AI-Driven Parametrization of Michaelis-Menten Maximal Velocity: Advancing in silico New Approach Methodologies (NAMs)

NAM journal., Journal Year: 2025, Volume and Issue: unknown, P. 100012 - 100012

Published: Feb. 1, 2025

Language: Английский

---

All-Atom Biomolecular Simulation in the Exascale Era

Journal of Chemical Theory and Computation, Journal Year: 2024, Volume and Issue: 20(5), P. 1777 - 1782

Published: Feb. 21, 2024

Exascale supercomputers have opened the door to dynamic simulations, facilitated by AI/ML techniques, that model biomolecular motions over unprecedented length and time scales. This new capability holds potential revolutionize our understanding of fundamental biological processes. Here we report on some major advances were discussed at a recent CECAM workshop in Pisa, Italy, topic with primary focus atomic-level simulations. First, highlight examples current large-scale simulations future possibilities enabled crossing exascale threshold. Next, discuss challenges be overcome optimizing usage these powerful resources. Finally, close listing several grand challenge problems could investigated this computer architecture.

Language: Английский

---

AiGPro: a multi-tasks model for profiling of GPCRs for agonist and antagonist

Journal of Cheminformatics, Journal Year: 2025, Volume and Issue: 17(1)

Published: Jan. 29, 2025

G protein-coupled receptors (GPCRs) play vital roles in various physiological processes, making them attractive drug discovery targets. Meanwhile, deep learning techniques have revolutionized by facilitating efficient tools for expediting the identification and optimization of ligands. However, existing models GPCRs often focus on single-target or a small subset employ binary classification, constraining their applicability high throughput virtual screening. To address these issues, we introduce AiGPro, novel multitask model designed to predict molecule agonists (EC50) antagonists (IC50) across 231 human GPCRs, it first-in-class solution large-scale GPCR profiling. Leveraging multi-scale context aggregation bidirectional multi-head cross-attention mechanisms, our approach demonstrates that ensemble may not be necessary predicting complex states interactions. Through extensive validation using stratified tenfold cross-validation, AiGPro achieves robust performance with Pearson's correlation coefficient 0.91, indicating broad generalizability. This breakthrough sets new standard studies, outperforming previous studies. Moreover, multi-tasking can agonist antagonist activities wide range offering comprehensive perspective ligand bioactivity within this diverse superfamily. facilitate easy accessibility, deployed web-based platform access at https://aicadd.ssu.ac.kr/AiGPro . Scientific Contribution We learning-based multi-task generalize prediction accurately. The is implemented user-friendly web server rapid screening small-molecule libraries, GPCR-targeted discovery. Covering set targets, delivers robust, scalable advancing GPCR-focused therapeutic development. proposed framework incorporates an innovative dual-label strategy, enabling simultaneous classification molecules as agonists, antagonists, both. Each further accompanied confidence score, quantitative measure activity likelihood. advancement moves beyond conventional focusing solely binding affinity, providing more understanding ligand-receptor At core lies Bi-Directional Multi-Head Cross-Attention (BMCA) module, architecture captures forward backward contextual embeddings protein features. By leveraging BMCA, effectively integrates structural sequence-level information, ensuring precise representation molecular Results show highly accurate affinity predictions consistent families. unifying into single architecture, bridge critical gap modeling. enhances accuracy accelerates workflows, valuable

Language: Английский

---

Ligand-Induced Biased Activation of GPCRs: Recent Advances and New Directions from In Silico Approaches

Molecules, Journal Year: 2025, Volume and Issue: 30(5), P. 1047 - 1047

Published: Feb. 25, 2025

G-protein coupled receptors (GPCRs) are the largest family of membrane proteins engaged in transducing signals from extracellular environment into cell. GPCR-biased signaling occurs when two different ligands, sharing same binding site, induce distinct pathways. This selective offers significant potential for design safer and more effective drugs. Although its molecular mechanism remains elusive, big efforts made to try explain this using a wide range methods. Recent advances computational techniques AI technology have introduced variety simulations machine learning tools that facilitate modeling GPCR signal transmission analysis ligand-induced biased signaling. In review, we present current state silico approaches elucidate structural includes dynamics capture main interactions causing bias. We also highlight major contributions impacts transmembrane domains, loops, mutations mediating Moreover, discuss impact models on bias prediction diffusion-based generative ligands. Ultimately, review addresses future directions studying problem through approaches.

Language: Английский

---

Structural Systems Biology Toolkit (SSBtoolkit): From Molecular Structure to Subcellular Signaling Pathways

Journal of Chemical Information and Modeling, Journal Year: 2025, Volume and Issue: unknown

Published: April 18, 2025

Here, we introduce the Structural Systems Biology (SSB) toolkit, a Python library that integrates structural macromolecular data with systems biology simulations to model signal-transduction pathways of G-protein-coupled receptors (GPCRs). Our framework streamlines simulation and analysis mathematical models GPCRs cellular pathways, facilitating exploration kinetics induced by ligand-GPCR interactions: dose-response ligand can be modeled, along corresponding change in concentration other signaling molecular species over time, like for instance [Ca2+] or [cAMP]. SSB toolkit brings light possibility easily investigating subcellular effects binding on receptor activation, even presence genetic mutations, thereby enhancing our understanding intricate relationship between ligand-target interactions at level higher-level (patho)physiological response mechanisms.

Language: Английский

---

Nano-oncology revisited: Insights on precise therapeutic advances and challenges in tumor

Fundamental Research, Journal Year: 2025, Volume and Issue: unknown

Published: May 1, 2025

Language: Английский

---

Kisspeptin Receptor Agonists and Antagonists: Strategies for Discovery and Implications for Human Health and Disease

International Journal of Molecular Sciences, Journal Year: 2025, Volume and Issue: 26(10), P. 4890 - 4890

Published: May 20, 2025

The kisspeptin/kisspeptin receptor (KISS1/KISS1R) system has emerged as a vital regulator of various physiological processes, including cancer progression, metabolic function, and reproduction. KISS1R, member the G protein-coupled family, is crucial for regulating hypothalamic/pituitary/gonadal axis. A growing number KISS1R agonists are currently being investigated in clinical trials, whereas antagonists remains limited. Most existing ligands synthetic peptides, with only few small-molecule compounds, such musk ambrette, having been identified. In this article, we provide an overview KISS1/KISS1R its involvement diseases reproductive disorders, cancer, diabetes, cardiovascular disease. We also highlight technologies used to identify ligands, radioligand binding assays, calcium flux IP1 formation ERK phosphorylation qRT-PCR, AI-based virtual screening. Furthermore, discuss latest advances identifying antagonists, highlighting ongoing challenges future directions research. These insights lay groundwork research aimed at leveraging developing innovative therapeutic strategies across range medical conditions.

Language: Английский

---

Structures, dynamics, complexes, and functions: From classic computation to artificial intelligence

Current Opinion in Structural Biology, Journal Year: 2024, Volume and Issue: 87, P. 102835 - 102835

Published: May 13, 2024

Computational approaches can provide highly detailed insight into the molecular recognition processes that underlie drug binding, assembly of protein complexes, and regulation biological functional processes. Classical simulation methods bridge a wide range length- time-scales typically involved in such Lately, automated learning artificial intelligence have shown potential to expand reach physics-based approaches, ushering possibility model even design complex architectures. The synergy between atomistic simulations AI is an emerging frontier with huge for advances structural biology. Herein, we explore various examples frameworks these providing select instances applications illustrate their impact on fundamental biomolecular problems.

Language: Английский

---

Artificial intelligence-based parametrization of Michaelis–Menten maximal velocity: Toward in silico New Approach Methodologies (NAMs)

Research Square (Research Square), Journal Year: 2024, Volume and Issue: unknown

Published: April 25, 2024

The development of mechanistic systems biology models necessitates the utilization numerous kinetic parameters once enzymatic mode action has been identified. Moreover, wet lab experimentation is associated with particularly high costs, does not adhere to principle reducing number animal tests, and a time-consuming procedure. Alternatively, an artificial intelligence-based method proposed that utilizes enzyme amino acid structures as input data. This combines NLP techniques molecular fingerprints catalyzed reaction determine Michaelis–Menten maximal velocities (Vmax). employed include RCDK standard (1024 bits), MACCS keys (166 PubChem (881 E-States (79 bits). These were integrated produce fingerprints. data sourced from SABIO RK, providing concrete framework support training procedures. After preprocessing stage, dataset was randomly split into set (70%), validation (10%), test (20%), ensuring unique sequences for each subset. points similar those used train model well uncommon reactions further. developed optimized during predict Vmax values efficiently reliably. By utilizing fully connected neural network, these can be applied all organisms. proportions enzymes also tested, which revealed content unreliable predictor Vmax. During testing, demonstrated better performance on known than unseen In given use case, trained solely representations achieved R-squared 0.45 0.70 structures. When fingerprints, 0.46 0.62

Language: Английский

---

Leveraging Artificial Intelligence in GPCR Activation Studies: Computational Prediction Methods as Key Drivers of Knowledge

Methods in molecular biology, Journal Year: 2024, Volume and Issue: unknown, P. 183 - 220

Published: Nov. 14, 2024

Language: Английский

---