Computational Biology and Chemistry最新文献_第5页

Integration of interpretable multi-features and multi-loss functions for multi-functional therapeutic peptide prediction via dataset construction 基于数据集构建的可解释多特征和多损失函数集成的多功能治疗肽预测

IF 3.1 4区生物学 Q2 BIOLOGY

Computational Biology and Chemistry

Pub Date : 2026-06-01 Epub Date: 2026-01-13 DOI: 10.1016/j.compbiolchem.2026.108893

Xinyi Wang , Minjie Zhou , Yunshan Su , Shunfang Wang

Multi-functional therapeutic peptides (MFTP) play a crucial role in drug development, exhibiting properties such as anti-cancer, anti-inflammatory effects, and more. Artificial intelligence-based predictors have been developed to identify MFTP and achieve satisfactory performance. However, these predictors often heavily rely on latent sequence features, overlooking physicochemical patterns and facing challenges with dataset imbalances. In this study, we propose MFTP-MFML, a model that combines interpretable multiple features and loss functions. Firstly, embedding features with positional information are input to the bi-directional long short-term memory (BiLSTM) network, generating latent representation information while preserving the original sequence information. Secondly, physicochemical attributes are utilized to supplement the amino acid composition and physicochemical properties across different functions of therapeutic peptides, and latent representation information are concatenated with these physicochemical attributes to enhance classification. Thirdly, addressing class imbalances and capturing label correlations, integration loss is employed, incorporating focal loss, binary cross entropy loss, and dice loss. Fourth, to enhance the diversity of functions of therapeutic peptides, MFTP-Mixed-90, a benchmark dataset comprising 27 functions, is constructed. Finally, to evaluate the performance of the model, we compare it with other methods on PrMFTP dataset and MFTP-Mixed-90 dataset. Experimental results demonstrate that MFTP-MFML outperforms existing methods, effectively utilizing integrated features and loss functions. Our code and the datasets are available at https://github.com/wongsing/MFTP-MFML.

多功能治疗肽（MFTP）在药物开发中发挥着至关重要的作用，具有抗癌、抗炎等作用。已经开发了基于人工智能的预测器来识别MFTP并获得令人满意的性能。然而，这些预测往往严重依赖于潜在的序列特征，忽视了物理化学模式，并面临着数据集不平衡的挑战。在这项研究中，我们提出了MFTP-MFML，一个结合了可解释的多特征和损失函数的模型。首先，将包含位置信息的嵌入特征输入到双向长短期记忆（BiLSTM）网络中，在保留原始序列信息的同时生成潜在的表示信息；其次，利用理化属性来补充不同功能治疗肽的氨基酸组成和理化性质，并将潜在表征信息与这些理化属性连接起来，增强分类能力；第三，解决类别不平衡和捕获标签相关性，采用积分损失，包括焦点损失，二元交叉熵损失和骰子损失。第四，为了增强治疗肽功能的多样性，构建了包含27个功能的基准数据集mftp - mix -90。最后，将该模型与其他方法在PrMFTP数据集和mftp - mix -90数据集上的性能进行了比较。实验结果表明，MFTP-MFML有效地利用了综合特征和损失函数，优于现有的方法。我们的代码和数据集可在https://github.com/wongsing/MFTP-MFML上获得。

{"title":"Integration of interpretable multi-features and multi-loss functions for multi-functional therapeutic peptide prediction via dataset construction","authors":"Xinyi Wang , Minjie Zhou , Yunshan Su , Shunfang Wang","doi":"10.1016/j.compbiolchem.2026.108893","DOIUrl":"10.1016/j.compbiolchem.2026.108893","url":null,"abstract":"<div><div>Multi-functional therapeutic peptides (MFTP) play a crucial role in drug development, exhibiting properties such as anti-cancer, anti-inflammatory effects, and more. Artificial intelligence-based predictors have been developed to identify MFTP and achieve satisfactory performance. However, these predictors often heavily rely on latent sequence features, overlooking physicochemical patterns and facing challenges with dataset imbalances. In this study, we propose MFTP-MFML, a model that combines interpretable multiple features and loss functions. Firstly, embedding features with positional information are input to the bi-directional long short-term memory (BiLSTM) network, generating latent representation information while preserving the original sequence information. Secondly, physicochemical attributes are utilized to supplement the amino acid composition and physicochemical properties across different functions of therapeutic peptides, and latent representation information are concatenated with these physicochemical attributes to enhance classification. Thirdly, addressing class imbalances and capturing label correlations, integration loss is employed, incorporating focal loss, binary cross entropy loss, and dice loss. Fourth, to enhance the diversity of functions of therapeutic peptides, MFTP-Mixed-90, a benchmark dataset comprising 27 functions, is constructed. Finally, to evaluate the performance of the model, we compare it with other methods on PrMFTP dataset and MFTP-Mixed-90 dataset. Experimental results demonstrate that MFTP-MFML outperforms existing methods, effectively utilizing integrated features and loss functions. Our code and the datasets are available at <span><span>https://github.com/wongsing/MFTP-MFML</span><svg><path></path></svg></span>.</div></div>","PeriodicalId":10616,"journal":{"name":"Computational Biology and Chemistry","volume":"122 ","pages":"Article 108893"},"PeriodicalIF":3.1,"publicationDate":"2026-06-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145974568","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"生物学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

AI-optimized sanguinarine derivatives inhibiting sortase A for combating AmpC β-lactamase resistance in Enterobacter cloacae: An integrated computational approach 人工智能优化的血氨酸衍生物抑制分类酶A对抗阴沟肠杆菌AmpC β-内酰胺酶耐药：一种综合计算方法

IF 3.1 4区生物学 Q2 BIOLOGY

Computational Biology and Chemistry

Pub Date : 2026-06-01 Epub Date: 2026-01-08 DOI: 10.1016/j.compbiolchem.2026.108891

Sonia Knawal , Ameer Mahmood Shaker , Mohammed Albahloul Rajab , Yatreb Omar Alkhbulli , Shifaa O. Alshammari , Emad Solouma , Mustafa Sabri Cheyad

The antimicrobial resistance is a serious health problem worldwide, and one of the causes of multidrug resistance is AmpC β-lactamase-producing Enterobacter cloacae. This work focused on analyzing the antibacterial properties of the plant-derived alkaloid sanguinarine and improve its derivatives with the help of artificial intelligence (AI) that may enhance antimicrobial activity. Therefore, molecular docking results demonstrate that AI-optimized Ligand 1 displayed the strongest binding affinity of −9.7 kcal/mol (AutoDock Vina) and −166.94 kcal/mol (HDOCK). Compared to Sanguinarine with a binding affinity of −9.2 kcal/mol. The lead AI-optimized Sanguinarine derivative with stable binding and good energetics was confirmed in molecular dynamics and in MMGBSA/MMPBSA analyses, suggesting that it may be a promising lead of AmpC β-lactamase inhibitors. Density Functional Theory (DFT) computations revealed that the lead AI-optimized compound had the HOMO-LUMO gap of 0.17089 eV and indicated moderate reactivity that would as a result of analysing the pharmacophore, key aromatic, hydrogen bond acceptor and hydrophobic sites were identified and the AI-optimized derivative was found to be a better drug-like assembly than natural Sanguinarine. The ADMET analysis showed potential lipophilicity, whole-GI absorption, and BBB permeability of the AI-optimized derivative and decreased toxicity in general, especially regarding neurotoxicity. The results indicate the possible improvement of the resistance to antibiotics using AI optimization that can aid in promoting the antimicrobial activity and safety set of Sanguinarine, which is a particularly promising additional tool that can be used to combat antibiotic resistance. These findings require further in vivo studies to validate the computational predictions that should prove their validity and confirm the possibility of using the results of trials with AI-optimized derivatives in clinical practice.

抗菌药物耐药性是世界范围内严重的健康问题，而产生AmpC β-内酰胺酶的阴沟肠杆菌是造成多重耐药的原因之一。本研究的重点是分析植物源生物碱血碱的抗菌特性，并利用人工智能技术对其衍生物进行改进，以期提高其抗菌活性。因此，分子对接结果表明，ai优化后的配体1的结合亲和力最强，分别为−9.7 kcal/mol （AutoDock Vina）和−166.94 kcal/mol （HDOCK）。结合亲和力为−9.2 kcal/mol。在分子动力学和MMGBSA/MMPBSA分析中证实，该ai优化先导物结合稳定，具有良好的能量学，可能是AmpC β-内酰胺酶抑制剂的先导物。密度泛函理论（DFT）计算结果表明，人工智能优化先导化合物的HOMO-LUMO间隙为0.17089 eV，反应性中等，通过对药团的分析，确定了关键的芳香族、氢键受体和疏水位点，人工智能优化的衍生物比天然血根碱具有更好的类药物组装性。ADMET分析显示ai优化的衍生物具有潜在的亲脂性、全gi吸收和血脑屏障通透性，总体毒性降低，尤其是神经毒性。结果表明，人工智能优化可能会改善抗生素耐药性，有助于提高Sanguinarine的抗菌活性和安全性，这是一个特别有前途的额外工具，可用于对抗抗生素耐药性。这些发现需要进一步的体内研究来验证计算预测，以证明其有效性，并确认在临床实践中使用人工智能优化衍生物的试验结果的可能性。

{"title":"AI-optimized sanguinarine derivatives inhibiting sortase A for combating AmpC β-lactamase resistance in Enterobacter cloacae: An integrated computational approach","authors":"Sonia Knawal , Ameer Mahmood Shaker , Mohammed Albahloul Rajab , Yatreb Omar Alkhbulli , Shifaa O. Alshammari , Emad Solouma , Mustafa Sabri Cheyad","doi":"10.1016/j.compbiolchem.2026.108891","DOIUrl":"10.1016/j.compbiolchem.2026.108891","url":null,"abstract":"<div><div>The antimicrobial resistance is a serious health problem worldwide, and one of the causes of multidrug resistance is <em>AmpC β-lactamase</em>-producing <em>Enterobacter cloacae</em>. This work focused on analyzing the antibacterial properties of the plant-derived alkaloid sanguinarine and improve its derivatives with the help of artificial intelligence (AI) that may enhance antimicrobial activity. Therefore, molecular docking results demonstrate that AI-optimized Ligand 1 displayed the strongest binding affinity of −9.7 kcal/mol (AutoDock Vina) and −166.94 kcal/mol (HDOCK). Compared to Sanguinarine with a binding affinity of −9.2 kcal/mol. The lead AI-optimized Sanguinarine derivative with stable binding and good energetics was confirmed in molecular dynamics and in MMGBSA/MMPBSA analyses, suggesting that it may be a promising lead of <em>AmpC β-lactamase</em> inhibitors. Density Functional Theory (DFT) computations revealed that the lead AI-optimized compound had the HOMO-LUMO gap of 0.17089 eV and indicated moderate reactivity that would as a result of analysing the pharmacophore, key aromatic, hydrogen bond acceptor and hydrophobic sites were identified and the AI-optimized derivative was found to be a better drug-like assembly than natural Sanguinarine. The ADMET analysis showed potential lipophilicity, whole-GI absorption, and BBB permeability of the AI-optimized derivative and decreased toxicity in general, especially regarding neurotoxicity. The results indicate the possible improvement of the resistance to antibiotics using AI optimization that can aid in promoting the antimicrobial activity and safety set of Sanguinarine, which is a particularly promising additional tool that can be used to combat antibiotic resistance. These findings require further in vivo studies to validate the computational predictions that should prove their validity and confirm the possibility of using the results of trials with AI-optimized derivatives in clinical practice.</div></div>","PeriodicalId":10616,"journal":{"name":"Computational Biology and Chemistry","volume":"122 ","pages":"Article 108891"},"PeriodicalIF":3.1,"publicationDate":"2026-06-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145974569","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"生物学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Cross-omics interpretable neural network for discovery of molecular markers in prostate cancer 交叉组学可解释神经网络用于前列腺癌分子标记的发现

IF 3.1 4区生物学 Q2 BIOLOGY

Computational Biology and Chemistry

Pub Date : 2026-06-01 Epub Date: 2026-01-08 DOI: 10.1016/j.compbiolchem.2026.108879

Xin Chen , Sheng Yi , Anwaier Yuemaierabola , Yuhan Liu , Liang He , Jing Ma , Wenjia Guo , Gang Sun

Determining molecular markers that mediate clinically aggressive phenotypes in prostate cancer is a significant challenge. While traditional linear models offer some interpretability, they often lack the precision needed for complex multi-omics data. Conversely, conventional deep learning methods provide robust predictions but typically remain opaque, hindering the identification of impactful molecular markers and biological mechanisms. To address this, we propose the Cross-omics Interpretable Neural Network (CINN), a biomimetic framework designed to predict prostate cancer states and identify key molecular markers by integrating diverse omics data.

CINN innovatively leverages prior biological knowledge from either pathway or protein–protein interaction (PPI) networks, combined with a novel trainable mask layer. This mask dynamically optimizes the strength of pre-defined biological connections, thereby enhancing both knowledge representation and model interpretability. The framework effectively integrates multi-omics data, including gene expression, somatic mutations, and copy number variations, to provide a holistic view of the disease.

Extensive experiments on a prostate cancer dataset demonstrate that CINN achieves substantial and statistically significant performance enhancements over a strong baseline (P-NET). Specifically, our best-performing variant, CINN-pw with a trainable mask, improved F1 scores by 13.1% to 0.843, Accuracy by 8.3% to 0.894, and AUC by 2.3% to 0.949. These gains were consistently statistically significant (

p < 0.0001

for most key metrics), underscoring the robustness of our approach. Crucially, CINN’s inherent interpretability facilitated the identification of pivotal molecular candidates, including TBP and TAF2, which are implicated in prostate cancer progression. These findings are supported by existing literature and provide valuable insights into the underlying mechanisms of prostate cancer, offering potential avenues for targeted therapeutic interventions and precision medicine.

确定介导前列腺癌临床侵袭性表型的分子标记是一个重大挑战。虽然传统的线性模型提供了一些可解释性，但它们往往缺乏复杂的多组学数据所需的精度。相反，传统的深度学习方法提供了强大的预测，但通常仍然不透明，阻碍了有效分子标记和生物机制的识别。为了解决这个问题，我们提出了交叉组学可解释神经网络（Cross-omics Interpretable Neural Network， CINN），这是一个旨在通过整合不同组学数据来预测前列腺癌状态和识别关键分子标记的仿生框架。CINN创新性地利用了来自通路或蛋白蛋白相互作用（PPI）网络的先前生物学知识，并结合了一种新的可训练的掩膜层。该掩模动态优化了预定义生物连接的强度，从而增强了知识表示和模型可解释性。该框架有效地整合了多组学数据，包括基因表达、体细胞突变和拷贝数变异，以提供疾病的整体视图。在前列腺癌数据集上的大量实验表明，CINN在强基线（P-NET）上实现了实质性的、统计上显著的性能增强。具体来说，我们表现最好的变体，带有可训练掩码的CINN-pw，将F1分数提高了13.1%至0.843，准确率提高了8.3%至0.894，AUC提高了2.3%至0.949。这些收益在统计上是显著的（p<；0.0001对于大多数关键指标），强调了我们方法的稳健性。至关重要的是，CINN固有的可解释性促进了关键候选分子的识别，包括TBP和TAF2，它们与前列腺癌的进展有关。这些发现得到了现有文献的支持，为前列腺癌的潜在机制提供了有价值的见解，为靶向治疗干预和精准医学提供了潜在的途径。

{"title":"Cross-omics interpretable neural network for discovery of molecular markers in prostate cancer","authors":"Xin Chen , Sheng Yi , Anwaier Yuemaierabola , Yuhan Liu , Liang He , Jing Ma , Wenjia Guo , Gang Sun","doi":"10.1016/j.compbiolchem.2026.108879","DOIUrl":"10.1016/j.compbiolchem.2026.108879","url":null,"abstract":"<div><div>Determining molecular markers that mediate clinically aggressive phenotypes in prostate cancer is a significant challenge. While traditional linear models offer some interpretability, they often lack the precision needed for complex multi-omics data. Conversely, conventional deep learning methods provide robust predictions but typically remain opaque, hindering the identification of impactful molecular markers and biological mechanisms. To address this, we propose the Cross-omics Interpretable Neural Network (CINN), a biomimetic framework designed to predict prostate cancer states and identify key molecular markers by integrating diverse omics data.</div><div>CINN innovatively leverages prior biological knowledge from either pathway or protein–protein interaction (PPI) networks, combined with a novel trainable mask layer. This mask dynamically optimizes the strength of pre-defined biological connections, thereby enhancing both knowledge representation and model interpretability. The framework effectively integrates multi-omics data, including gene expression, somatic mutations, and copy number variations, to provide a holistic view of the disease.</div><div>Extensive experiments on a prostate cancer dataset demonstrate that CINN achieves substantial and statistically significant performance enhancements over a strong baseline (P-NET). Specifically, our best-performing variant, CINN-pw with a trainable mask, improved F1 scores by <strong>13.1%</strong> to <strong>0.843</strong>, Accuracy by <strong>8.3%</strong> to <strong>0.894</strong>, and AUC by <strong>2.3%</strong> to <strong>0.949</strong>. These gains were consistently statistically significant (<span><math><mrow><mi>p</mi><mo><</mo><mn>0</mn><mo>.</mo><mn>0001</mn></mrow></math></span> for most key metrics), underscoring the robustness of our approach. Crucially, CINN’s inherent interpretability facilitated the identification of pivotal molecular candidates, including <strong>TBP</strong> and <strong>TAF2</strong>, which are implicated in prostate cancer progression. These findings are supported by existing literature and provide valuable insights into the underlying mechanisms of prostate cancer, offering potential avenues for targeted therapeutic interventions and precision medicine.</div></div>","PeriodicalId":10616,"journal":{"name":"Computational Biology and Chemistry","volume":"122 ","pages":"Article 108879"},"PeriodicalIF":3.1,"publicationDate":"2026-06-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145941033","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"生物学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Quality evaluation of Epmedii Folium from different species based on spectrum-efficacy relationship 基于光谱-功效关系的黄颡鱼叶质量评价。

IF 3.1 4区生物学 Q2 BIOLOGY

Computational Biology and Chemistry

Pub Date : 2026-06-01 Epub Date: 2026-01-09 DOI: 10.1016/j.compbiolchem.2026.108892

Zhichen Cai , Jia Xue , Yongyi Zhou , Haijie Chen , Jingjing Shi , Lisi Zou , Cuihua Chen , Xunhong Liu

Epimedii Folium (EF) is a widely used traditional Chinese drug that encompasses a variety of species. It has been reported that the active constituents in EF vary, leading to the uneven quality of commercial medicinal materials. To investigate the specific differences, we developed a comprehensive evaluation method to access the quality of EF from four species. First, UPLC-Triple TOF-MS/MS was used to generate characteristic fingerprints of Epimedium samples; second, adenine-induced Kidney Yang deficiency model was established to evaluate the quality of four Epimedium varieties by evaluating biochemical markers and morphology; third, multivariate statistical analysis, including gray correlation analysis and bivariate correlation analysis was combined; and finally, UPLC-QTRAP MS/MS identified a potential biomarker. The results showed that 12 common peaks were identified in 40 batches derived from four Epimedium species. The severity of kidney and testicular lesions in experimental groups of rats showed significant improvement compared with the model. GCA and BCA indicated that three ingredients, icariin, quercitrin, and epimedin B were potential biomarkers, confirmed using LC-MS. In addition, epimedin B and icariin were significantly higher in EBM compared to the other three species, consistent with the pharmacological tests. The quality and efficacy of EF from different origins were stable, and all of them had protective effects on Kidney Yang deficiency of rats. Especially, all data suggested that EBM possesses superior quality than the other three. Overall, our work offers fundamental data for the thorough assessment and a fresh viewpoint on the quality control of EF from several species.

淫羊藿（Epimedii Folium， EF）是一种广泛应用的中药，其种类繁多。据报道，其有效成分各不相同，导致商品药材质量参差不齐。为了研究不同树种间的差异，我们建立了一种综合评价方法来评价四种树种的EF质量。首先，采用uplc -三重TOF-MS/MS方法，建立淫羊藿样品的特征指纹图谱；其次，建立腺嘌呤诱导肾阳虚模型，通过生化指标和形态评价4个淫羊藿品种的品质；三是多变量统计分析，包括灰色关联分析和双变量关联分析相结合；最后，UPLC-QTRAP MS/MS鉴定出潜在的生物标志物。结果表明，从4个淫羊藿属植物中分离得到的40批药材中鉴定出12个共有峰。各实验组大鼠肾脏和睾丸病变严重程度均较模型有明显改善。GCA和BCA表明淫羊藿苷、槲皮苷和淫羊藿苷B是潜在的生物标志物，经LC-MS证实。此外，EBM中epimedin B和淫羊藿苷含量明显高于其他三种，与药理学试验一致。不同来源的枳实质量和疗效稳定，均对肾阳虚大鼠有保护作用。特别是，所有数据都表明循证医学的质量优于其他三种。总之，我们的工作为深入评估和对不同物种EF的质量控制提供了基础数据和新的观点。

{"title":"Quality evaluation of Epmedii Folium from different species based on spectrum-efficacy relationship","authors":"Zhichen Cai , Jia Xue , Yongyi Zhou , Haijie Chen , Jingjing Shi , Lisi Zou , Cuihua Chen , Xunhong Liu","doi":"10.1016/j.compbiolchem.2026.108892","DOIUrl":"10.1016/j.compbiolchem.2026.108892","url":null,"abstract":"<div><div>Epimedii Folium (EF) is a widely used traditional Chinese drug that encompasses a variety of species. It has been reported that the active constituents in EF vary, leading to the uneven quality of commercial medicinal materials. To investigate the specific differences, we developed a comprehensive evaluation method to access the quality of EF from four species. First, UPLC-Triple TOF-MS/MS was used to generate characteristic fingerprints of Epimedium samples; second, adenine-induced Kidney Yang deficiency model was established to evaluate the quality of four Epimedium varieties by evaluating biochemical markers and morphology; third, multivariate statistical analysis, including gray correlation analysis and bivariate correlation analysis was combined; and finally, UPLC-QTRAP MS/MS identified a potential biomarker. The results showed that 12 common peaks were identified in 40 batches derived from four Epimedium species. The severity of kidney and testicular lesions in experimental groups of rats showed significant improvement compared with the model. GCA and BCA indicated that three ingredients, icariin, quercitrin, and epimedin B were potential biomarkers, confirmed using LC-MS. In addition, epimedin B and icariin were significantly higher in EBM compared to the other three species, consistent with the pharmacological tests. The quality and efficacy of EF from different origins were stable, and all of them had protective effects on Kidney Yang deficiency of rats. Especially, all data suggested that EBM possesses superior quality than the other three. Overall, our work offers fundamental data for the thorough assessment and a fresh viewpoint on the quality control of EF from several species.</div></div>","PeriodicalId":10616,"journal":{"name":"Computational Biology and Chemistry","volume":"122 ","pages":"Article 108892"},"PeriodicalIF":3.1,"publicationDate":"2026-06-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145968016","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"生物学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

From virtual screening to bench: A dual-validation framework for drug repurposing against PI3K 从虚拟筛选到实验：针对PI3K药物再利用的双重验证框架。

IF 3.1 4区生物学 Q2 BIOLOGY

Computational Biology and Chemistry

Pub Date : 2026-06-01 Epub Date: 2026-01-29 DOI: 10.1016/j.compbiolchem.2026.108934

Kavita Tewani , Zunnun Narmawala , Deepshikha Rathore , Heena Dave

Virtual screening has emerged as one of the most impactful in silico approaches for the identification of novel drug candidates, substantially reducing the cost and time associated with high-throughput screening (HTS). Ongoing efforts focus on exploring large-scale libraries of drug-like molecules to identify candidates with favourable pharmacological properties. In this study, we propose an applicability domain-based virtual screening strategy that extends beyond conventional approaches by prioritising compounds with ADMET profiles comparable to marketed drugs. To further enhance predictive performance, we developed a QSAR model on PI3K ligands using Light Gradient Boosting Machine (LGBM), which achieved an R2 value of 0.799, thereby providing an additional layer of validation for compound selection. The phosphoinositide 3-kinase (PI3K) pathway, a critical regulator of cell growth, survival, metabolism, and proliferation, is frequently dysregulated in multiple cancers and other diseases. Repurposing existing drugs that modulate PI3K activity offers the potential to accelerate therapeutic development while mitigating the challenges of de novo drug discovery.

To demonstrate the utility of our approach, we screened two compound libraries from Enamine—a hit-like locator library (

>

400,000 molecules) and a kinase-focused library (

>

64,000 molecules)—against the PI3K-

α

isoform. In addition, a set of 1367 FDA-approved drugs was screened to identify potential candidates for repurposing. From these extensive datasets, three small molecules from the Enamine libraries were identified with favourable drug-like properties and synthetic accessibility compared with existing PI3K-

α

inhibitors. Furthermore, one FDA-approved drug demonstrated potential PI3K-

α

inhibitory activity. Pharmacophore mapping provided additional validation of their drug-likeness. Importantly, wet-lab evaluation of the FDA-approved drug confirmed its inhibitory activity, thereby supporting the computational predictions.

Overall, our integrated in silico and experimental framework highlights promising PI3K-

α

inhibitors, underscoring the potential of applicability domain–based virtual screening and QSAR modelling for both drug discovery and repurposing.

虚拟筛选已成为识别新型候选药物的最具影响力的计算机方法之一，大大降低了与高通量筛选（HTS）相关的成本和时间。目前的工作重点是探索大规模的药物样分子文库，以确定具有良好药理特性的候选药物。在这项研究中，我们提出了一种基于适用性域的虚拟筛选策略，该策略超越了传统方法，优先考虑ADMET谱与上市药物相当的化合物。为了进一步提高预测性能，我们使用光梯度增强机（LGBM）建立了PI3K配体的QSAR模型，其R2值为0.799，从而为化合物选择提供了额外的验证层。磷酸肌肽3-激酶（PI3K）通路是细胞生长、存活、代谢和增殖的关键调节因子，在多种癌症和其他疾病中经常失调。重新利用现有的调节PI3K活性的药物提供了加速治疗开发的潜力，同时减轻了新药物发现的挑战。为了证明我们的方法的实用性，我们从enamine中筛选了两个化合物文库——一个类似于hit的定位文库（>400,000个分子）和一个激酶聚焦文库（>64,000个分子）——针对PI3K-α亚型。此外，一组1367 fda批准的药物进行筛选，以确定潜在的候选药物重新利用。从这些广泛的数据集中，从Enamine文库中鉴定出三个小分子，与现有的PI3K-α抑制剂相比，它们具有良好的药物样性质和合成可及性。此外，一种fda批准的药物显示出潜在的PI3K-α抑制活性。药效团图谱进一步验证了它们的药物相似性。重要的是，fda批准的药物的湿实验室评估证实了其抑制活性，从而支持了计算预测。总体而言，我们的集成硅和实验框架突出了有前途的PI3K-α抑制剂，强调了基于域的虚拟筛选和QSAR建模在药物发现和再利用方面的适用性潜力。

{"title":"From virtual screening to bench: A dual-validation framework for drug repurposing against PI3K","authors":"Kavita Tewani , Zunnun Narmawala , Deepshikha Rathore , Heena Dave","doi":"10.1016/j.compbiolchem.2026.108934","DOIUrl":"10.1016/j.compbiolchem.2026.108934","url":null,"abstract":"<div><div>Virtual screening has emerged as one of the most impactful in silico approaches for the identification of novel drug candidates, substantially reducing the cost and time associated with high-throughput screening (HTS). Ongoing efforts focus on exploring large-scale libraries of drug-like molecules to identify candidates with favourable pharmacological properties. In this study, we propose an applicability domain-based virtual screening strategy that extends beyond conventional approaches by prioritising compounds with ADMET profiles comparable to marketed drugs. To further enhance predictive performance, we developed a QSAR model on PI3K ligands using Light Gradient Boosting Machine (LGBM), which achieved an R2 value of 0.799, thereby providing an additional layer of validation for compound selection. The phosphoinositide 3-kinase (PI3K) pathway, a critical regulator of cell growth, survival, metabolism, and proliferation, is frequently dysregulated in multiple cancers and other diseases. Repurposing existing drugs that modulate PI3K activity offers the potential to accelerate therapeutic development while mitigating the challenges of de novo drug discovery.</div><div>To demonstrate the utility of our approach, we screened two compound libraries from Enamine—a hit-like locator library (<span><math><mo>></mo></math></span>400,000 molecules) and a kinase-focused library (<span><math><mo>></mo></math></span>64,000 molecules)—against the PI3K-<span><math><mi>α</mi></math></span> isoform. In addition, a set of 1367 FDA-approved drugs was screened to identify potential candidates for repurposing. From these extensive datasets, three small molecules from the Enamine libraries were identified with favourable drug-like properties and synthetic accessibility compared with existing PI3K-<span><math><mi>α</mi></math></span> inhibitors. Furthermore, one FDA-approved drug demonstrated potential PI3K-<span><math><mi>α</mi></math></span> inhibitory activity. Pharmacophore mapping provided additional validation of their drug-likeness. Importantly, wet-lab evaluation of the FDA-approved drug confirmed its inhibitory activity, thereby supporting the computational predictions.</div><div>Overall, our integrated in silico and experimental framework highlights promising PI3K-<span><math><mi>α</mi></math></span> inhibitors, underscoring the potential of applicability domain–based virtual screening and QSAR modelling for both drug discovery and repurposing.</div></div>","PeriodicalId":10616,"journal":{"name":"Computational Biology and Chemistry","volume":"122 ","pages":"Article 108934"},"PeriodicalIF":3.1,"publicationDate":"2026-06-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"146115046","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"生物学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Repurposing sulfonamide drugs as anticancer ligands and understanding its properties through density functional theory 利用密度泛函理论研究磺胺类药物抗癌配体的特性。

IF 3.1 4区生物学 Q2 BIOLOGY

Computational Biology and Chemistry

Pub Date : 2026-06-01 Epub Date: 2026-02-01 DOI: 10.1016/j.compbiolchem.2026.108933

Palanisamy Deepa , Balasubramanian Sundarakannan , Duraisamy Thirumeignanam

Drug repurposing represents a promising approach towards drug discovery that has the potential to improve patient outcomes and address unmet medical needs. This study attempted to repurpose existing sulfonamide drugs in search of novel anticancer drugs because of their effectiveness in treating bacterial infections. A search was made in DrugBank for Sulfonamide, and 25 drugs with functional groups like SH, OSO, CS, and -S- were chosen for our study. The drug properties, such as dipole moment, volume, polarisability, highest occupied molecular orbital (HOMO), lowest unoccupied molecular orbital (LUMO), and electrostatic potential map, were analysed through a quantum mechanical approach at different functionals: M062X, M06HF, and B3LYP with basis sets (6–31 +G*, LANL2DZ). The electrostatic potential map was analyzed to determine the magnitude, size, and distribution of the electron cloud surrounding the sulfur atoms. Analysis of NBO (Natural Bond Orbital) and NCI (Non-Covalent Interaction) plots confirmed the presence of intramolecular hydrogen bonding in the sulfonamide drugs. Furthermore, the frontier molecular orbitals (HOMO and LUMO) and the band gap were thoroughly examined for all drugs to identify the best electron acceptors and donors. Docking analysis was performed to have a lock-and-key model of 25 sulfonamide drugs with the most promising cancer-targeted protein (1ZZ1): histone deacetylases (HDACs). The best drug orientation (optimal position) was discussed and compared with the control ligand SHH based on the analysis of binding affinity and root mean square deviation (RMSD). Binding affinity of control ligand SHH is −8.1 kcal/mol for the 2nd pose, which matches exactly with 1ZZ1 SHH ligand. The drugs Tolazamide, Fezolinetant, Ensulizole, Taurolidine, Acetohexamide, Isoxicam, Sulfamethizole, Sulfamethoxazole, Sulfapyridine, Sulfaphenazole, and Dodecyl sulphate were observed to exhibit high molecular volume, polarizability, dipole moment and significant HOMO, LUMO values, which are recommended for further quantum mechanical calculations. The findings of this study will be essential for evaluating the properties of sulfonamide drugs from a drugbank using a variety of analyses in order to repurpose them as novel anticancer drugs. Quantum mechanical calculations will be performed on the optimal docking poses in future work. Keywords: Sulfonamide drugs, Docking, Histone deacetylases, Lipinsk’s rule, Binding affinity

药物再利用是一种很有前途的药物发现方法，有可能改善患者的治疗效果并解决未满足的医疗需求。由于磺胺类药物在治疗细菌感染方面的有效性，本研究试图重新利用现有的磺胺类药物来寻找新的抗癌药物。我们在DrugBank中检索了磺胺类药物，选取了含有SH、OSO、CS、- s -等功能基团的25种药物作为研究对象。利用量子力学方法分析了M062X、M06HF和B3LYP不同官能团（6-31 +G*, LANL2DZ）上的偶极矩、体积、极化率、最高占据分子轨道（HOMO）、最低未占据分子轨道（LUMO）和静电势图等药物性质。分析静电势图以确定硫原子周围电子云的大小、大小和分布。NBO（天然键轨道）和NCI（非共价相互作用）图的分析证实了磺胺类药物分子内氢键的存在。此外，对所有药物的前沿分子轨道（HOMO和LUMO）和带隙进行了彻底的检查，以确定最佳的电子受体和给体。对接分析25种磺胺类药物与最有希望的癌症靶向蛋白（1ZZ1）：组蛋白去乙酰化酶（hdac）建立锁-钥匙模型。通过结合亲和力和均方根偏差（RMSD）分析，讨论了最佳药物取向（最佳位置），并与对照配体SHH进行了比较。控制配体SHH第二位姿的结合亲和力为-8.1 kcal/mol，与1ZZ1 SHH配体完全匹配。药物Tolazamide、Fezolinetant、ensullizole、taaurolidine、Acetohexamide、Isoxicam、sulfameethizole、Sulfamethoxazole、Sulfapyridine、Sulfaphenazole和Dodecyl sulphate表现出较高的分子体积、极化率、偶极矩和显著的HOMO、LUMO值，建议进一步进行量子力学计算。本研究的发现对于利用各种分析方法评估药库中磺胺类药物的特性，以便将其重新用作新型抗癌药物至关重要。在未来的工作中，将对最佳对接姿态进行量子力学计算。关键词：磺胺类药物，对接，组蛋白去乙酰化酶，利平斯克规则，结合亲和力

{"title":"Repurposing sulfonamide drugs as anticancer ligands and understanding its properties through density functional theory","authors":"Palanisamy Deepa , Balasubramanian Sundarakannan , Duraisamy Thirumeignanam","doi":"10.1016/j.compbiolchem.2026.108933","DOIUrl":"10.1016/j.compbiolchem.2026.108933","url":null,"abstract":"<div><div>Drug repurposing represents a promising approach towards drug discovery that has the potential to improve patient outcomes and address unmet medical needs. This study attempted to repurpose existing sulfonamide drugs in search of novel anticancer drugs because of their effectiveness in treating bacterial infections. A search was made in DrugBank for Sulfonamide, and 25 drugs with functional groups like SH, O<img>S<img>O, C<img>S, and -S- were chosen for our study. The drug properties, such as dipole moment, volume, polarisability, highest occupied molecular orbital (HOMO), lowest unoccupied molecular orbital (LUMO), and electrostatic potential map, were analysed through a quantum mechanical approach at different functionals: M062X, M06HF, and B3LYP with basis sets (6–31 +G*, LANL2DZ). The electrostatic potential map was analyzed to determine the magnitude, size, and distribution of the electron cloud surrounding the sulfur atoms. Analysis of NBO (Natural Bond Orbital) and NCI (Non-Covalent Interaction) plots confirmed the presence of intramolecular hydrogen bonding in the sulfonamide drugs. Furthermore, the frontier molecular orbitals (HOMO and LUMO) and the band gap were thoroughly examined for all drugs to identify the best electron acceptors and donors. Docking analysis was performed to have a lock-and-key model of 25 sulfonamide drugs with the most promising cancer-targeted protein (1ZZ1): histone deacetylases (HDACs). The best drug orientation (optimal position) was discussed and compared with the control ligand SHH based on the analysis of binding affinity and root mean square deviation (RMSD). Binding affinity of control ligand SHH is −8.1 kcal/mol for the 2nd pose, which matches exactly with 1ZZ1 SHH ligand. The drugs Tolazamide, Fezolinetant, Ensulizole, Taurolidine, Acetohexamide, Isoxicam, Sulfamethizole, Sulfamethoxazole, Sulfapyridine, Sulfaphenazole, and Dodecyl sulphate were observed to exhibit high molecular volume, polarizability, dipole moment and significant HOMO, LUMO values, which are recommended for further quantum mechanical calculations. The findings of this study will be essential for evaluating the properties of sulfonamide drugs from a drugbank using a variety of analyses in order to repurpose them as novel anticancer drugs. Quantum mechanical calculations will be performed on the optimal docking poses in future work. <strong>Keywords:</strong> Sulfonamide drugs, Docking, Histone deacetylases, Lipinsk’s rule, Binding affinity</div></div>","PeriodicalId":10616,"journal":{"name":"Computational Biology and Chemistry","volume":"122 ","pages":"Article 108933"},"PeriodicalIF":3.1,"publicationDate":"2026-06-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"146121383","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"生物学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

UriPred: Machine learning prediction of urinary proteins and identification of biomarkers for liver cancer UriPred：尿蛋白的机器学习预测和肝癌生物标志物的鉴定。

IF 3.1 4区生物学 Q2 BIOLOGY

Computational Biology and Chemistry

Pub Date : 2026-06-01 Epub Date: 2026-02-08 DOI: 10.1016/j.compbiolchem.2026.108946

Dahrii Paul, Vigneshwar Suriya Prakash Sinnarasan, Rajesh Das, Md Mujibur Rahman Sheikh, Santhosh Manickannan, Amouda Venkatesan

Urinary proteins are promising non-invasive biomarkers, but their low abundance and wide dynamic range make detection challenging. This study presents UriPred, a computational tool that integrates machine learning (ML), BLAST, and Motif-EmeRging and Classes-Identification (MERCI) to predict urinary proteins and facilitate the identification of liver cancer (LC) biomarkers. A dataset of 10588 urinary and non-urinary proteins was curated, from which two feature types were generated: 10074 compositional and 20 evolutionary features. Seven feature selection methods were applied to compositional features, and 11 ML algorithms were trained on different feature sets. Evolutionary features achieved the highest training performance (AUC 0.79, accuracy 71.99 %), whereas amino acid composition (AAC) with 20 features achieved identical validation AUC (0.74) and comparable accuracy while being computationally less expensive and consistently selected. The ML-AAC model was therefore chosen as the final model. This optimal model was integrated with BLAST and MERCI to create UriPred, which reduced false positives from 34.59 % (ML) to 3.12 % (hybrid) on the validation dataset and from 5.8 % (ML) to zero (hybrid) on an external dataset. Using UriPred, 53 LC differentially expressed protein-coding genes were predicted as urinary proteins. Protein-protein interaction analysis, AUROC evaluation (AUC > 0.80), survival analysis, and cross-verification of urine detectability with the Human Protein Atlas and Human Urine PeptideAtlas databases identified five proteins (KIF23, COL15A1, CTHRC1, MMP9, and SPP1) as potential LC biomarkers. UriPred efficiently predicts urinary proteins using AAC features and enables biomarker discovery for LC. The tool is publicly available at https://github.com/Dahrii-Paul/UriPred.

尿蛋白是一种很有前途的非侵入性生物标志物，但其低丰度和宽动态范围给检测带来了挑战。本研究提出了UriPred，一种集成了机器学习（ML）， BLAST和Motif-EmeRging and Classes-Identification （MERCI）的计算工具，用于预测尿蛋白并促进肝癌（LC）生物标志物的鉴定。收集了10588个尿蛋白和非尿蛋白的数据集，从中生成了两种特征类型：10074个组成特征和20个进化特征。将7种特征选择方法应用于组合特征，并在不同的特征集上训练了11种 ML算法。进化特征获得了最高的训练性能（AUC 0.79，准确率71.99 %），而氨基酸组成（AAC）与20个特征获得相同的验证AUC（0.74）和相当的准确性，同时计算成本更低，选择一致。因此选择ML-AAC模型作为最终模型。该优化模型与BLAST和MERCI集成创建了UriPred，在验证数据集上将假阳性从34.59 % （ML）减少到3.12 %（混合），在外部数据集上将假阳性从5.8 % （ML）减少到零（混合）。利用UriPred预测了53个LC差异表达蛋白编码基因作为尿蛋白。蛋白-蛋白相互作用分析、AUROC评估（AUC > 0.80）、生存分析以及与Human Protein Atlas和Human urine PeptideAtlas数据库交叉验证尿液可检出性，确定了5种蛋白（KIF23、COL15A1、CTHRC1、MMP9和SPP1）作为潜在的LC生物标志物。UriPred利用AAC特征有效地预测尿蛋白，并使LC的生物标志物发现成为可能。该工具可在https://github.com/Dahrii-Paul/UriPred上公开获取。

{"title":"UriPred: Machine learning prediction of urinary proteins and identification of biomarkers for liver cancer","authors":"Dahrii Paul, Vigneshwar Suriya Prakash Sinnarasan, Rajesh Das, Md Mujibur Rahman Sheikh, Santhosh Manickannan, Amouda Venkatesan","doi":"10.1016/j.compbiolchem.2026.108946","DOIUrl":"10.1016/j.compbiolchem.2026.108946","url":null,"abstract":"<div><div>Urinary proteins are promising non-invasive biomarkers, but their low abundance and wide dynamic range make detection challenging. This study presents UriPred, a computational tool that integrates machine learning (ML), BLAST, and Motif-EmeRging and Classes-Identification (MERCI) to predict urinary proteins and facilitate the identification of liver cancer (LC) biomarkers. A dataset of 10588 urinary and non-urinary proteins was curated, from which two feature types were generated: 10074 compositional and 20 evolutionary features. Seven feature selection methods were applied to compositional features, and 11 ML algorithms were trained on different feature sets. Evolutionary features achieved the highest training performance (AUC 0.79, accuracy 71.99 %), whereas amino acid composition (AAC) with 20 features achieved identical validation AUC (0.74) and comparable accuracy while being computationally less expensive and consistently selected. The ML-AAC model was therefore chosen as the final model. This optimal model was integrated with BLAST and MERCI to create UriPred, which reduced false positives from 34.59 % (ML) to 3.12 % (hybrid) on the validation dataset and from 5.8 % (ML) to zero (hybrid) on an external dataset. Using UriPred, 53 LC differentially expressed protein-coding genes were predicted as urinary proteins. Protein-protein interaction analysis, AUROC evaluation (AUC > 0.80), survival analysis, and cross-verification of urine detectability with the Human Protein Atlas and Human Urine PeptideAtlas databases identified five proteins (KIF23, COL15A1, CTHRC1, MMP9, and SPP1) as potential LC biomarkers. UriPred efficiently predicts urinary proteins using AAC features and enables biomarker discovery for LC. The tool is publicly available at <span><span>https://github.com/Dahrii-Paul/UriPred</span><svg><path></path></svg></span>.</div></div>","PeriodicalId":10616,"journal":{"name":"Computational Biology and Chemistry","volume":"122 ","pages":"Article 108946"},"PeriodicalIF":3.1,"publicationDate":"2026-06-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"146183832","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"生物学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Targeting the MPO/LCN2/GMPPB axis in IBS-depression comorbidity: Integrated multi-omics and bidirectional network pharmacology for precision diagnostics and therapeutics 靶向MPO/LCN2/GMPPB轴在IBS-depression共病中的作用：用于精确诊断和治疗的多组学和双向网络药理学整合

IF 3.1 4区生物学 Q2 BIOLOGY

Computational Biology and Chemistry

Pub Date : 2026-06-01 Epub Date: 2026-02-08 DOI: 10.1016/j.compbiolchem.2026.108939

Shirui Li , Feng Jiang , Xiuyang Li

Background

Irritable Bowel Syndrome (IBS) and Major Depressive Disorder (MDD) exhibit high comorbidity, driven by dysregulation of gut-brain axis interactions. Despite evidence of shared pathophysiology, the core molecular mechanisms and therapeutic targets remain elusive, largely due to clinical heterogeneity and fragmented research approaches.

Methods

We established an integrated framework combining: (1) Bidirectional epidemiological analysis using the CHARLS cohort; (2) Multi-tissue transcriptomics (intestinal mucosa/prefrontal cortex) from GEO datasets using differential expression analysis, WGCNA, and machine learning (LASSO/RF/SVM-RFE); (3) PPI network reconstruction followed by multi-algorithm topological validation; (4) Functional enrichment and immune deconvolution (CIBERSORTx); (5) Bidirectional pharmacology (CTD-based compounds screening and TCM network pharmacology); (6) Molecular docking and short-term molecular dynamics (MD) simulations for binding stability assessment; (7) ADME/Tox Profiling.

Results

Epidemiological analysis indicated bidirectional IBS-MDD risk (Digestive to Mental: OR=1.82(95%CI:1.65-6.79), Mental to Digestive: OR=3.34(95%CI:1.17-2.82)). Integrated transcriptomics identified MPO, LCN2, and GMPPB as core comorbidity genes, validated across cohorts and linked to neutrophil activation, iron dysregulation, and glycosylation defects. Immune profiling revealed tissue-specific dysregulation, with gut-dominated neutrophil/M2 macrophage infiltration in IBS versus brain-enriched CD8⁺ T/NK cells in MDD. Bidirectional pharmacology prioritized bisphenol A/lipopolysaccharide (pathogenic) and resveratrol/quercetin (therapeutic) as high-affinity binders to core targets (ΔG < –7.0 kcal/mol). Short-term MD simulations provided preliminary support for the binding of key therapeutic compounds to targets GMPPB and MPO, supported by TCM herbs (e.g., Jujubae Fructus).

Conclusion

Our study analyzes neuro-immune-endocrine crosstalk underlying IBS-MDD comorbidity, nominating MPO/LCN2/GMPPB as diagnostic biomarkers and therapeutic targets. Environmental toxins and natural compounds offer actionable strategies for gut-brain axis modulation.

背景：肠易激综合征（IBS）和重度抑郁症（MDD）表现出高的合并症，由肠-脑轴相互作用失调驱动。尽管有共同的病理生理学证据，但主要由于临床异质性和零散的研究方法，核心分子机制和治疗靶点仍然难以捉摸。方法：(1)采用CHARLS队列进行双向流行病学分析；(2)利用差异表达分析、WGCNA和机器学习（LASSO/RF/SVM-RFE）对GEO数据集进行多组织转录组学（肠黏膜/前额叶皮质）；(3)重构PPI网络，并进行多算法拓扑验证；(4)功能富集和免疫反褶积（CIBERSORTx）；(5)双向药理学（基于ctd的化合物筛选和中医网络药理学）；(6)结合稳定性评估的分子对接和短期分子动力学（MD）模拟；(7) ADME/Tox分析。结果：流行病学分析显示双向IBS-MDD风险(消化系统与精神系统：OR=1.82（95%CI:1.65 ~ 6.79)，精神系统与消化系统：OR=3.34(95%CI:1.17 ~ 2.82)）。整合转录组学鉴定MPO、LCN2和GMPPB为核心共病基因，跨队列验证并与中性粒细胞激活、铁调节失调和糖基化缺陷相关。免疫谱显示了组织特异性失调，IBS中肠道主导的中性粒细胞/M2巨噬细胞浸润与MDD中脑富集的CD8 + T/NK细胞相比。双向药理学优先考虑双酚A/脂多糖（致病）和白藜芦醇/槲皮素（治疗）作为核心靶点的高亲和力结合物（ΔG < -7.0 kcal/mol）。短期MD模拟为关键治疗化合物与目标GMPPB和MPO的结合提供了初步支持，并得到中药（如枣果）的支持。结论：本研究分析了IBS-MDD合并症的神经-免疫-内分泌串扰，推荐MPO/LCN2/GMPPB作为诊断生物标志物和治疗靶点。环境毒素和天然化合物为肠脑轴调节提供了可行的策略。

{"title":"Targeting the MPO/LCN2/GMPPB axis in IBS-depression comorbidity: Integrated multi-omics and bidirectional network pharmacology for precision diagnostics and therapeutics","authors":"Shirui Li , Feng Jiang , Xiuyang Li","doi":"10.1016/j.compbiolchem.2026.108939","DOIUrl":"10.1016/j.compbiolchem.2026.108939","url":null,"abstract":"<div><h3>Background</h3><div>Irritable Bowel Syndrome (IBS) and Major Depressive Disorder (MDD) exhibit high comorbidity, driven by dysregulation of gut-brain axis interactions. Despite evidence of shared pathophysiology, the core molecular mechanisms and therapeutic targets remain elusive, largely due to clinical heterogeneity and fragmented research approaches.</div></div><div><h3>Methods</h3><div>We established an integrated framework combining: (1) Bidirectional epidemiological analysis using the CHARLS cohort; (2) Multi-tissue transcriptomics (intestinal mucosa/prefrontal cortex) from GEO datasets using differential expression analysis, WGCNA, and machine learning (LASSO/RF/SVM-RFE); (3) PPI network reconstruction followed by multi-algorithm topological validation; (4) Functional enrichment and immune deconvolution (CIBERSORTx); (5) Bidirectional pharmacology (CTD-based compounds screening and TCM network pharmacology); (6) Molecular docking and short-term molecular dynamics (MD) simulations for binding stability assessment; (7) ADME/Tox Profiling.</div></div><div><h3>Results</h3><div>Epidemiological analysis indicated bidirectional IBS-MDD risk (Digestive to Mental: OR=1.82(95%CI:1.65-6.79), Mental to Digestive: OR=3.34(95%CI:1.17-2.82)). Integrated transcriptomics identified MPO, LCN2, and GMPPB as core comorbidity genes, validated across cohorts and linked to neutrophil activation, iron dysregulation, and glycosylation defects. Immune profiling revealed tissue-specific dysregulation, with gut-dominated neutrophil/M2 macrophage infiltration in IBS versus brain-enriched CD8⁺ T/NK cells in MDD. Bidirectional pharmacology prioritized bisphenol A/lipopolysaccharide (pathogenic) and resveratrol/quercetin (therapeutic) as high-affinity binders to core targets (ΔG < –7.0 kcal/mol). Short-term MD simulations provided preliminary support for the binding of key therapeutic compounds to targets GMPPB and MPO, supported by TCM herbs (e.g., Jujubae Fructus).</div></div><div><h3>Conclusion</h3><div>Our study analyzes neuro-immune-endocrine crosstalk underlying IBS-MDD comorbidity, nominating MPO/LCN2/GMPPB as diagnostic biomarkers and therapeutic targets. Environmental toxins and natural compounds offer actionable strategies for gut-brain axis modulation.</div></div>","PeriodicalId":10616,"journal":{"name":"Computational Biology and Chemistry","volume":"122 ","pages":"Article 108939"},"PeriodicalIF":3.1,"publicationDate":"2026-06-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"146183922","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"生物学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Optimal path reconstruction of plant chromosome evolution 植物染色体进化的最优路径重建

IF 3.1 4区生物学 Q2 BIOLOGY

Computational Biology and Chemistry

Pub Date : 2026-06-01 Epub Date: 2026-01-15 DOI: 10.1016/j.compbiolchem.2026.108911

Yunfei Li , weiwei Liu , Yiting Tian , Ying Wang , Yuelong Jia , Xiyin Wang

DNA rearrangements contribute to the formation of new chromosomes, which are often the foundation of speciation. Deciphering the order of DNA rearrangements facilitates the reconstruction of evolutionary trajectories from extant to ancestral chromosomes, a task that is computationally NP-hard. Here, a mathematically rigorous chromosome-rearrangement model is integrated with the “Chromosomal Inversion Path Exploration via Monte-Carlo Tree Search (CIPE-MCTS)” framework in a deeply coupled manner. This integration yields a robust analytical framework that ensures both global optimality and computational efficiency, thereby enabling the precise reconstruction of complex chromosomal evolutionary trajectories. The framework rigorously characterizes elementary rearrangement operations—inversion, translocation, fusion, and fission—within a strict graph-theoretic and group-theoretic formalism. On this basis, MCTS is introduced, guided by a domain-specific heuristic evaluation function that integrates a decay function and a dynamic hash table. The Upper Confidence Bound for Trees (UCT) serves as the node-selection criterion, while an extensible sampling strategy enables rapid convergence toward near-optimal solutions within the vast state space. Comprehensive benchmark tests demonstrate that, under identical hardware constraints, this method achieves significantly higher reconstruction accuracy than exhaustive global search, while its overall running time is markedly shorter than that of a single heuristic algorithm, thereby achieving simultaneous improvements in both accuracy and efficiency. This study introduces, for the first time, a scalable, reproducible, and mathematically guaranteed tool for accurate analysis of complex plant genomes, offering a novel quantitative perspective for elucidating the chromosomal basis of plant diversification and adaptive evolution.

DNA重排有助于新染色体的形成，这通常是物种形成的基础。破译DNA重排的顺序有助于重建从现存到祖先染色体的进化轨迹，这是一项计算np困难的任务。在这里，一个数学上严谨的染色体重排模型以一种深度耦合的方式与“通过蒙特卡洛树搜索进行染色体倒置路径探索（CIPE-MCTS）”框架相结合。这种整合产生了一个强大的分析框架，确保了全局最优性和计算效率，从而能够精确地重建复杂的染色体进化轨迹。该框架严格地描述了基本重排操作——倒置、易位、融合和裂变——在严格的图论和群论形式体系内。在此基础上，引入了MCTS，并以集成衰减函数和动态哈希表的特定领域启发式评估函数为指导。树的上置信度界（UCT）作为节点选择准则，而可扩展的采样策略可以在巨大的状态空间内快速收敛到接近最优解。综合基准测试表明，在相同硬件约束条件下，该方法重构精度明显高于穷举全局搜索，总体运行时间明显短于单一启发式算法，实现了精度和效率的同步提升。该研究首次为复杂植物基因组的精确分析提供了一种可扩展、可重复、数学上有保证的工具，为阐明植物多样化和适应性进化的染色体基础提供了新的定量视角。

{"title":"Optimal path reconstruction of plant chromosome evolution","authors":"Yunfei Li , weiwei Liu , Yiting Tian , Ying Wang , Yuelong Jia , Xiyin Wang","doi":"10.1016/j.compbiolchem.2026.108911","DOIUrl":"10.1016/j.compbiolchem.2026.108911","url":null,"abstract":"<div><div>DNA rearrangements contribute to the formation of new chromosomes, which are often the foundation of speciation. Deciphering the order of DNA rearrangements facilitates the reconstruction of evolutionary trajectories from extant to ancestral chromosomes, a task that is computationally NP-hard. Here, a mathematically rigorous chromosome-rearrangement model is integrated with the “Chromosomal Inversion Path Exploration via Monte-Carlo Tree Search (CIPE-MCTS)” framework in a deeply coupled manner. This integration yields a robust analytical framework that ensures both global optimality and computational efficiency, thereby enabling the precise reconstruction of complex chromosomal evolutionary trajectories. The framework rigorously characterizes elementary rearrangement operations—inversion, translocation, fusion, and fission—within a strict graph-theoretic and group-theoretic formalism. On this basis, MCTS is introduced, guided by a domain-specific heuristic evaluation function that integrates a decay function and a dynamic hash table. The Upper Confidence Bound for Trees (UCT) serves as the node-selection criterion, while an extensible sampling strategy enables rapid convergence toward near-optimal solutions within the vast state space. Comprehensive benchmark tests demonstrate that, under identical hardware constraints, this method achieves significantly higher reconstruction accuracy than exhaustive global search, while its overall running time is markedly shorter than that of a single heuristic algorithm, thereby achieving simultaneous improvements in both accuracy and efficiency. This study introduces, for the first time, a scalable, reproducible, and mathematically guaranteed tool for accurate analysis of complex plant genomes, offering a novel quantitative perspective for elucidating the chromosomal basis of plant diversification and adaptive evolution.</div></div>","PeriodicalId":10616,"journal":{"name":"Computational Biology and Chemistry","volume":"122 ","pages":"Article 108911"},"PeriodicalIF":3.1,"publicationDate":"2026-06-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145974567","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"生物学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Structural basis of long non-coding RNA PVT1 interactions with select mRNAs universal in pan-cancer system: A computational study 长链非编码RNA PVT1与泛癌症系统中普遍存在的mrna相互作用的结构基础：一项计算研究

IF 3.1 4区生物学 Q2 BIOLOGY

Computational Biology and Chemistry

Pub Date : 2026-06-01 Epub Date: 2026-01-08 DOI: 10.1016/j.compbiolchem.2026.108878

Euphinia Tiberius Kharsyiemiong, Bhupal Haribhakta Raghunandan, Seema Mishra

PVT1 lncRNA can regulate multi-gene expression through diverse mechanisms, one of which is through binding interactions with mRNAs. Our previous study highlighted its regulatory role in pan-cancer systems through predicted interactions with select mRNAs that are significantly differentially expressed across 15 cancer types. The structural basis of these interactions are yet to be propounded. Here, in order to identify and compare secondary structural features that may mediate PVT1 binding to select cancer-relevant mRNAs, and to determine evidence of evolutionary conservation, if any, we adopted the secondary structure folding information to identify key intermolecular interactions mediating the binding of PVT1 to specific regions, 5′UTR, coding region, and 3′UTR, of these select mRNAs, which may influence the translation process. Forming stable secondary structures, using both Watson-Crick and non-Watson-Crick base pairs, the flexibility of PVT1 lncRNA in interacting with these varied molecules at specific locations is deduced at the secondary structure level. To demonstrate the possible presence of conserved structural elements in PVT1 secondary structure generated based on 7SL non-coding RNA seed sequences, covariation analysis identified 10 significantly co-varying base pairs, suggesting structural conservation. The location of the start point of these lncRNA-mRNA interactions is majorly in the open loop regions. A-U nucleotides in the loops are observed to be higher in number than G-C nucleotides in PVT1 secondary structure. This may initiate multiple base-pairing interactions with other macromolecules more readily, owing to a lesser strength of the hydrogen bonding interactions between A-U base pairs. In the case of these mRNAs, comparatively speaking, there is a variability in the number of purines in the loop regions in their respective secondary structures. Since GC content correlates with the stability of mRNA secondary structures, our analysis shows that even though there is a variable sequence length, some of these mRNAs may demonstrate a higher stability of their specific secondary structures based on a higher GC content. Further, in order to potentially correlate with high protein expression, the distal segment of CDS and the 3′UTR regions of mRNAs require the presence of increased secondary structure. In our analysis, we found the same underlying pattern in a few of our select mRNA molecules. Exploration of the sequence and structural details of these lncRNA-mRNA interactions led us to an insight on a probable mechanism of a single PVT1 molecule being able to bind multiple mRNAs simultaneously or sequentially, in a spatio-temporal manner. Our research also seeks to further elucidate the contribution of bases and intermolecular interactions in the formation of these complexes.

PVT1 lncRNA可以通过多种机制调控多基因表达，其中一种机制是通过与mrna的结合相互作用。我们之前的研究通过预测与15种癌症类型中显著差异表达的精选mrna的相互作用，强调了其在泛癌症系统中的调节作用。这些相互作用的结构基础尚未被提出。在这里，为了鉴定和比较可能介导PVT1结合选择癌症相关mrna的二级结构特征，并确定进化保守的证据，如果有的话，我们采用二级结构折叠信息来鉴定介导PVT1与这些选择mrna的特定区域（5'UTR，编码区和3'UTR）结合的关键分子间相互作用，这些区域可能影响翻译过程。形成稳定的二级结构，使用沃森-克里克和非沃森-克里克碱基对，在二级结构水平上推导了PVT1 lncRNA在特定位置与这些不同分子相互作用的灵活性。为了证明基于7SL非编码RNA种子序列生成的PVT1二级结构中可能存在保守的结构元件，共变分析发现了10个显著共变的碱基对，表明结构保守。这些lncRNA-mRNA相互作用的起点位置主要在开环区。观察到环中的A-U核苷酸数量高于PVT1二级结构中的G-C核苷酸。由于a - u碱基对之间的氢键相互作用强度较小，这可能更容易引发与其他大分子的多重碱基对相互作用。在这些mrna的情况下，相对而言，在各自的二级结构中，环区嘌呤的数量存在可变性。由于GC含量与mRNA二级结构的稳定性相关，我们的分析表明，即使存在可变的序列长度，其中一些mRNA可能基于较高的GC含量而表现出更高的特定二级结构稳定性。此外，为了潜在地与高蛋白表达相关，CDS的远端片段和mrna的3'UTR区域需要增加二级结构的存在。在我们的分析中，我们在一些我们选择的mRNA分子中发现了相同的潜在模式。对这些lncRNA-mRNA相互作用的序列和结构细节的探索使我们深入了解了单个PVT1分子能够以时空方式同时或顺序结合多个mrna的可能机制。我们的研究还试图进一步阐明这些复合物形成过程中碱基和分子间相互作用的作用。

{"title":"Structural basis of long non-coding RNA PVT1 interactions with select mRNAs universal in pan-cancer system: A computational study","authors":"Euphinia Tiberius Kharsyiemiong, Bhupal Haribhakta Raghunandan, Seema Mishra","doi":"10.1016/j.compbiolchem.2026.108878","DOIUrl":"10.1016/j.compbiolchem.2026.108878","url":null,"abstract":"<div><div><em>PVT1</em> lncRNA can regulate multi-gene expression through diverse mechanisms, one of which is through binding interactions with mRNAs. Our previous study highlighted its regulatory role in pan-cancer systems through predicted interactions with select mRNAs that are significantly differentially expressed across 15 cancer types. The structural basis of these interactions are yet to be propounded. Here, in order to identify and compare secondary structural features that may mediate <em>PVT1</em> binding to select cancer-relevant mRNAs, and to determine evidence of evolutionary conservation, if any, we adopted the secondary structure folding information to identify key intermolecular interactions mediating the binding of <em>PVT1</em> to specific regions, 5′UTR, coding region, and 3′UTR, of these select mRNAs, which may influence the translation process. Forming stable secondary structures, using both Watson-Crick and non-Watson-Crick base pairs, the flexibility of <em>PVT1</em> lncRNA in interacting with these varied molecules at specific locations is deduced at the secondary structure level. To demonstrate the possible presence of conserved structural elements in <em>PVT1</em> secondary structure generated based on 7SL non-coding RNA seed sequences, covariation analysis identified 10 significantly co-varying base pairs, suggesting structural conservation. The location of the start point of these lncRNA-mRNA interactions is majorly in the open loop regions. A-U nucleotides in the loops are observed to be higher in number than G-C nucleotides in <em>PVT1</em> secondary structure. This may initiate multiple base-pairing interactions with other macromolecules more readily, owing to a lesser strength of the hydrogen bonding interactions between A-U base pairs. In the case of these mRNAs, comparatively speaking, there is a variability in the number of purines in the loop regions in their respective secondary structures. Since GC content correlates with the stability of mRNA secondary structures, our analysis shows that even though there is a variable sequence length, some of these mRNAs may demonstrate a higher stability of their specific secondary structures based on a higher GC content. Further, in order to potentially correlate with high protein expression, the distal segment of CDS and the 3′UTR regions of mRNAs require the presence of increased secondary structure. In our analysis, we found the same underlying pattern in a few of our select mRNA molecules. Exploration of the sequence and structural details of these lncRNA-mRNA interactions led us to an insight on a probable mechanism of a single <em>PVT1</em> molecule being able to bind multiple mRNAs simultaneously or sequentially, in a spatio-temporal manner. Our research also seeks to further elucidate the contribution of bases and intermolecular interactions in the formation of these complexes.</div></div>","PeriodicalId":10616,"journal":{"name":"Computational Biology and Chemistry","volume":"122 ","pages":"Article 108878"},"PeriodicalIF":3.1,"publicationDate":"2026-06-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145974168","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"生物学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0