Search Results (155)

Background: The rapid escalation of multidrug-resistant P. aeruginosa necessitates anti-virulence strategies targeting quorum sensing rather than bacterial survival; however, integrating scaffold repurposing with generative AI to inhibit LasR remains underexplored. Here, we address this gap by combining triazole scaffold mining with transformer-based de novo molecular generation to systematically identify putative LasR inhibitors. Methods: An integrated computational pipeline involving Structure-based inhibitor design using Generative Transformer Architecture, deep learning-assisted GNINA rescoring, density functional theory optimization, and molecular dynamics simulations was employed, followed by MM-GBSA binding free energy estimation. Results: Screening of 2666 triazole derivatives and 19,861 DrugGPT-generated compounds yielded top hits with superior binding affinities (−11.59 to −13.81 kcal/mol) compared to the reference ligand (−8.50 kcal/mol). MD simulations yielded stable protein–ligand complexes with RMSD values of 2.24–3.01 Å, while key interactions involving residues Tyr50, Asp67, and Ser123 were consistently maintained. Binding free energy calculations further confirmed strong thermodynamic stability, with MM-GBSA ΔG_bind values significantly favorable, supporting robust ligand–receptor affinity. Conclusions: Collectively, these findings establish a powerful AI-integrated framework for anti-virulence drug discovery and identify structurally diverse, high-affinity triazole-based and de novo compounds as promising lead candidates for disrupting LasR-mediated quorum sensing in P. aeruginosa. Full article

Background: The C-reactive protein/albumin ratio (CAR) has emerged as a prognostic biomarker across several malignancies, but its role in de novo metastatic colorectal cancer (mCRC) has not been evaluated using a propensity score-matched design. Methods: This retrospective study included 212 patients with mCRC who received first-line fluoropyrimidine-based chemotherapy plus a biological agent. The optimal CAR cutoff (2.69) was determined by ROC analysis (AUC = 0.714). Propensity score matching (PSM), incorporating 16 clinical, pathological, and molecular covariates, was performed to balance low-CAR and high-CAR groups. Progression-free survival (PFS) and overall survival (OS) were compared using Kaplan–Meier and multivariate Cox regression analyses. Results: PSM yielded 65 matched pairs (130 patients), eliminating all baseline imbalances present in the unmatched cohort. In the matched cohort, high CAR remained associated with shorter PFS (median 8.4 vs. 12.9 months; HR = 1.58, 95% CI 1.10–2.30, p = 0.014) and OS (median 22.7 vs. 31.1 months; HR = 1.70, 95% CI 1.15–2.50, p = 0.006). On multivariate analysis, CAR retained independent significance for OS (HR = 1.792, 95% CI 1.157–2.775, p = 0.009) alongside CEA level (HR = 3.331, p < 0.001) and BRAF mutation status (HR = 4.543, p = 0.040), but not for PFS, where mucinous component (HR = 2.335, p = 0.015) and CEA level (HR = 2.231, p = 0.045) were the dominant independent predictors. Conclusions: Baseline CAR is an independent prognostic factor for OS in mCRC, even after adjusting for established clinicopathological and molecular risk factors through propensity score matching. As an inexpensive and readily available biomarker, CAR may help identify patients with more aggressive disease biology who warrant closer monitoring, although prospective validation is needed before clinical implementation. Full article

Protein–protein interactions (PPIs) represent a vast and largely underexplored landscape of therapeutic targets, yet their structural features—including large, flat, and dynamic interfaces—have historically limited their druggability. In this context, cyclic peptides have emerged as a powerful class of PPI modulators, sitting at the interface between biologics and small molecules, and thus garnering key advantages of both classes. Their conformational constraint enhances binding affinity, proteolytic stability and, in some instances, cell permeability, thus enabling access to intracellular targets. This review provides an updated overview of cyclic peptides as modulators of PPIs, focusing on both conceptual foundations and practical strategies for their discovery and optimization. The main discovery approaches include natural sources, de novo design based on secondary structure mimetics, high-throughput screening, and computational approaches. Integration of these complementary strategies is crucial to enhance success rates in the discovery of effective and developable cyclic peptides. Accordingly, the present review aims to provide a practical guide for researchers entering this rapidly growing field, outlining current opportunities, methodological advances, and remaining challenges in the development of cyclic peptide-based PPI modulators. Full article

The combination of computational and experimental methods has become indispensable for optimization and rational enzyme design. Recently, the development of artificial intelligence (AI)-based tools has further streamlined enzyme engineering pipelines, enabling more accurate designs, while reducing the number of variants required for experimental validation. However, due to the intricate complexity of enzymatic systems, significant challenges must be addressed before we take the next step to fully optimize the use of these AI-guided enzyme design methodologies. These challenges include un-curated datasets, the need to consider both the static and dynamic structure of enzymes, and the requirement for effective interdisciplinary collaborations to ensure the integration of computational and experimental approaches. Here, we present recent advances in AI-based computational enzyme design, discussing the main challenges in the field and how a combination with classical physics-based methods could help overcome them. We further explore novel trends that could completely modulate the future of protein design and provide our outlook on the key concepts and future opportunities that will shape the next steps of enzyme design. Full article

Soybean is a strategic crop for global protein and vegetable oil supply chains; however, genetic improvement remains constrained by genotype-dependent regeneration, variable transformation efficiency, and regulatory concerns regarding stable transgene integration. This review synthesizes emerging DNA-free and genotype-independent genome-editing frameworks for soybean, where genotype independence is defined as the ability to recover fertile, non-chimeric edited plants across elite germplasm. We critically examine the soybean genome-editing toolbox, including CRISPR-Cas9, Cas12a, multiplex editing systems, base editing, and prime editing, and discuss persistent bottlenecks associated with target selection, off-target assessment, editability, and plant recovery. Particular emphasis is placed on artificial intelligence (AI)-assisted approaches that integrate genomic, epigenomic, chromatin-accessibility, and multi-omics datasets to improve target prioritization, guide RNA design, off-target prediction, and locus- and genotype-specific editability assessment. We further evaluate DNA-free genome-editing technologies, including CRISPR-Cas ribonucleoproteins, transient RNA-based systems, and nanocarrier-mediated delivery platforms, highlighting their potential to generate non-integrative edits while reducing prolonged nuclease exposure. In addition, we discuss regeneration reprogramming strategies based on developmental regulators and morphogenic modules, including BBM-WUS, GRF-GIF, de novo meristem induction, and somatic embryogenesis, as enabling technologies for overcoming cultivar-dependent regeneration barriers. Importantly, this review proposes an integrated AI-to-field framework that connects target discovery, editability prediction, DNA-free editing, regeneration reprogramming, phenotypic validation, and breeding deployment into a unified soybean improvement pipeline. We further highlight emerging opportunities in multi-omics-guided target discovery, genotype-aware prediction models, regeneration-aware editing strategies, and closed-loop machine-learning systems that continuously improve editing decisions through experimental feedback. Collectively, these convergent innovations provide a practical foundation for accelerating the development of climate-resilient, nutritionally enhanced, and industry-ready soybean cultivars. Full article

The majority of disease-associated proteins are considered “undruggable” due to the absence of well-defined binding pockets, the presence of extended interaction surfaces, and intrinsic structural disorder, which collectively limit the effectiveness of conventional small molecules and biologics. Representative examples include KRAS, p53, and c-MYC. Peptide therapeutics, particularly macrocyclic peptides, occupy a unique chemical space capable of targeting such recalcitrant protein–protein interactions (PPIs) where small molecules often fail. However, traditional peptide discovery, which relies heavily on high-throughput screening, is labor-intensive and frequently yields candidates with suboptimal pharmacological properties. The integration of artificial intelligence has begun to transform peptide discovery from a largely empirical process into a rational and design-driven paradigm. Modern deep learning approaches, including diffusion-based generative models, enable the de novo design of peptide binders with high affinity and structural precision, even for disordered or previously intractable targets. In this perspective, we highlight key structural and biological challenges associated with undruggable proteins and consider how peptide-based modalities are beginning to overcome these longstanding barriers. We further explore how advances in artificial intelligence and computational modeling may reshape the rational design of next-generation peptide therapeutics and propose an integrated experimental–computational framework to facilitate the development of clinically actionable candidates. Full article

Recently, there has been great interest in AI-based approaches for de novo design of novel drug candidates. However, the generation of useful lead drug candidate compounds requires more than predicting engagement with the desired protein target. Candidate molecules must also be anchored in the real world of medicinal chemistry for their synthesis and modification as well as satisfying multiple drug development-related criteria. Here, we present Nevermore, an AI target-conditioned, database-grounded workflow for prioritizing candidate ligands from large compound libraries. Nevermore uses a geometry-aware protein–ligand affinity oracle to score target-specific binding and perform sparse integer edits in count-based Morgan fingerprint space. Nevermore then retrieves the most structurally similar molecules from public chemical databases. This design enables multi-objective search over predicted affinity and absorption, distribution, metabolism, excretion, and toxicity (ADMET) proxies while keeping all candidates anchored to valid database compounds. We evaluated Nevermore’s performance across three biologically distinct targets: Menin, a protein-interaction target relevant to leukemia; SARS-CoV-2 M^pro, a viral cysteine protease relevant to antiviral discovery; and epidermal growth factor receptor (EGFR), a kinase-superfamily oncology target with extensive experimentally tested compounds. Nevermore retrieved candidate sets with favorable predicted affinity–property trade-offs. These results support database-grounded fingerprint steering as a practical computational strategy for lead prioritization and for generating testable molecular hypotheses, although the prioritized candidates remain predictions, requiring follow-up experimental validation. Full article

To clarify the cellular origin of matrix metalloproteinase-9 (MMP9) and explore targeted research, we utilized single-cell RNA sequencing analysis, which revealed that MMP9 is predominantly enriched in specific macrophages within the activated B-cell-like (ABC) subtype. Guided by this target information, we applied a generative AI pipeline incorporating RFdiffusion, ProteinMPNN, and AlphaFold to de novo design protein binders targeting the hemopexin (PEX) domain of MMP9. ELISA experiments confirmed the in vitro binding capability of these designs; among them, MMP9-30 displayed the strongest binding, with an apparent EC50 of approximately 1.1 μM, followed by MMP9-34, while MMP9-97 showed the weakest interaction. This study successfully integrates single-cell sequencing with AI-assisted protein design, providing a preliminary exploratory framework for subsequent MMP9-targeted research and protein binder development. Full article

Artificial intelligence has evolved to generative AI (GenAI), a paradigm shift that has shifted the emphasis away from the evaluation of existing patterns to the generation of novel biological and medical material. This study examines GenAI achievements in biosciences and medical fields the last five years in these fields using databases such as PubMed and Scopus. The paper highlights the recent evolution in biomedical research from virtual screening to de novo design. It illustrates how models like RFdiffusion and ProteinMPNN leverage “inverse folding” to assemble novel of proteins and drugs. Ultimately, these generative methods yield candidate with enhanced binding affinity and structural stability. For example, exploratory studies suggest GenAI has the potential to address inefficiencies via automatic documentation in the therapeutic sector, and it may enhance research capabilities by using Generative Adversarial Networks (GANs) and Variational Autoencoders (VAEs) to generate synthetic clinical trial data that preserves confidentiality. In addition, the review argues that though GenAI democratizes medical education through scalable simulations, it raises questions about long-term knowledge retention. Finally, GenAI also offers a transformative “write” capability for biology, but its responsible application will require addressing model “hallucinations” and building Explainable AI (XAI) and robust ethical frameworks. Full article

Peptide therapeutics occupy a unique chemical space between small molecules and biologics, combining high target specificity with structural programmability and favorable safety profiles. Recent regulatory approvals and expanding clinical pipelines underscore the growing therapeutic and commercial relevance of peptide-based drugs. This review outlines chemical modification approaches and contemporary design strategies, and evaluates their impact on proteolytic stability, pharmacokinetics, membrane permeability, and target engagement. We then highlight recent advances in artificial intelligence (AI)-guided peptide drug design, including machine learning models, protein language models, and generative architectures that enable high-throughput activity prediction, property optimization, and de novo sequence generation. These approaches collectively accelerate the traditional discovery–design–validation cycle while reducing experimental attrition through data-driven, structure-informed modeling frameworks. Among these applications, AI also enables the rational design of cell-penetrating peptides (CPPs) to enhance intracellular delivery and biological activity. Building on these methodological advances, we further examine their application to peptide therapeutics, with particular emphasis on AI-based predictive models for CPPs as well as on therapeutic applications within the central nervous and pulmonary systems. We conclude by outlining future perspectives and emphasize that the systematic integration of AI-enabled sequence design with rational chemical engineering and advanced delivery technologies, supported by rigorous experimental validation, will be critical for developing robust and clinically durable peptide-based medicines. Full article

Pirfenidone (PFD) shows therapeutic potential for liver fibrosis, but its molecular mechanisms are not fully elucidated. Activation of hepatic stellate cells (HSCs) is central to liver fibrosis, making their targeted elimination a prime therapeutic strategy. Since amino acid metabolism governs both HSC activation and ferroptosis, we investigated whether PFD acts by reprogramming these metabolic pathways. Analysis of primary rat HSCs revealed that their in vitro activation induced fibrotic markers, including collagen type I and α-smooth muscle actin, as well as key metabolic enzymes. Specifically, we observed upregulation of glutaminase 1, initiating glutaminolysis to produce glutamate; serine hydroxymethyltransferase 2, which generates glycine from serine; and pyrroline-5-carboxylate synthase, the rate-limiting enzyme for de novo proline synthesis. Treatment with PFD suppressed HSC activation by reducing protein levels of these enzymes, an effect consistent with PFD’s inhibition of activating transcription factor 4 nuclear accumulation. This created a dual metabolic vulnerability, limiting amino acid precursors for both collagen synthesis and the master antioxidant glutathione (GSH). Consequently, while PFD alone was not cytotoxic, GSH depletion sensitized activated HSCs to ferroptosis. Co-treatment with the ferroptosis inducer erastin triggered a synergistic increase in reactive oxygen species, labile iron, and lipid peroxidation, culminating in cell death. This synergistic lethality was abrogated by the ferroptosis inhibitor ferrostatin-1 and the antioxidant N-acetylcysteine, confirming ferroptosis as the specific cell death modality. Our study uncovers a dual anti-fibrotic mechanism for PFD: PFD inhibits collagen synthesis by limiting key amino acid precursors and depletes GSH. This compromises antioxidant defenses, creating vulnerability to ferroptosis. Our findings establish a rationale for using PFD in combination therapies designed to eliminate activated HSCs. Full article

Peptide binders, serving as a critical drug modality bridging small-molecule compounds and protein macromolecules, can effectively mimic the secondary structural elements of natural proteins. Peptides exhibit unique physicochemical advantages when targeting protein protein interaction (PPI) interfaces, which are typically characterized by flat surfaces and extensive contact areas. Recently, diffusion models represented by RFdiffusion have established a new computational paradigm for protein backbone generation by defining a denoising process over the rigid-body transformation group. However, in the de novo design of binders targeting “undruggable” PPI targets, this general paradigm encounters significant adaptability bottlenecks. First, its underlying rigid-body assumption struggles to accurately describe the dynamic induced-fit process of peptides at the binding interface. Second, it lacks sufficient robustness to the experimental resolution heterogeneity inherent in training data. Furthermore, the decoupled two-stage generation of sequence and structure severs the synergy of physicochemical properties, leading to backbones with idealized, singular secondary structures that lack authentic spatial binding capacity and reasonable side-chain physicochemical features. To address these challenges, this study proposes PPI-Diff, a novel generative framework. While preserving the generative capability of diffusion models, PPI-Diff introduces three core mechanisms: (1) a resolution-aware constraint mechanism that maps the measurement precision of experimental data into explicit contextual constraints to dynamically suppress geometric noise from low-resolution samples; (2) an internal-coordinate-driven manifold diffusion model that performs conformational evolution on a Riemannian manifold constructed by dihedral angles, balancing local stereochemical validity with the precise capture of flexible peptide conformations; and (3) a geometry-semantic synergistic modeling mechanism that leverages the evolutionary embeddings of a pre-trained protein language model (ESM-2) as latent variables to align structure generation with biophysical functions. Systematic benchmarking demonstrates that, on a strictly non-homologous test set, the binders generated by PPI-Diff significantly outperform existing baseline models in terms of interface contact density, stereochemical validity, and sequence novelty. Full article

Background/Objectives: Protein–protein interactions (PPIs) involving oncogenic drivers remain among the most intractable targets in cancer biology due to their dynamic conformations and limited accessibility to conventional small molecules. Although antibodies and inhibitors have achieved clinical success against targets such as PD-1/PD-L1 and MYC, challenges persist related to tissue penetration, intracellular delivery, resistance, and incomplete blockade of key interface hotspots. The objective of this study is to develop an integrated computational framework for systematically designing hotspot-conditioned de novo miniprotein binders to target these interfaces. Methods: We present DesignForge, a computational protein design pipeline that integrates energetic hotspot identification, generative backbone design, sequence optimization, and structural confidence evaluation. The framework combines hotspot mapping using an open force-field-based energetic analysis module with generative backbone sampling using BindCraft, sequence optimization using ProteinMPNN, and structural validation using AlphaFold2. This in silico pipeline was applied to three representative oncogenic interfaces: PD-1/PD-L1, MYC/MAX, and KRAS/RAF. Results: Computationally generated designs exhibited high predicted structural confidence, favorable interface energetics, and consistent engagement of identified hotspot residues across targets. AlphaFold2-Multimer structural modeling indicated that the candidate PD-1 mimetic scaffolds, MYC/MAX interface binders, and KRAS interaction candidates can adopt conformations compatible with the target interfaces. Energetic contact analysis further supported predicted engagement of key hotspot residues. These findings support the computational feasibility of hotspot-conditioned binder generation using a unified design workflow. Conclusions: DesignForge provides a reproducible computational framework for hotspot-guided de novo protein binder design targeting oncogenic protein–protein interfaces. The designs reported here represent computational predictions derived from structural modeling and energetic analysis. Experimental biochemical and cellular validation will be required to determine the functional activity of the proposed binders. Full article

This article offers a reassessment of the Intelligent Design doctrine by engaging probability theory, complexity theory and contemporary artificial intelligence. Andrey Kolmogorov’s work shows that chance belongs to an intelligible mathematical order and that complex structures can arise from patterns that admit concise description. This challenges the assumption that improbability signals an external designer and instead points to a creation whose inner rationality is stable and fruitful. Insights from self-organizing systems strengthen this view by showing how new forms of order emerge from the interaction of fluctuation and natural constraint. Recent advances in artificial intelligence including AlphaFold, de novo protein design and the Brain-Derived Hebbian architecture make aspects of this intelligibility visible by modeling and predicting biological form and basic patterns of reasoning without recourse to explicit foresight. Their capacity to generate coherent structures under learned constraints reflects the rational order of creation, which Christian theology identifies with the Divine Logos. This order provides a deeper account of divine action than interpretations of Intelligent Design grounded solely in structural improbability. Full article

Allosteric modulation has emerged as a powerful strategy for achieving superior selectivity and safety in drug discovery and protein function regulation. Unlike highly conserved orthosteric sites, allosteric pockets are structurally diverse and less evolutionarily constrained, making them particularly suitable for modulation by designed miniproteins. Miniproteins can provide extended binding interfaces and high affinity for shallow, dynamic, or cryptic regulatory surfaces that are often inaccessible to small molecules. Recent advances in artificial intelligence (AI) are transforming this field through deep learning-based structure prediction and generative modeling. These AI-driven approaches enable the identification of allosteric hotspots, characterization of conformational ensembles, and de novo design of structured miniprotein binders. They are rapidly expanding the landscape for designing selective modulators across diverse allosteric targets, including GPCRs, receptor tyrosine kinases, nuclear receptors, ion channels, and other protein–protein interaction systems. This review summarizes state-of-the-art AI-driven computational methodologies for designing miniproteins as potential allosteric modulators and discusses their current challenges and future opportunities in allosteric drug discovery. Full article