and ensure that predicted coordinates, denoising directions, and velocity fields transform consistently with molecular geometry. This symmetry-aware design supports direct modeling of conformations and protein–ligand spatial relationships while remaining compatible with diffusion, flow-based, autoregressive, and hybrid generative frameworks. Consequently, EGNNs enable more geometrically consistent generation and facilitate controllable design conditioned on binding pockets, pharmacophores, fragments, scaffolds, reference ligands, or molecular properties. Their benefits, however, are bounded by what equivariance encodes. Coordinate-frame consistency does not itself enforce valid bonds, valence, stereochemistry, topology–geometry agreement, synthesizability, or biological activity. Moreover, practical models must reconcile discrete chemical variables with continuous geometric dynamics, represent long-range interactions without prohibitive computational cost, and account for protein flexibility, solvent, metal ions, and induced fit. Performance is further constrained by biased docking or property predictors, heterogeneous benchmarks, and limited prospective validation. This review synthesizes how EGNNs function across major generative paradigms and argues that future drug-oriented systems should combine joint 2D–3D graph generation, physically credible interaction modeling, synthesis and ADMET constraints, multi-objective optimization, and closed-loop experimental feedback. EGNNs should therefore be viewed not as a complete chemical solution, but as the symmetry-aware foundation upon which more testable and biologically relevant molecular design systems can be built.

Virtual screening (VS) on small molecules aims to identify promising drug candidates against protein targets from expansive chemical libraries by balancing the core requirements of accurate scoring and efficient search against the inherent trade-off between accuracy and speed. This survey provides a comprehensive review of how Artificial Intelligence and Machine Learning (AI/ML) are redefining this landscape across three critical dimensions. First, we examine the evolution of AI-driven scoring functions, which utilize AI/ML models to capture complex structure–activity relationships from massive biochemical datasets, significantly enhancing structure- and ligand-based evaluations beyond traditional heuristics. Second, we summarize the emergence of efficient search algorithms that iteratively prioritize informative compounds to reduce search efforts by orders of magnitude. Third, we review the paradigm shift toward generative molecular design, making VS transition from screening fixed libraries to the de novo generation of molecules optimized for specific structural contexts and multi-objective properties. This review outlines the transition toward end-to-end, adaptive discovery systems that ensure computational hits are biologically potent, structurally optimized, and synthetically accessible.

Modern chemistry is pushing the limits of traditional Artificial Intelligence (AI) models, placing unprecedented demands on data availability to address humanity's most pressing challenges. One particular concern is AI's dependence on large, curated data and its tendency to deviate from or misrepresent fundamental chemistry principles. Nonetheless, this concern is often overshadowed by the urgent demand for emergent solutions to real-world problems. This perspective describes the incorporation of a domain-specific knowledge representation & reasoning (KR&R) framework with machine learning (ML) for predictive chemistry. KR&R is presented as a framework to represent chemical knowledge, making a formal connection between inductive hypothesis generation and deductive reasoning. By integrating scientific rules into data-driven processes, upholding a “chemist in the loop” approach, KR&R ensures that ML models are understandable and consistent with existing chemical theory. These concepts are illustrated by case studies where KR&R improves the interpretability of ML predictive models targeting thermodynamic properties (ΔG_sol, Δ_vapH_m°), reaction yields, and catalytic performance. These examples also show KR&R's importance in managing the complexity of modern computational chemistry, establishing it as a key component of explainable AI in the field.

G protein-coupled receptors (GPCRs) are the largest superfamily of membrane proteins and remain one of the most important classes of therapeutic targets. Although structural biology has provided valuable static structures for drug discovery, the translation of these insights into effective therapeutic strategies remains challenging because GPCR function and regulation are governed by complex conformational dynamics, metastable-state transitions, and long-range allosteric coupling. This review examines how computational strategies, including molecular dynamics (MD) simulations, artificial intelligence (AI) methods, and their integration, are being used to bridge this gap across three major areas, namely structural landscape mapping, mechanistic elucidation, and drug discovery. First, computational approaches for defining receptor architecture are summarized, combining static structure prediction with dynamic refinement to map conformational ensembles. Then, binding-site discovery and dynamic characterization are discussed, along with the identification of metastable states and the reconstruction of free energy landscapes. Subsequently, the use of AI and MD in elucidating signaling mechanisms is examined. Particular attention is given to decoding ligand binding and dissociation, allosteric communication networks, and the structural basis of biased signaling and effector coupling. Finally, computational workflows for therapeutic development, including virtual screening and the optimization of lead compounds, are highlighted. Future directions for integrating AI and MD are discussed, with the goal of moving GPCR research from static structural description toward dynamic mechanistic understanding and next-generation therapeutic development.

The dynamic interactions between enzymes and substrates, as well as the accurate elucidation of enzymatic catalytic mechanisms, are key scientific questions in modern enzymology. The development of quantum chemical calculation methods provides a new technical means for the study of enzymatic catalytic mechanisms. Among these, the theozyme model can enormously simplify the required computations and has become an important method for exploring enzymatic catalysis. However, as catalytic reactions discovered grow increasingly complex, the theozyme model built from static structures is increasingly found to be inadequate for accurate explanations of catalytic mechanisms. We have now coupled molecular dynamics (MD) simulations with the theozyme model to create a dynamic theozyme model for the study of complex catalytic systems. The new strategy utilizes MD simulations of transition states or key intermediates to obtain structural information during catalysis, providing reliable guidance for constructing the theozyme model. We have applied this strategy to several distinct catalytic systems, successfully revealing catalytic mechanisms leading to regioselectivity and stereoselectivity. The strategy proposed herein effectively integrates dynamic interaction analysis with quantitative mechanism evaluation, providing a research paradigm that balances accuracy and efficiency for elucidating enzymatic catalytic mechanisms.

This article is categorized under:

• Structure and Mechanism > Computational Biochemistry and Biophysics
• Structure and Mechanism > Reaction Mechanisms and Catalysis

Spatial confinement radically changes the collective behavior of amphiphilic molecules, generating self-assembled entities that significantly exceed bulk systems in morphological and topological complexity. Under geometric persuasion (and a modest nudge from entropy), these molecules orchestrate unexpected architectures: vesicle-in-vesicle nesting, fenestrated structures, and convoluted hydrophilic/hydrophobic chambers with curious properties for intracellular transport. Remarkably, confined systems echo the cell's spatially restricted nature, depicting fusion, fission, and engulfment events, suggesting that confinement does not merely impose boundary constraints but natural design principles: tune the cage, and the molecules will improvise a new choreography. Within nanosciences, recognizing confinement as a controllable parameter extends the traditional concept of self-assembly from a passive chemical consequence into an engineerable route toward hierarchical 3D structures with programmable form and function, each one a little kingdom with its own rules for curvature and asymmetry.

This article is categorized under:

• Structure and Mechanism > Computational Biochemistry and Biophysics
• Structure and Mechanism > Computational Materials Science
• Molecular and Statistical Mechanics > Molecular Dynamics and Monte-Carlo Methods

Effectively constructing an enzyme active site model is not only the most important task involved in quantum mechanical (QM) modeling of enzymes; it is also one of the most difficult. Without widely agreed-upon standards for selecting, truncating, and constraining residues, QM-cluster models have traditionally been built by bespoke approaches that are rarely consistent between labs. The Residue Interaction Network ResidUe Selector (RINRUS) is a software toolkit that automates and standardizes QM-cluster model construction with clear rules guaranteeing predictability and reproducibility. By selecting active site residues with interatomic contact networks, RINRUS replaces tedious visual inspection and chemical intuition with a data-driven approach. Algorithmic protocols are used for structure trimming, capping, and constraining to ensure the protein is truncated in chemically sensible places while keeping the system size reasonable. RINRUS can also prepare QM input files for the Gaussian, ORCA, Q-Chem, and Psi4 software packages. This fast and user-friendly automation of QM-cluster model design and input file preparation eliminates many of the learning barriers, hidden labor costs of project design/implementation, and inconsistencies that have historically plagued QM-cluster modeling. RINRUS is open-source and available for download at https://github.com/natedey/RINRUS.