Pub Date : 2026-03-01Epub Date: 2026-02-03DOI: 10.1016/j.compbiomed.2026.111507
Vishnu Malakar , S.P. Dhanabal , Dhritiman Roy , Chandi C. Malakar , Pratik Khona , Antony Justin
Mangifera indica has been utilized as an adjunct therapy for Alzheimer's disease (AD) due to its anti-Alzheimer's phytoconstituents. However, the underlying molecular mechanisms remain largely elusive. This research aimed to investigate the mechanism of action of Mangifera indica phytoconstituents in AD therapy. Anti-Alzheimer's phytoconstituents were identified from the literature and database, their related targets and associated pathways relevant to AD. Protein-protein interaction (PPI) networks were constructed using the STRING database and visualised through Cytoscape software. Target cluster module analysis was performed using the MCODE plugin in Cytoscape. Additionally, Gene Ontology and KEGG analyses were conducted to identify targets associated with Mangifera indica and AD. Furthermore, computational studies were conducted using AutoDock Vina tools, GROMACS, and Gaussian software. In this study, 15 active phytoconstituents and their 157 common targets were analysed. Based on topological parameters such as degree, closeness, and betweenness, the top five targets: Nrf2, Keap1, GSK-3β, APP, and PTPN1 were identified as critical nodes associated with regulation of Nrf2 signalling involving Keap1 and GSK-3β in the context of AD therapy. Molecular docking, MD simulations (1000 ns), PCA, DFT, and MM-PBSA analyses of Nrf2, Keap1, and GSK-3β demonstrated that the compound Mangiferin exhibited favourable predicted binding, stable interaction behaviour, and consistent equilibrium dynamics in comparison with reference ligands. This research highlights that Mangifera indica-related AD therapy involves a complex interplay of multiple phytoconstituents, molecular targets, and signalling pathways and offers significant molecular insights of Mangifera indica into potential antioxidant, anti-inflammatory, and neuroprotective mechanisms relevant to neuronal cells.
{"title":"Network pharmacology and computational-based approaches to activate NRF2 pathway via KEAP1 and GSK-3β inhibition: Exploring the possible molecular insights of mangiferin for Alzheimer's","authors":"Vishnu Malakar , S.P. Dhanabal , Dhritiman Roy , Chandi C. Malakar , Pratik Khona , Antony Justin","doi":"10.1016/j.compbiomed.2026.111507","DOIUrl":"10.1016/j.compbiomed.2026.111507","url":null,"abstract":"<div><div><em>Mangifera indica</em> has been utilized as an adjunct therapy for Alzheimer's disease (AD) due to its anti-Alzheimer's phytoconstituents. However, the underlying molecular mechanisms remain largely elusive. This research aimed to investigate the mechanism of action of <em>Mangifera indica</em> phytoconstituents in AD therapy. Anti-Alzheimer's phytoconstituents were identified from the literature and database, their related targets and associated pathways relevant to AD. Protein-protein interaction (PPI) networks were constructed using the STRING database and visualised through Cytoscape software. Target cluster module analysis was performed using the MCODE plugin in Cytoscape. Additionally, Gene Ontology and KEGG analyses were conducted to identify targets associated with <em>Mangifera indica</em> and AD. Furthermore, computational studies were conducted using AutoDock Vina tools, GROMACS, and Gaussian software. In this study, 15 active phytoconstituents and their 157 common targets were analysed. Based on topological parameters such as degree, closeness, and betweenness, the top five targets: Nrf2, Keap1, GSK-3β, APP, and PTPN1 were identified as critical nodes associated with regulation of Nrf2 signalling involving Keap1 and GSK-3β in the context of AD therapy. Molecular docking, MD simulations (1000 ns), PCA, DFT, and MM-PBSA analyses of Nrf2, Keap1, and GSK-3β demonstrated that the compound Mangiferin exhibited favourable predicted binding, stable interaction behaviour, and consistent equilibrium dynamics in comparison with reference ligands. This research highlights that <em>Mangifera indica</em>-related AD therapy involves a complex interplay of multiple phytoconstituents, molecular targets, and signalling pathways and offers significant molecular insights of <em>Mangifera indica</em> into potential antioxidant, anti-inflammatory, and neuroprotective mechanisms relevant to neuronal cells.</div></div>","PeriodicalId":10578,"journal":{"name":"Computers in biology and medicine","volume":"204 ","pages":"Article 111507"},"PeriodicalIF":6.3,"publicationDate":"2026-03-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"146118183","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2026-03-01Epub Date: 2026-02-02DOI: 10.1016/j.compbiomed.2026.111506
Noor Rahman , Humaira Zafar , Thirugnanasambandam Rajendran , Ruby Sharif , Ahmed Almehdi , Atia tul-Wahab , Sumbla Sheikh , M. Iqbal Choudhary
Acute myeloid leukemia (AML) is the predominant form of acute leukemia, affecting elderly individuals, typically diagnosed at an average age of 68 years. AML cells rely on the Bcl-2 protein for their survival. Overexpression of Bcl-2 protein in various cancer types renders it as a potential candidate for targeted therapies. The present study aimed to identify natural compounds as Bcl-2 inhibitors using in vitro, biophysical, and integrated computational approaches. The MTT assay was performed for cell proliferation, followed by apoptosis and gene expression analysis. STD-NMR spectroscopy, molecular docking and molecular dynamics simulations were performed for protein-ligand interactions. In the in vitro anti-proliferative assay, three natural compounds, gossypol (1), camptothecin (2), and jaceidin (3), were found active against the HL-60 cell line with IC50 concentrations of 1.634 ± 0.072, 0.137 ± 0.029, and 13.492 ± 2.292 μM, respectively. These compounds triggered apoptosis and decreased cellular viability in a dose-dependent manner. The gene expression analysis of Bax, Bcl-2, and Caspase 3 in HL-60 cells revealed that these compounds induce apoptosis by regulating essential apoptotic genes. Among the three identified potential hits, only gossypol (1) was buffer soluble and subjected to STD-NMR experiment to evaluate its protein-ligand interactions. Furthermore, molecular docking, binding free energies and MD simulation analyses demonstrated stable interactions of these compounds with the Bcl-2 protein. The three natural products showed potent to significant activity, effectively inducing apoptosis in the HL-60 cell line. Hence, this study identifies three potential lead candidates for drug discovery against Bcl-2-related cancers after further mechanistic and pre-clinical studies.
{"title":"Exploring natural products as Bcl-2 inhibitors for acute myeloid leukemia therapy using In vitro, STD-NMR spectroscopy, and In silico approaches","authors":"Noor Rahman , Humaira Zafar , Thirugnanasambandam Rajendran , Ruby Sharif , Ahmed Almehdi , Atia tul-Wahab , Sumbla Sheikh , M. Iqbal Choudhary","doi":"10.1016/j.compbiomed.2026.111506","DOIUrl":"10.1016/j.compbiomed.2026.111506","url":null,"abstract":"<div><div>Acute myeloid leukemia (AML) is the predominant form of acute leukemia, affecting elderly individuals, typically diagnosed at an average age of 68 years. AML cells rely on the Bcl-2 protein for their survival. Overexpression of Bcl-2 protein in various cancer types renders it as a potential candidate for targeted therapies. The present study aimed to identify natural compounds as Bcl-2 inhibitors using <em>in vitro</em>, biophysical, and integrated computational approaches. The MTT assay was performed for cell proliferation, followed by apoptosis and gene expression analysis. STD-NMR spectroscopy, molecular docking and molecular dynamics simulations were performed for protein-ligand interactions. In the <em>in vitro</em> anti-proliferative assay, three natural compounds, gossypol (<strong>1</strong>), camptothecin (<strong>2</strong>), and jaceidin (<strong>3</strong>), were found active against the HL-60 cell line with IC<sub>50</sub> concentrations of 1.634 ± 0.072, 0.137 ± 0.029, and 13.492 ± 2.292 μM, respectively. These compounds triggered apoptosis and decreased cellular viability in a dose-dependent manner. The gene expression analysis of <em>Bax</em>, <em>Bcl-2</em>, and <em>Caspase 3</em> in HL-60 cells revealed that these compounds induce apoptosis by regulating essential apoptotic genes. Among the three identified potential hits, only gossypol (<strong>1</strong>) was buffer soluble and subjected to STD-NMR experiment to evaluate its protein-ligand interactions. Furthermore, molecular docking, binding free energies and MD simulation analyses demonstrated stable interactions of these compounds with the Bcl-2 protein. The three natural products showed potent to significant activity, effectively inducing apoptosis in the HL-60 cell line. Hence, this study identifies three potential lead candidates for drug discovery against Bcl-2-related cancers after further mechanistic and pre-clinical studies.</div></div>","PeriodicalId":10578,"journal":{"name":"Computers in biology and medicine","volume":"204 ","pages":"Article 111506"},"PeriodicalIF":6.3,"publicationDate":"2026-03-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"146112384","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Cigarette smoking is known to cause airway remodelling leading to loss of lung plasticity, a key feature of chronic obstructive pulmonary disease (COPD). Despite the availability of several disease management approaches, an effective cure is elusive due to a lack of clear molecular insight into COPD pathogenesis. Thus, utilizing bioinformatics tools, this study aimed to identify crucial hub genes in COPD pathogenesis and validate them using in-vitro experiments and COPD patient samples.
Study methodology
In-silico identification of molecular interactions was analysed using bioinformatics tools like String, GEO datasets, CTD, Genecards, Disgenet, Opentargets, and Cytoscape. Airway epithelial cells (AECs) were exposed to different concentrations of cigarette smoke extract (CSE), followed by assessments of fibrosis and EMT-related parameters and markers using cellular and molecular biology techniques such as the MTT assay, AO/EtBr assay, trypan blue assay, the migration and invasion assays, morphological analysis, immunoblotting, immunocytochemistry, and RT-qPCR. Further, key genes expression and cytokines profile were assessed in PBMCs and plasma from COPD patients and healthy volunteers via RT-qPCR and ELISA, respectively.
Key findings
Four online databases (CTD, Genecards, Opentargets, and Disgenet) and a clinical dataset from the Gene Expression Omnibus were utilized to identify upregulated differentially expressed genes (DEGs). Subsequently, ten hub genes for COPD were identified using MCODE and cytohubba indices of Cytoscape, of which NOTCH3 and matrix metalloprotease (MMP) 2 were selected for further validation owing to their crucial role in COPD. CSE exposure of AECs caused alteration in cellular morphology, induced fibrous phenotype, upregulation of fibrosis and EMT markers, and increased expression of NOTCH3 and MMP2. Furthermore, chemical inhibition of MMP2 downregulated NOTCH3, suggesting NOTCH pathway upregulation by CSE-induced MMP2 activation. Inhibition of either MMP2 or NOTCH3 reversed CSE-induced fibrotic or EMT-related changes in AECs. PBMCs derived from COPD patients showed modulation of NOTCH3 and MMP2. JAG1, a NOTCH ligand, and many inflammatory markers were also significantly upregulated in COPD patient samples compared to healthy volunteers.
Significance
Our multi-level holistic approach, combining in-silico and in-vitro studies elucidated that MMP2 and NOTCH3 could be key mediators in CSE-induced airway epithelial cell remodelling, which was also confirmed through COPD patients’ sample analysis. We, thus, identify MMP2 and NOTCH3 as important gene targets for controlling CS-induced COPD pathophysiology.
{"title":"Systems medicine approach unravels MMP2 and NOTCH3 as key mediators of cigarette smoke-induced airway remodelling in COPD","authors":"Anupama Dubey , Md Shamim Akhtar , Anamika , Suneel Kateriya , Umesh C.S. Yadav","doi":"10.1016/j.compbiomed.2026.111508","DOIUrl":"10.1016/j.compbiomed.2026.111508","url":null,"abstract":"<div><h3>Background and aim</h3><div>Cigarette smoking is known to cause airway remodelling leading to loss of lung plasticity, a key feature of chronic obstructive pulmonary disease (COPD). Despite the availability of several disease management approaches, an effective cure is elusive due to a lack of clear molecular insight into COPD pathogenesis. Thus, utilizing bioinformatics tools, this study aimed to identify crucial hub genes in COPD pathogenesis and validate them using in-vitro experiments and COPD patient samples.</div></div><div><h3>Study methodology</h3><div><em>In-silico</em> identification of molecular interactions was analysed using bioinformatics tools like String, GEO datasets, CTD, Genecards, Disgenet, Opentargets, and Cytoscape. Airway epithelial cells (AECs) were exposed to different concentrations of cigarette smoke extract (CSE), followed by assessments of fibrosis and EMT-related parameters and markers using cellular and molecular biology techniques such as the MTT assay, AO/EtBr assay, trypan blue assay, the migration and invasion assays, morphological analysis, immunoblotting, immunocytochemistry, and RT-qPCR. Further, key genes expression and cytokines profile were assessed in PBMCs and plasma from COPD patients and healthy volunteers via RT-qPCR and ELISA, respectively.</div></div><div><h3>Key findings</h3><div>Four online databases (CTD, Genecards, Opentargets, and Disgenet) and a clinical dataset from the Gene Expression Omnibus were utilized to identify upregulated differentially expressed genes (DEGs). Subsequently, ten hub genes for COPD were identified using MCODE and cytohubba indices of Cytoscape, of which NOTCH3 and matrix metalloprotease (MMP) 2 were selected for further validation owing to their crucial role in COPD. CSE exposure of AECs caused alteration in cellular morphology, induced fibrous phenotype, upregulation of fibrosis and EMT markers, and increased expression of NOTCH3 and MMP2. Furthermore, chemical inhibition of MMP2 downregulated NOTCH3<em>,</em> suggesting NOTCH pathway upregulation by CSE-induced MMP2 activation. Inhibition of either MMP2 or NOTCH3 reversed CSE-induced fibrotic or EMT-related changes in AECs. PBMCs derived from COPD patients showed modulation of NOTCH3 and MMP2. JAG1, a NOTCH ligand, and many inflammatory markers were also significantly upregulated in COPD patient samples compared to healthy volunteers.</div></div><div><h3>Significance</h3><div>Our multi-level holistic approach, combining <em>in-silico</em> and <em>in-vitro</em> studies elucidated that MMP2 and NOTCH3 could be key mediators in CSE-induced airway epithelial cell remodelling, which was also confirmed through COPD patients’ sample analysis. We, thus, identify MMP2 and NOTCH3 as important gene targets for controlling CS-induced COPD pathophysiology.</div></div>","PeriodicalId":10578,"journal":{"name":"Computers in biology and medicine","volume":"204 ","pages":"Article 111508"},"PeriodicalIF":6.3,"publicationDate":"2026-03-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"146112412","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2026-03-01Epub Date: 2026-02-03DOI: 10.1016/j.compbiomed.2026.111518
Md.Mahfuz Ahmed , Md.Maruf Hossain , Md.Rakibul Hasan Rakib , Ronok Hashan , Md.Touhid Hasan Nirob , Md.Khairul Islam
Stroke is one of the leading causes of mortality and long-term disability worldwide, primarily resulting from the sudden disruption of cerebral blood flow. Early and accurate diagnosis plays a crucial role in minimizing neurological damage and improving recovery outcomes. This study proposes a comprehensive multimodal framework integrating a hybrid Swin Transformer–Bidirectional Long Short-Term Memory (SwinT–BiLSTM) model and an ensemble learning-based classifier for automated stroke detection and risk prediction from medical image and tabular clinical data. This study utilizes two brain stroke Computed Tomography (CT) datasets, including a primary dataset named BrSCTHD-2025, collected from hospitals in Dhaka and Faridpur, Bangladesh, and a secondary Kaggle CT dataset. In addition, a primary clinical tabular dataset was collected from Kushtia Medical College Hospital for multimodal analysis. The proposed SwinT–BiLSTM model efficiently extracts global spatial and sequential dependencies from CT images, while the ensemble classifier predicts stroke risk based on clinical and lifestyle parameters. Experimental results demonstrate that the model achieves 98% accuracy with an AUC of 1.00 on the BrSCTHD-2025 dataset and 97% accuracy with an AUC of 0.99 on the secondary Kaggle dataset, outperforming standalone SwinT by 2.5% and Convolutional Neural Network (CNN) architectures such as VGG16 and ResNet50 by 3%–4%. The ensemble classifier trained on tabular data achieved 80.36% accuracy, identifying critical stroke risk factors such as heart disease, prolonged sitting duration, and cholesterol level. Furthermore, Explainable Artificial Intelligence (XAI) techniques such as LIME, SHAP, enhanced Grad-CAM, and attention maps enhance interpretability by identifying the most influential visual and clinical features. Overall, the proposed SwinT–BiLSTM–Ensemble framework establishes a robust foundation for accurate, interpretable, and clinically reliable stroke diagnosis and personalized risk assessment in real-world healthcare environments.
{"title":"A hybrid swin transformer–BiLSTM framework and ensemble learning for multimodal brain stroke detection and risk prediction","authors":"Md.Mahfuz Ahmed , Md.Maruf Hossain , Md.Rakibul Hasan Rakib , Ronok Hashan , Md.Touhid Hasan Nirob , Md.Khairul Islam","doi":"10.1016/j.compbiomed.2026.111518","DOIUrl":"10.1016/j.compbiomed.2026.111518","url":null,"abstract":"<div><div>Stroke is one of the leading causes of mortality and long-term disability worldwide, primarily resulting from the sudden disruption of cerebral blood flow. Early and accurate diagnosis plays a crucial role in minimizing neurological damage and improving recovery outcomes. This study proposes a comprehensive multimodal framework integrating a hybrid Swin Transformer–Bidirectional Long Short-Term Memory (SwinT–BiLSTM) model and an ensemble learning-based classifier for automated stroke detection and risk prediction from medical image and tabular clinical data. This study utilizes two brain stroke Computed Tomography (CT) datasets, including a primary dataset named BrSCTHD-2025, collected from hospitals in Dhaka and Faridpur, Bangladesh, and a secondary Kaggle CT dataset. In addition, a primary clinical tabular dataset was collected from Kushtia Medical College Hospital for multimodal analysis. The proposed SwinT–BiLSTM model efficiently extracts global spatial and sequential dependencies from CT images, while the ensemble classifier predicts stroke risk based on clinical and lifestyle parameters. Experimental results demonstrate that the model achieves 98% accuracy with an AUC of 1.00 on the BrSCTHD-2025 dataset and 97% accuracy with an AUC of 0.99 on the secondary Kaggle dataset, outperforming standalone SwinT by 2.5% and Convolutional Neural Network (CNN) architectures such as VGG16 and ResNet50 by 3%–4%. The ensemble classifier trained on tabular data achieved 80.36% accuracy, identifying critical stroke risk factors such as heart disease, prolonged sitting duration, and cholesterol level. Furthermore, Explainable Artificial Intelligence (XAI) techniques such as LIME, SHAP, enhanced Grad-CAM, and attention maps enhance interpretability by identifying the most influential visual and clinical features. Overall, the proposed SwinT–BiLSTM–Ensemble framework establishes a robust foundation for accurate, interpretable, and clinically reliable stroke diagnosis and personalized risk assessment in real-world healthcare environments.</div></div>","PeriodicalId":10578,"journal":{"name":"Computers in biology and medicine","volume":"204 ","pages":"Article 111518"},"PeriodicalIF":6.3,"publicationDate":"2026-03-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"146118139","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Toxin–antitoxin systems are central to bacterial persistence, promoting drug tolerance and infection relapse, and therefore demand a clear mechanistic understanding of their regulation. It is thus intriguing to investigate the possible routes to persister cell formation through mathematical modelling and to assess whether their emergence can be anticipated using statistical measures. For this dual purpose, a mathematical model describing the fundamental biochemical interactions among the operon, mRNA, toxin, antitoxin, and two associated protein complexes is considered in this study. The uncertainty in the steady-state behaviour of the deterministic model outcomes is analysed using two complementary forms of global sensitivity analysis. Both these techniques identify six key parameters that substantially influence transcription, translation, and the turnover of antitoxins. Among these, the parameter controlling the quadratic repression of antitoxin through toxin binding has opposite effects on the two species, thereby driving hysteresis between alternate physiological states. Intrinsic noise is introduced into the deterministic model via the chemical master equation. Subsequent Gillespie simulations reveal a critical transition from normal to persister cells, which is then detected using twelve multivariate statistical indicators within moving- and expanding-window frameworks. Sensitivity analyses define hyperparameter ranges that ensure reliable predictions, and robustness tests across repeated simulations show consistent performance for most moving-window indicators, except for some variance–covariance and information-based measures. The expanding-window approach reveals different types of warnings—flickering, sustained, and spurious—quantified by true-positive rates, lead times, and total warning counts. Together, these results demonstrate that multivariate measures can reliably predict critical transitions and provide a solid framework for understanding the loss of resilience in complex biological systems.
{"title":"Parameter sensitivity and critical transition anticipation in bistable toxin-antitoxin dynamics","authors":"Shankha Narayan Chattopadhyay , Inayat Ullah Irshad , Ajeet K. Sharma , Arvind Kumar Gupta","doi":"10.1016/j.compbiomed.2026.111535","DOIUrl":"10.1016/j.compbiomed.2026.111535","url":null,"abstract":"<div><div>Toxin–antitoxin systems are central to bacterial persistence, promoting drug tolerance and infection relapse, and therefore demand a clear mechanistic understanding of their regulation. It is thus intriguing to investigate the possible routes to persister cell formation through mathematical modelling and to assess whether their emergence can be anticipated using statistical measures. For this dual purpose, a mathematical model describing the fundamental biochemical interactions among the operon, mRNA, toxin, antitoxin, and two associated protein complexes is considered in this study. The uncertainty in the steady-state behaviour of the deterministic model outcomes is analysed using two complementary forms of global sensitivity analysis. Both these techniques identify six key parameters that substantially influence transcription, translation, and the turnover of antitoxins. Among these, the parameter controlling the quadratic repression of antitoxin through toxin binding has opposite effects on the two species, thereby driving hysteresis between alternate physiological states. Intrinsic noise is introduced into the deterministic model via the chemical master equation. Subsequent Gillespie simulations reveal a critical transition from normal to persister cells, which is then detected using twelve multivariate statistical indicators within moving- and expanding-window frameworks. Sensitivity analyses define hyperparameter ranges that ensure reliable predictions, and robustness tests across repeated simulations show consistent performance for most moving-window indicators, except for some variance–covariance and information-based measures. The expanding-window approach reveals different types of warnings—flickering, sustained, and spurious—quantified by true-positive rates, lead times, and total warning counts. Together, these results demonstrate that multivariate measures can reliably predict critical transitions and provide a solid framework for understanding the loss of resilience in complex biological systems.</div></div>","PeriodicalId":10578,"journal":{"name":"Computers in biology and medicine","volume":"204 ","pages":"Article 111535"},"PeriodicalIF":6.3,"publicationDate":"2026-03-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"146141350","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2026-03-01Epub Date: 2026-02-10DOI: 10.1016/j.compbiomed.2026.111534
Joonhyeon Park , Kyubo Shin , Jongchan Kim , Jaemin Park , Jae Hoon Moon , JaeSang Ko
{"title":"Comment on: “Multimodal large language models as assistance for evaluation of thyroid-associated ophthalmopathy”","authors":"Joonhyeon Park , Kyubo Shin , Jongchan Kim , Jaemin Park , Jae Hoon Moon , JaeSang Ko","doi":"10.1016/j.compbiomed.2026.111534","DOIUrl":"10.1016/j.compbiomed.2026.111534","url":null,"abstract":"","PeriodicalId":10578,"journal":{"name":"Computers in biology and medicine","volume":"204 ","pages":"Article 111534"},"PeriodicalIF":6.3,"publicationDate":"2026-03-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"146164567","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
MmpL3 protein plays a vital role in cell wall synthesis in Mycobacterium. Novel benzoxazole carboxamide derivatives were designed to inhibit cell wall formation by targeting the MmpL3 and combat tuberculosis. Fourteen benzoxazole carboxamide derivatives (BXZ-I to BXZ-XIV) were synthesised, and their structures were confirmed using both experimental and computational methods. Techniques such as molecular docking, ADME, toxicity prediction, deep learning-based docking, and molecular dynamics simulation were used to analyse these compounds. Molecules with promising antimycobacterial activity were selected for MDS, MM-GBSA, and FEP analyses. BXZ-IX and BXZ-XIV exhibited potent activity against Mycobacterium smegmatis, with a minimum inhibitory concentration (MIC) of 15.62 μg/mL, compared with SQ109 (standard MmpL3 inhibitor), which had an MIC of 10.0 μg/mL. Overall, ten of the selected benzoxazole compounds significantly inhibited the growth of M. smegmatis, with MICs ranging from 15.62 to 62.5 μg/mL in laboratory tests, demonstrating greater effectiveness against the MmpL3 protein.
{"title":"Molecular modelling assisted identification of novel Benzoxazole derivatives as hit molecules targeting Mycobacterial Membrane Protein Large 3 (MmpL3)","authors":"Rupesh Chikhale , Vikramsinh Sardarsinh Suryawanshi , Shweta Sharma , Vivek Kumar Gupta , Pramod B. Khedekar","doi":"10.1016/j.compbiomed.2026.111521","DOIUrl":"10.1016/j.compbiomed.2026.111521","url":null,"abstract":"<div><div>MmpL3 protein plays a vital role in cell wall synthesis in Mycobacterium. Novel benzoxazole carboxamide derivatives were designed to inhibit cell wall formation by targeting the MmpL3 and combat tuberculosis. Fourteen benzoxazole carboxamide derivatives (BXZ-I to BXZ-XIV) were synthesised, and their structures were confirmed using both experimental and computational methods. Techniques such as molecular docking, ADME, toxicity prediction, deep learning-based docking, and molecular dynamics simulation were used to analyse these compounds. Molecules with promising antimycobacterial activity were selected for MDS, MM-GBSA, and FEP analyses. BXZ-IX and BXZ-XIV exhibited potent activity against <em>Mycobacterium smegmatis,</em> with a minimum inhibitory concentration (MIC) of 15.62 μg/mL, compared with SQ109 (standard MmpL3 inhibitor), which had an MIC of 10.0 μg/mL. Overall, ten of the selected benzoxazole compounds significantly inhibited the growth of <em>M. smegmatis</em>, with MICs ranging from 15.62 to 62.5 μg/mL in laboratory tests, demonstrating greater effectiveness against the MmpL3 protein.</div></div>","PeriodicalId":10578,"journal":{"name":"Computers in biology and medicine","volume":"204 ","pages":"Article 111521"},"PeriodicalIF":6.3,"publicationDate":"2026-03-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"146104231","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2026-02-15Epub Date: 2026-01-21DOI: 10.1016/j.compbiomed.2026.111471
Sajid Naveed , Mujtaba Husnain , Najah Alsubaie
A variety of AI-based approaches have been employed to analyze complex genomic datasets. Predicting the synergy of drug combinations is a critical step toward optimizing cancer treatment by identifying the most effective drug pairs. This study presents HybridDeepSynergy, a novel hybrid deep learning model that integrates Convolutional Neural Networks (CNNs), Long Short-Term Memory (LSTM), and Transformer attention mechanisms to predict drug synergy across diverse drug combinations and cancer cell lines. The model is designed to enhance precision medicine and cancer treatment outcomes.
HybridDeepSynergy leverages CNNs to capture local feature interactions, LSTMs to model sequential dependencies, and attention mechanisms to extract long-range relationships within the data. The model was trained and evaluated on a comprehensive dataset containing numerous drug combinations, using five established synergy scoring models: Bliss Independence (BLISS), Zero Interaction Potency (ZIP), Loewe Additivity (LOEWE), Highest Single Agent (HSA), and General Synergy (S).
Our model demonstrated superior performance compared to existing approaches, achieving a lower Root Mean Squared Error (RMSE = 3.911) and Mean Absolute Error (MAE = 2.922), along with higher coefficients of determination ( = 0.953), Pearson correlation (0.917), and Spearman correlation (0.886). These results confirm its predictive efficiency and consistency across multiple synergy scoring models. Furthermore, the incorporation of attention mechanisms provides interpretability by highlighting significant features associated with drug resistance.
Future work will focus on incorporating additional cancer datasets, enhancing model predictive capabilities, and validating the approach in clinical settings to support personalized medicine. The findings suggest that HybridDeepSynergy has the potential to substantially improve treatment strategies for cancer and may be applicable to other disease contexts.
{"title":"HybridDeepSynergy: A hybrid deep learning model integrating CNN, LSTM, and attention mechanisms for cancer drug synergy prediction","authors":"Sajid Naveed , Mujtaba Husnain , Najah Alsubaie","doi":"10.1016/j.compbiomed.2026.111471","DOIUrl":"10.1016/j.compbiomed.2026.111471","url":null,"abstract":"<div><div>A variety of AI-based approaches have been employed to analyze complex genomic datasets. Predicting the synergy of drug combinations is a critical step toward optimizing cancer treatment by identifying the most effective drug pairs. This study presents HybridDeepSynergy, a novel hybrid deep learning model that integrates Convolutional Neural Networks (CNNs), Long Short-Term Memory (LSTM), and Transformer attention mechanisms to predict drug synergy across diverse drug combinations and cancer cell lines. The model is designed to enhance precision medicine and cancer treatment outcomes.</div><div>HybridDeepSynergy leverages CNNs to capture local feature interactions, LSTMs to model sequential dependencies, and attention mechanisms to extract long-range relationships within the data. The model was trained and evaluated on a comprehensive dataset containing numerous drug combinations, using five established synergy scoring models: Bliss Independence (BLISS), Zero Interaction Potency (ZIP), Loewe Additivity (LOEWE), Highest Single Agent (HSA), and General Synergy (S).</div><div>Our model demonstrated superior performance compared to existing approaches, achieving a lower Root Mean Squared Error (RMSE = 3.911) and Mean Absolute Error (MAE = 2.922), along with higher coefficients of determination (<span><math><msup><mi>R</mi><mn>2</mn></msup></math></span> = 0.953), Pearson correlation (0.917), and Spearman correlation (0.886). These results confirm its predictive efficiency and consistency across multiple synergy scoring models. Furthermore, the incorporation of attention mechanisms provides interpretability by highlighting significant features associated with drug resistance.</div><div>Future work will focus on incorporating additional cancer datasets, enhancing model predictive capabilities, and validating the approach in clinical settings to support personalized medicine. The findings suggest that HybridDeepSynergy has the potential to substantially improve treatment strategies for cancer and may be applicable to other disease contexts.</div></div>","PeriodicalId":10578,"journal":{"name":"Computers in biology and medicine","volume":"203 ","pages":"Article 111471"},"PeriodicalIF":6.3,"publicationDate":"2026-02-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"146028663","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2026-02-15Epub Date: 2026-01-21DOI: 10.1016/j.compbiomed.2026.111467
Zeyuan Song , Xiao-Cong Zhen
Large language models (LLMs) have demonstrated impressive proficiency in various science and engineering applications. However, due to the innate multi-scale property of biological systems, existing LLMs face severe limitations in capturing hierarchical relationships and context-dependent interactions across molecular, cellular, tissue, and systemic levels. These models often lack the architectural mechanisms needed to reason effectively across different biological scales, resulting in reduced accuracy and limited interpretability when applied to complex tasks. Here, we introduce a novel framework named multi-scale chain-of-thought fusion (MS-CoTF), which fuses reasoning at molecular, cellular, tissue, and system scales to enhance accuracy and interpretability when solving biological tasks. Through adaptive reasoning depth control, multi-scale integration, bi-directional flow and dynamic fusion strategies, our MS-CoTF model effectively processes queries of varying complexity, enabling scalable and interpretable reasoning across multiple biological levels. Ablation studies demonstrate that these components function synergistically to enhance model accuracy while simultaneously providing biologically meaningful insights. Furthermore, our MS-CoTF model consistently outperforms state-of-the-art reasoning models by 10–15% across three benchmark problems and two case studies in terms of accuracy, expert ratings, and the capacity to produce reasonable inference chains. Technically, MS-CoTF orchestrates a frozen biomedical LLM backbone with trainable cross-scale modules, employing a precise definition of per-step chain-of-thought (CoT) construction and linking. To ensure rigorous evaluation, we implement an explicit dataset splitting protocol (entity-disjoint and temporal) and utilize the Reasoning Coherence Score strictly as a post-hoc metric to ensure fair comparisons. We further validate the framework through extended baselines, including structure-conditioned and multimodal biomedical LLMs, alongside detailed human evaluation protocols and hallucination stress tests.
{"title":"MS-CoTF: Multi-scale chain-of-thought fusion for interpretable biological reasoning with large language models","authors":"Zeyuan Song , Xiao-Cong Zhen","doi":"10.1016/j.compbiomed.2026.111467","DOIUrl":"10.1016/j.compbiomed.2026.111467","url":null,"abstract":"<div><div>Large language models (LLMs) have demonstrated impressive proficiency in various science and engineering applications. However, due to the innate multi-scale property of biological systems, existing LLMs face severe limitations in capturing hierarchical relationships and context-dependent interactions across molecular, cellular, tissue, and systemic levels. These models often lack the architectural mechanisms needed to reason effectively across different biological scales, resulting in reduced accuracy and limited interpretability when applied to complex tasks. Here, we introduce a novel framework named multi-scale chain-of-thought fusion (MS-CoTF), which fuses reasoning at molecular, cellular, tissue, and system scales to enhance accuracy and interpretability when solving biological tasks. Through adaptive reasoning depth control, multi-scale integration, bi-directional flow and dynamic fusion strategies, our MS-CoTF model effectively processes queries of varying complexity, enabling scalable and interpretable reasoning across multiple biological levels. Ablation studies demonstrate that these components function synergistically to enhance model accuracy while simultaneously providing biologically meaningful insights. Furthermore, our MS-CoTF model consistently outperforms state-of-the-art reasoning models by 10–15% across three benchmark problems and two case studies in terms of accuracy, expert ratings, and the capacity to produce reasonable inference chains. Technically, MS-CoTF orchestrates a frozen biomedical LLM backbone with trainable cross-scale modules, employing a precise definition of per-step chain-of-thought (CoT) construction and linking. To ensure rigorous evaluation, we implement an explicit dataset splitting protocol (entity-disjoint and temporal) and utilize the Reasoning Coherence Score strictly as a post-hoc metric to ensure fair comparisons. We further validate the framework through extended baselines, including structure-conditioned and multimodal biomedical LLMs, alongside detailed human evaluation protocols and hallucination stress tests.</div></div>","PeriodicalId":10578,"journal":{"name":"Computers in biology and medicine","volume":"203 ","pages":"Article 111467"},"PeriodicalIF":6.3,"publicationDate":"2026-02-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"146028623","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2026-02-15Epub Date: 2026-01-19DOI: 10.1016/j.compbiomed.2026.111484
Wu Yinhang , Zhuang Jing , Qu Zhanbo , Liu Jiang , Zhou Qing , Zhang Qi , Jin Yin , Song Jianwen , Wu Wei , Han Shuwen
Background
Being involved in the occurrence of colorectal cancer (CRC), gut microbes are potential targets for early diagnosis of CRC. Defining the threshold of these characteristic bacteria could provide a basis for the clinical application of microorganisms as novel tumor markers for CRC.
Objective
To sort out and define the threshold of related bacteria and the ecological characteristics of gut bacteria.
Methods
A total of 8021 fecal samples from healthy people and 497 from CRC patients in the public database were collected to analyse the reference range. CRC-related bacteria and gut microbial characteristics were screened by literature review and analysed. CRC related bacteria and 5–95 % medians of gut microbial characteristics in healthy populations were used as reference value. 16S rRNA Miseq sequencing (175 CRC patients and 175 healthy people) and PacBio sequencing (200 CRC patients and 200 healthy people) were used to detect stool DNA sequence. The community composition of gut microbiota between CRC and healthy subjects was plotted; the species differences were analysed by Lefse analysis. R studio software was used to analyse CRC-related bacteria and gut microbial characteristics.
Results
A total of 218 CRC-associated bacteria and 15 gut microbial characteristics, such as enterotypes and Firmicutes/Bacteroidetes ratio, were reviewed and analysed. A 5–95 % threshold for these 218 CRC-associated bacteria and 15 gut microbiome signatures was developed to provide criteria for the normal range of gut bacteria. The CRC evaluation intelligent system software was developed and it could quickly calculate the value of 218 CRC related bacteria and 15 gut microbial characteristics using sequencing data, and assess whether they are within the threshold. And this software has the function of predicting CRC risk. The accuracy of CRC risk assessment ranged from 89.14 % to 91.50 %.
Conclusion
We established, for the first time, quantitative thresholds for CRC-associated bacteria and have driven advances in microbial risk prediction for CRC.
{"title":"Establishment of threshold of human gut microbes and risk assessment system for colorectal cancer","authors":"Wu Yinhang , Zhuang Jing , Qu Zhanbo , Liu Jiang , Zhou Qing , Zhang Qi , Jin Yin , Song Jianwen , Wu Wei , Han Shuwen","doi":"10.1016/j.compbiomed.2026.111484","DOIUrl":"10.1016/j.compbiomed.2026.111484","url":null,"abstract":"<div><h3>Background</h3><div>Being involved in the occurrence of colorectal cancer (CRC), gut microbes are potential targets for early diagnosis of CRC. Defining the threshold of these characteristic bacteria could provide a basis for the clinical application of microorganisms as novel tumor markers for CRC.</div></div><div><h3>Objective</h3><div>To sort out and define the threshold of related bacteria and the ecological characteristics of gut bacteria.</div></div><div><h3>Methods</h3><div>A total of 8021 fecal samples from healthy people and 497 from CRC patients in the public database were collected to analyse the reference range. CRC-related bacteria and gut microbial characteristics were screened by literature review and analysed. CRC related bacteria and 5–95 % medians of gut microbial characteristics in healthy populations were used as reference value. 16S rRNA Miseq sequencing (175 CRC patients and 175 healthy people) and PacBio sequencing (200 CRC patients and 200 healthy people) were used to detect stool DNA sequence. The community composition of gut microbiota between CRC and healthy subjects was plotted; the species differences were analysed by Lefse analysis. R studio software was used to analyse CRC-related bacteria and gut microbial characteristics.</div></div><div><h3>Results</h3><div>A total of 218 CRC-associated bacteria and 15 gut microbial characteristics, such as enterotypes and Firmicutes/Bacteroidetes ratio, were reviewed and analysed. A 5–95 % threshold for these 218 CRC-associated bacteria and 15 gut microbiome signatures was developed to provide criteria for the normal range of gut bacteria. The CRC evaluation intelligent system software was developed and it could quickly calculate the value of 218 CRC related bacteria and 15 gut microbial characteristics using sequencing data, and assess whether they are within the threshold. And this software has the function of predicting CRC risk. The accuracy of CRC risk assessment ranged from 89.14 % to 91.50 %.</div></div><div><h3>Conclusion</h3><div>We established, for the first time, quantitative thresholds for CRC-associated bacteria and have driven advances in microbial risk prediction for CRC.</div></div>","PeriodicalId":10578,"journal":{"name":"Computers in biology and medicine","volume":"203 ","pages":"Article 111484"},"PeriodicalIF":6.3,"publicationDate":"2026-02-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"146008911","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}