NPJ drug discovery最新文献_第2页

Interpretable machine learning framework for peptide lead optimization in ultralarge design spaces. 超大设计空间中肽导联优化的可解释机器学习框架。

NPJ drug discovery Pub Date : 2026-10-12 DOI: 10.1038/s44386-026-00069-y

Yixiang Mao, Ruibo Zhang, Robert Sheridan, Dan Sindhikara, Dipannita Kalyani, Li Xiao, Amy B McCracken, Bryant Kim, Shasha Feng, Tianxiong Mi, Christopher W Plummer, Jennifer M Johnston, Liying Zhang

{"title":"Interpretable machine learning framework for peptide lead optimization in ultralarge design spaces.","authors":"Yixiang Mao, Ruibo Zhang, Robert Sheridan, Dan Sindhikara, Dipannita Kalyani, Li Xiao, Amy B McCracken, Bryant Kim, Shasha Feng, Tianxiong Mi, Christopher W Plummer, Jennifer M Johnston, Liying Zhang","doi":"10.1038/s44386-026-00069-y","DOIUrl":"10.1038/s44386-026-00069-y","url":null,"abstract":"In modern peptide drug discovery, the incorporation of noncanonical amino acids (ncAAs) has become a powerful strategy to improve potency and key developability properties, including stability, solubility, and oral bioavailability. This shift has expanded peptide chemical diversity and design spaces well beyond natural residues, creating fundamental challenges for predictive methods (e.g., quantitative structure-activity relationship modeling) used in lead optimization: limited transferability from natural residues to novel ncAAs, limited interpretability, and poor scalability to ultralarge peptide spaces (>1050). We present PepOpt, an interpretable machine learning framework that unifies molecule-based and sequence-based peptide representations in linear models and exploits dynamic programming to efficiently explore ultralarge peptide spaces. PepOpt achieves high predictive accuracy and interpretability, supports training and inference across diverse ncAAs, including ncAAs absent from the training set whose constituent substructures are represented in the training data. Its linear structure decomposes predictions into residue-specific contributions, enabling systematic optimization without exhaustive sequence enumeration. Notably, PepOpt is inherently capable of multi-parameter optimization without specialized modifications. Together, these capabilities position PepOpt as a valuable platform for prediction, interpretation, ideation, and rapid exploration in ncAA-rich ultralarge peptide chemical spaces, accelerating the rational design of novel peptide therapeutics.","PeriodicalId":520448,"journal":{"name":"NPJ drug discovery","volume":"3 1","pages":""},"PeriodicalIF":0.0,"publicationDate":"2026-10-12","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC13658789/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"149170251","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Computational design and translational development of TCR-mimic scFv antibodies for peptide-HLA neoantigen targeting. 针对多肽hla新抗原靶向的tcr -模拟scFv抗体的计算设计和翻译开发。

NPJ drug discovery Pub Date : 2026-10-12 DOI: 10.1038/s44386-026-00084-z

Sarfaraz K Niazi, Matthias Magoola, Zamara Mariam

{"title":"Computational design and translational development of TCR-mimic scFv antibodies for peptide-HLA neoantigen targeting.","authors":"Sarfaraz K Niazi, Matthias Magoola, Zamara Mariam","doi":"10.1038/s44386-026-00084-z","DOIUrl":"10.1038/s44386-026-00084-z","url":null,"abstract":"Single-chain variable fragments (scFvs) penetrate tissue more effectively than full antibodies and are simpler to produce, but the mutant proteins that define a tumor are predominantly intracellular and therefore inaccessible to conventional antibodies. What is accessible is the mutation-derived peptide displayed on cell-surface HLA. This review concerns TCR-mimic scFvs directed at those peptide-HLA complexes. It separates three questions that are frequently conflated: which mutations are predicted to be presented, which presented complexes are genuinely targetable given surface density and cross-reactivity risk, and how immunogenic the resulting therapeutic molecule will itself be. We review computational strategies spanning target identification, AI-driven structure prediction, de novo design, affinity and specificity engineering, half-life extension and deimmunization, and distinguish throughout between the scFv as a recognition module and the effector architecture, such as a CD3 bispecific, a chimeric antigen receptor binding domain or an Fc-containing format, in which it must ultimately be deployed.","PeriodicalId":520448,"journal":{"name":"NPJ drug discovery","volume":"3 1","pages":""},"PeriodicalIF":0.0,"publicationDate":"2026-10-12","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC13658221/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"149170839","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

TechBio 3.0 closes the drug discovery loop with multimodal AI and generative chemistry. TechBio 3.0通过多模态人工智能和生成化学完成了药物发现循环。

NPJ drug discovery Pub Date : 2026-10-09 DOI: 10.1038/s44386-026-00078-x

Krish Ramadurai, Abhirup Banerjee

引用次数: 0

Personalized drug discovery from rare diseases to cancer. 从罕见疾病到癌症的个性化药物发现。

NPJ drug discovery Pub Date : 2026-09-30 DOI: 10.1038/s44386-026-00085-y

Vladimir L Katanaev

引用次数: 0

Harnessing snake venom cardiotoxins for antimicrobial peptide discovery. 利用蛇毒心脏毒素发现抗菌肽。

NPJ drug discovery Pub Date : 2026-09-16 DOI: 10.1038/s44386-026-00073-2

Bruno Mendes, José R Almeida, Valeria Castelletto, Ian W Hamley, Glyn Barrett

{"title":"Harnessing snake venom cardiotoxins for antimicrobial peptide discovery.","authors":"Bruno Mendes, José R Almeida, Valeria Castelletto, Ian W Hamley, Glyn Barrett","doi":"10.1038/s44386-026-00073-2","DOIUrl":"10.1038/s44386-026-00073-2","url":null,"abstract":"Snake venoms are a rich source of bioactive molecules with considerable potential for drug discovery. Cardiotoxins (CTXs) from cobras (Naja spp.) are membrane-active venom proteins and promising templates for antimicrobial peptide development. Here, we combined two computational toxin-mining strategies to identify antimicrobial peptide candidates from CTXs. First, AMPA-guided sequence mining was used to detect encrypted antimicrobial regions within CTX sequences. Second, sequence alignment and consensus-sequence analysis were employed to generate peptides from a conserved CTX scaffold. Candidate peptides were prioritised using multiple machine learning- and deep learning-based antimicrobial peptide prediction tools, resulting in fourteen CTX-inspired peptides, including two modified derivatives of CTX-p5. Antimicrobial screening showed limited activity across the peptide panel, with CTX-p5 exhibiting the strongest activity against both Gram-negative and Gram-positive bacteria. Neither truncation nor chemical modification improved its antibacterial potency. Mechanistic studies, including atomic force microscopy, flow cytometry, fluorescence microscopy and membrane permeabilisation assays, indicated that CTX-p5 disrupts bacterial membrane integrity. These findings identify CTXs as a source of antimicrobial peptide templates while highlighting the limitations of current computational prioritisation approaches and the need for more accurate toxin-mining workflows to enhance peptide selection during toxin-inspired drug discovery.","PeriodicalId":520448,"journal":{"name":"NPJ drug discovery","volume":"3 1","pages":""},"PeriodicalIF":0.0,"publicationDate":"2026-09-16","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC13582888/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148984564","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

RUNX2 inhibitor CADD522 improves bone microarchitecture and lipid metabolism in post-menopausal bone loss. RUNX2抑制剂CADD522改善绝经后骨质流失的骨微结构和脂质代谢。

NPJ drug discovery Pub Date : 2026-09-07 DOI: 10.1038/s44386-026-00076-z

Adel Ersek, Myoung Sook Kim, Caterina Suelzu, Isabelle Piec, Aerielle Matsangos, Elmar Nurmemmedov, Emma C Bull, Tabitha Bartlett, Rachel Dunn, James R Dick, Rena Lapidus, Danielle M Copeman, Jonathan C Y Tang, Paul G Crichton, Matthew G Pontifex, William D Fraser, Nicole J Horwood, Antonino Passaniti, Darrell Green

{"title":"RUNX2 inhibitor CADD522 improves bone microarchitecture and lipid metabolism in post-menopausal bone loss.","authors":"Adel Ersek, Myoung Sook Kim, Caterina Suelzu, Isabelle Piec, Aerielle Matsangos, Elmar Nurmemmedov, Emma C Bull, Tabitha Bartlett, Rachel Dunn, James R Dick, Rena Lapidus, Danielle M Copeman, Jonathan C Y Tang, Paul G Crichton, Matthew G Pontifex, William D Fraser, Nicole J Horwood, Antonino Passaniti, Darrell Green","doi":"10.1038/s44386-026-00076-z","DOIUrl":"10.1038/s44386-026-00076-z","url":null,"abstract":"Osteoporosis is a leading cause of age-related morbidity, yet existing antiresorptive and anabolic therapies remain limited by safety concerns, contraindications and poor long-term adherence. CADD522 is a small molecule inhibitor of the RUNX2 transcription factor currently under development for cancer therapy. Here, we investigated whether RUNX2 inhibition could protect against post-menopausal bone loss. In an ovariectomy-induced mouse model, CADD522 (25 mg/kg, three times weekly for eight weeks) enhanced bone formation, preserved trabecular microarchitecture and reduced marrow and peripheral adiposity. Cross-species pharmacokinetic and toxicological studies demonstrated oral bioavailability, favourable short-term tolerability and target engagement despite rapid systemic clearance, while cellular thermal shift assays confirmed direct engagement of RUNX2. Together, these findings identify RUNX2 inhibition as a therapeutic strategy that simultaneously improves skeletal integrity and metabolic homeostasis, supporting further development of CADD522 for osteoporosis and other RUNX2-driven diseases.","PeriodicalId":520448,"journal":{"name":"NPJ drug discovery","volume":"3 1","pages":""},"PeriodicalIF":0.0,"publicationDate":"2026-09-07","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC13550584/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148908592","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Closing the recapitulation gap in obesity-conditioned TNBC drug development. 缩小肥胖条件下TNBC药物开发的重述差距。

NPJ drug discovery Pub Date : 2026-09-03 DOI: 10.1038/s44386-026-00074-1

Karim I Budhwani, Erica M Stringer-Reasor, Anna G Sorace, Donald J Buchsbaum, Chelsea L Crawford

引用次数: 0

Discovery and preclinical validation of a translationally optimized mitochondrial complex I modulator for Alzheimer's disease. 阿尔茨海默病翻译优化线粒体复合体I调节剂的发现和临床前验证。

NPJ drug discovery Pub Date : 2026-09-02 DOI: 10.1038/s44386-026-00072-3

Sergey Trushin, Thi Kim Oanh Nguyen, Mark Ostroot, Jarred J Nesbitt, Tetiana Kovalenko, Alexander Galkin, Toshihiko Nambara, Wenyan Lu, Takahisa Kanekiyo, Graham Johnson, Eugenia Trushina

{"title":"Discovery and preclinical validation of a translationally optimized mitochondrial complex I modulator for Alzheimer's disease.","authors":"Sergey Trushin, Thi Kim Oanh Nguyen, Mark Ostroot, Jarred J Nesbitt, Tetiana Kovalenko, Alexander Galkin, Toshihiko Nambara, Wenyan Lu, Takahisa Kanekiyo, Graham Johnson, Eugenia Trushina","doi":"10.1038/s44386-026-00072-3","DOIUrl":"10.1038/s44386-026-00072-3","url":null,"abstract":"Alzheimer's disease (AD) is characterized by progressive metabolic failure, impaired mitochondrial function, and diminished adaptive stress responses, highlighting the need for disease-modifying therapies that restore cellular resilience rather than target downstream pathology. Here, we report the discovery and preclinical validation of C273, a translationally optimized, brain-penetrant mitochondrial complex I (mtCI) modulator developed through medicinal chemistry optimization of our first-generation compounds. C273 retained nanomolar neuroprotective activity against Aβ-induced toxicity while exhibiting favorable drug-like properties, including high oral bioavailability, efficient brain penetration, microsomal stability, minimal CYP and off-target pharmacology liabilities, and selective mild modulation of mtCI. Mechanistic studies demonstrated that C273 activated AMP-activated protein kinase (AMPK) and coordinated antioxidant, autophagic, anti-inflammatory, and mitochondrial quality-control pathways in cultured cells and mouse brain. These responses were absent in AMPKα1/α2-deficient cells, establishing AMPK as an essential mediator, while rotenone pretreatment abolished C273-mediated neuroprotection, supporting engagement of the mtCI quinone-binding site. Repeated administration to wild-type mice for 30 days produced no detectable cardiac or hepatic toxicity. Importantly, C273 activated the same neuroprotective pathways and reduced Aβ and p-Tau levels in induced pluripotent stem cell-derived cerebral organoids from patients with sporadic AD. Together, these findings establish mild modulation of mtCI as a therapeutic strategy to restore metabolic resilience and identify C273 as a promising disease-modifying candidate for AD treatment.","PeriodicalId":520448,"journal":{"name":"NPJ drug discovery","volume":"3 1","pages":""},"PeriodicalIF":0.0,"publicationDate":"2026-09-02","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC13538682/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148883311","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Data driven selection of consensus docking pipelines for structure based hit identification. 基于结构的命中识别共识对接管道的数据驱动选择。

NPJ drug discovery Pub Date : 2026-09-02 DOI: 10.1038/s44386-026-00063-4

Hamza Agha, Youssef Ibrahim, Michael Backenköhler, Antoine Lacour, Mostafa M Hamed, Anna K H Hirsch, Andrea Volkamer

{"title":"Data driven selection of consensus docking pipelines for structure based hit identification.","authors":"Hamza Agha, Youssef Ibrahim, Michael Backenköhler, Antoine Lacour, Mostafa M Hamed, Anna K H Hirsch, Andrea Volkamer","doi":"10.1038/s44386-026-00063-4","DOIUrl":"10.1038/s44386-026-00063-4","url":null,"abstract":"Structure-based virtual screening (SBVS) is a cornerstone of computer-aided drug design, yet its success depends on selecting a combination of docking tools, scoring function (SF), and ranking strategies. MolDockLab addresses this challenge with an automated, data-driven framework that optimizes SBVS workflows for a protein target, balancing predictive performance and computational efficiency. It systematically explores combinations of five docking engines, 15 SF, and three consensus ranking strategies using a calibration set of ≈ 200 compounds with known bioactivity, and applies the best-correlating workflow to the larger screening library. Final hit selection from the top 1% integrates protein-ligand interaction profiler (PLIP)-derived interaction fingerprints, structural-diversity assessment, and expert visual inspection. In a retrospective evaluation on the epidermal growth factor receptor (EGFR), the chosen pipeline achieved a Spearman correlation of 0.36 and an enrichment factor (EF) at 10% of 1.57, consistent with calibration. Prospectively, for the energy coupling factor transporters (ECF-T)-a challenging transmembrane target with a cryptic binding site and no co-crystallized ligand-the pipeline reached a correlation of 0.45 and enrichment of 3.13. Post-processing enabled in vitro confirmation of two chemically novel inhibitors rivaling the most potent ECF-T inhibitors reported to date.","PeriodicalId":520448,"journal":{"name":"NPJ drug discovery","volume":"3 1","pages":""},"PeriodicalIF":0.0,"publicationDate":"2026-09-02","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC13538370/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148883320","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Pushing the boundaries of virtual screening scale of combinatorial spaces with the V-SYNTHES approach. 利用V-SYNTHES方法突破组合空间虚拟筛选尺度的界限。

NPJ drug discovery Pub Date : 2026-09-02 DOI: 10.1038/s44386-026-00067-0

Mykola Protopopov, Olha Semenenko, Maryna Vasylchuk, Anastasiia V Sadybekov, Arman A Sadybekov, Kateryna Horbatok, Oleksii Hrabovskyi, Anna Kapeliukha, Antonina Nazarova, Dmytro Radchenko, Vsevolod Katritch, Olga Tarkhanova

{"title":"Pushing the boundaries of virtual screening scale of combinatorial spaces with the V-SYNTHES approach.","authors":"Mykola Protopopov, Olha Semenenko, Maryna Vasylchuk, Anastasiia V Sadybekov, Arman A Sadybekov, Kateryna Horbatok, Oleksii Hrabovskyi, Anna Kapeliukha, Antonina Nazarova, Dmytro Radchenko, Vsevolod Katritch, Olga Tarkhanova","doi":"10.1038/s44386-026-00067-0","DOIUrl":"10.1038/s44386-026-00067-0","url":null,"abstract":"Computational screening of giga-scale chemical spaces opens a cost-effective path to high-quality hit identification, providing entry points for drug discovery. As these on-demand spaces grow and successful applications multiply, rigorous blind benchmarks like CACHE Challenges provide important performance metrics for computational tools. Here, we report the first application of the V-SYNTHES2 synthon-based screening approach to the 173-billion-compound Enamine xREAL Space, a 16-fold expansion beyond its previous benchmarks, demonstrating near-linear computational scaling with only a 10-15% increase in cost relative to the 11-billion-compound REAL Space. We applied this workflow in CACHE Challenge #2, targeting the RNA-binding site of NSP13 (SARS-CoV-2), and CACHE Challenge #4, targeting the tyrosine kinase-binding domain of CBLB, both pockets lacking established pharmacology and representing extreme hit-finding challenges. Under blinded, independently validated conditions, V-SYNTHES2 ranked among the top-performing submissions: the 8% hit rate for NSP13 exceeded the field average of 2.3% and placed the approach among the top three workflows, while for CBLB, one compound meeting predefined hit criteria was identified. These results demonstrate that V-SYNTHES2 maintains robust performance at giga-scale on ligand-depleted targets, precisely the conditions where data-driven approaches would face fundamental limitations, and establish a quantitative performance baseline for synthon-based screening of hundred-billion-compound chemical spaces.","PeriodicalId":520448,"journal":{"name":"NPJ drug discovery","volume":"3 1","pages":""},"PeriodicalIF":0.0,"publicationDate":"2026-09-02","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC13538404/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148883329","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0