Therapeutic peptide development revolutionized: Harnessing the power of artificial intelligence for drug discovery

Exploring the impact of bioactive peptides from fermented Milk proteins: A review with emphasis on health implications and artificial intelligence integration

This review explores the health benefits of bioactive peptides (BAPs) from fermented milk proteins, emphasizing the transformative role of artificial intelligence (AI) and machine learning (ML) in advancing this field. BAPs exhibit diverse biological activities, including antimicrobial, antihypertensive, antioxidant, and immunomodulatory effects, making them promising for functional foods and nutraceuticals. However, challenges such as stability, bioavailability, and cost-effective production remain. This review highlights how AI/ML-driven tools, including data mining, sequence analysis, and predictive modeling, revolutionize peptide discovery, optimize fermentation, and enable personalized nutrition. By accelerating the identification of novel peptides and enhancing production efficiency, AI/ML offers innovative solutions to overcome existing limitations. The integration of AI/ML not only improves research efficiency but also opens new avenues for personalized nutrition and therapeutic applications. This review underscores the potential of interdisciplinary collaboration to harness the benefits of BAPs fully, driving future advancements in functional foods and health promotion.

Beyond Efficacy: Ensuring Safety in Peptide Therapeutics through Immunogenicity Assessment

Peptides are gaining remarkable popularity in clinical diagnosis and treatment due to their high selectivity and minimal side effects. Over 11% of all new pharmaceutical chemical entities authorised by the FDA between 2016 and 2024 were synthetically manufactured peptides. A critical factor that can potentially limit the efficacy and safety of peptide-based therapeutics or biologics is immunogenicity, defined as an unintended or adverse immune response to a protein or peptide therapy. This response may be triggered by the peptide itself or by impurities in the production or formulation steps, leading to the production of antidrug antibodies (ADAs). To address this, current regulatory guidelines require the assessment of risks in market authorization applications, which include identifying drug impurity levels and immunogenicity. The development and critical evaluation of appropriate immunogenicity assays is therefore highly warranted. Such assays must consider the fine complexities of the immune response, as well as its variation within the human population. Moreover, immunogenicity testing is expected to remain a priority as the shift toward greener chemistries in peptide synthesis may require reassessment of novel impurities in peptide formulations.

Machine Learning-Driven Consensus Modeling for Activity Ranking and Chemical Landscape Analysis of HIV-1 Inhibitors

Background/Objective: This study aimed to develop a predictive model to classify and rank highly active compounds that inhibit HIV-1 integrase (IN). Methods: A total of 2271 potential HIV-1 inhibitors were selected from the ChEMBL database. The most relevant molecular descriptors were identified using a hybrid GA-SVM-RFE approach. Predictive models were built using Random Forest (RF), eXtreme Gradient Boosting (XGBoost), Support Vector Machines (SVM), and Multi-Layer Perceptron (MLP). The models underwent a comprehensive evaluation employing calibration, Y-randomization, and Net Gain methodologies. Results: The four models demonstrated intense calibration, achieving an accuracy greater than 0.88 and an area under the curve (AUC) exceeding 0.90. Net Gain at a high probability threshold indicates that the models are both effective and highly selective, ensuring more reliable predictions with greater confidence. Additionally, we combine the predictions of multiple individual models by using majority voting to determine the final prediction for each compound. The Rank Score (weighted sum) serves as a confidence indicator for the consensus prediction, with the majority of highly active compounds identified through high scores in both the 2D descriptors and ECFP4-based models, highlighting the models' effectiveness in predicting potent inhibitors. Furthermore, cluster analysis identified significant classes associated with vigorous biological activity. Conclusions: Some clusters were found to be enriched in highly potent compounds while maintaining moderate scaffold diversity, making them promising candidates for exploring unique chemical spaces and identifying novel lead compounds. Overall, this study provides valuable insights into predicting integrase binders, thereby enhancing the accuracy of predictive models.

Peptide classification landscape: An in-depth systematic literature review on peptide types, databases, datasets, predictors architectures and performance

Peptides are gaining significant attention in diverse fields such as the pharmaceutical market has seen a steady rise in peptide-based therapeutics over the past six decades. Peptides have been utilized in the development of distinct applications including inhibitors of SARS-COV-2 and treatments for conditions like cancer and diabetes. Distinct types of peptides possess unique characteristics, and development of peptide-specific applications require the discrimination of one peptide type from others. To the best of our knowledge, approximately 230 Artificial Intelligence (AI) driven applications have been developed for 22 distinct types of peptides, yet there remains significant room for development of new predictors. A Comprehensive review addresses the critical gap by providing a consolidated platform for the development of AI-driven peptide classification applications. This paper offers several key contributions, including presenting the biological foundations of 22 unique peptide types and categorizes them into four main classes: Regulatory, Therapeutic, Nutritional, and Delivery Peptides. It offers an in-depth overview of 47 databases that have been used to develop peptide classification benchmark datasets. It summarizes details of 288 benchmark datasets that are used in development of diverse types AI-driven peptide classification applications. It provides a detailed summary of 197 sequence representation learning methods and 94 classifiers that have been used to develop 230 distinct AI-driven peptide classification applications. Across 22 distinct types peptide classification tasks related to 288 benchmark datasets, it demonstrates performance values of 230 AI-driven peptide classification applications. It summarizes experimental settings and various evaluation measures that have been employed to assess the performance of AI-driven peptide classification applications. The primary focus of this manuscript is to consolidate scattered information into a single comprehensive platform. This resource will greatly assist researchers who are interested in developing new AI-driven peptide classification applications.

Trends in the research and development of peptide drug conjugates: artificial intelligence aided design

Peptide-drug conjugates (PDCs) represent an emerging class of targeted therapeutic agents that consist of small molecular drugs coupled to multifunctional peptides through cleavable or non-cleavable linkers. The principal advantage of PDCs lies in their capacity to deliver drugs to diseased tissues at increased local concentrations, thereby reducing toxicity and mitigating adverse effects by limiting damage to non-diseased tissues. Despite the increasing number of PDCs being developed for various diseases, their advancements remain relatively slow due to several development constraints, which include limited available peptides and linkers, narrow therapeutic applications, and incomplete evaluation and information platforms for PDCs. Marked by the recent Nobel Prize awarded to artificial intelligence (AI) and de novo protein design for "protein design and structure prediction," AI is playing an increasingly important role in drug discovery and development. In this review, we summarize the recent developments and limitations of PDCs, highlights the potential of AI in revolutionizing the design and evaluation of PDC.

Natural Cyclic Peptides: Synthetic Strategies and Biomedical Applications

Natural cyclic peptides, a diverse class of bioactive compounds, have been isolated from various natural sources and are renowned for their extensive structural variability and broad spectrum of medicinal properties. Over 40 cyclic peptides or their derivatives are currently approved as medicines, underscoring their significant therapeutic potential. These compounds are employed in diverse roles, including antibiotics, antifungals, antiparasitics, immune modulators, and anti-inflammatory agents. Their unique ability to combine high specificity with desirable pharmacokinetic properties makes them valuable tools in addressing unmet medical needs, such as combating drug-resistant pathogens and targeting challenging biological pathways. Due to the typically low concentrations of cyclic peptides in nature, effective synthetic strategies are indispensable for their acquisition, characterization, and biological evaluation. Cyclization, a critical step in their synthesis, enhances metabolic stability, bioavailability, and receptor binding affinity. Advances in synthetic methodologies-such as solid-phase peptide synthesis (SPPS), chemoenzymatic approaches, and orthogonal protection strategies-have transformed cyclic peptide production, enabling greater structural complexity and precision. This review compiles recent progress in the total synthesis and biological evaluation of natural cyclic peptides from 2017 onward, categorized by cyclization strategies: head-to-tail; head-to-side-chain; tail-to-side-chain; and side-chain-to-side-chain strategies. Each account includes retrosynthetic analyses, synthetic advancements, and biological data to illustrate their therapeutic relevance and innovative methodologies. Looking ahead, the future of cyclic peptides in drug discovery is bright. Emerging trends, including integrating computational tools for rational design, novel cyclization techniques to improve pharmacokinetic profiles, and interdisciplinary collaboration among chemists, biologists, and computational scientists, promise to expand the scope of cyclic peptide-based therapeutics. These advancements can potentially address complex diseases and advance the broader field of biological drug development.

Evaluation of Structure Prediction and Molecular Docking Tools for Therapeutic Peptides in Clinical Use and Trials Targeting Coronary Artery Disease

This study evaluates the performance of various structure prediction tools and molecular docking platforms for therapeutic peptides targeting coronary artery disease (CAD). Structure prediction tools, including AlphaFold 3, I-TASSER 5.1, and PEP-FOLD 4, were employed to generate accurate peptide conformations. These methods, ranging from deep-learning-based (AlphaFold) to template-based (I-TASSER 5.1) and fragment-based (PEP-FOLD), were selected for their proven capabilities in predicting reliable structures. Molecular docking was conducted using four platforms (HADDOCK 2.4, HPEPDOCK 2.0, ClusPro 2.0, and HawDock 2.0) to assess binding affinities and interactions. A 100 ns molecular dynamics (MD) simulation was performed to evaluate the stability of the peptide-receptor complexes, along with Molecular Mechanics/Poisson-Boltzmann Surface Area (MM/PBSA) calculations to determine binding free energies. The results demonstrated that Apelin, a therapeutic peptide, exhibited superior binding affinities and stability across all platforms, making it a promising candidate for CAD therapy. Apelin's interactions with key receptors involved in cardiovascular health were notably stronger and more stable compared to the other peptides tested. These findings underscore the importance of integrating advanced computational tools for peptide design and evaluation, offering valuable insights for future therapeutic applications in CAD. Future work should focus on in vivo validation and combination therapies to fully explore the clinical potential of these therapeutic peptides.

In Silico Design of Peptide Inhibitors Targeting HER2 for Lung Cancer Therapy

BACKGROUND/OBJECTIVES

Human epidermal growth factor receptor 2 (HER2) is overexpressed in several malignancies, such as breast, gastric, ovarian, and lung cancers, where it promotes aggressive tumor proliferation and unfavorable prognosis. Targeting HER2 has thus emerged as a crucial therapeutic strategy, particularly for HER2-positive malignancies. The present study focusses on the design and optimization of peptide inhibitors targeting HER2, utilizing machine learning to identify and enhance peptide candidates with elevated binding affinities. The aim is to provide novel therapeutic options for malignancies linked to HER2 overexpression.

METHODS

This study started with the extraction and structural examination of the HER2 protein, succeeded by designing the peptide sequences derived from essential interaction residues. A machine learning technique (XGBRegressor model) was employed to predict binding affinities, identifying the top 20 peptide possibilities. The candidates underwent further screening via the FreeSASA methodology and binding free energy calculations, resulting in the selection of four primary candidates (pep-17, pep-7, pep-2, and pep-15). Density functional theory (DFT) calculations were utilized to evaluate molecular and reactivity characteristics, while molecular dynamics simulations were performed to investigate inhibitory mechanisms and selectivity effects. Advanced computational methods, such as QM/MM simulations, offered more understanding of peptide-protein interactions.

RESULTS

Among the four principal peptides, pep-7 exhibited the most elevated DFT values (-3386.93 kcal/mol) and the maximum dipole moment (10,761.58 Debye), whereas pep-17 had the lowest DFT value (-5788.49 kcal/mol) and the minimal dipole moment (2654.25 Debye). Molecular dynamics simulations indicated that pep-7 had a steady binding free energy of -12.88 kcal/mol and consistently bound inside the HER2 pocket during a 300 ns simulation. The QM/MM simulations showed that the overall total energy of the system, which combines both QM and MM contributions, remained around -79,000 ± 400 kcal/mol, suggesting that the entire protein-peptide complex was in a stable state, with pep-7 maintaining a strong, well-integrated binding.

CONCLUSIONS

Pep-7 emerged as the most promising therapeutic peptide, displaying strong binding stability, favorable binding free energy, and molecular stability in HER2-overexpressing cancer models. These findings suggest pep-7 as a viable therapeutic candidate for HER2-positive cancers, offering a potential novel treatment strategy against HER2-driven malignancies.