Computational Biology and Chemistry最新文献_第8页

Insilico analysis of the pathogenic deletion mutations (CFTR-F508del) associated with cystic fibrosis disease 与囊性纤维化病相关的致病性缺失突变（CFTR-F508del）的计算机分析

IF 3.1 4区生物学 Q2 BIOLOGY

Computational Biology and Chemistry

Pub Date : 2026-04-01 Epub Date: 2025-12-20 DOI: 10.1016/j.compbiolchem.2025.108865

Maryam Dhary Kamel

Background

Cystic fibrosis(CF) is a life-threatening autosomal recessive disorder primaily caused by mutations in the CFTR gene that disrupt chloride ion transport across epithelial membranes. The deletion of phenylalanine at position 508 (F508del) which is the most pathogenic variant frequent, leading to misfolding of protein, defective channel gating, and accelerated degradtion of the CFTR protein. Although lvacaftor is an FDA-approved CFTR potentiator that improves channel opening probability, its exact interaction pattern with the F508del - mutated CFTR remains incompletely understood at the molecular level.

Objectives

The present study aimed to elucidate the genetic and structural consequences of the F508del mutation and to characterize lvacaftor binding th the mutant CFTR using an integrative in silico workflow. By combining sequence analysis, physicochemical profiling, homolgy modeling, and molecular docking, the study sought to (i) identify conformational alterations caused by the F508del deletion, (ii) map key residues involved in lvacaftor binding, and (iii) provide a structural rationale that could guide the optimization of CFTR-targeted therapies and personalized treatment approaches for CF patients carrying the F508del mutation.

Methods

The CFTR gene and protein sequences were analyzed using NCBI, BLASTx, and ExPASy tools. Structural models of wild-type and F508del-mutant CFTR proteins were constructed via SWISS-Model, and physicochemical properties were computed. Molecular docking of Ivacaftor with the mutant CFTR was conducted using MOE (2024.06), and ligand–receptor interactions were analyzed.

Results

The F508del mutation was confirmed to cause a three-nucleotide (CTT) deletion in exon 10, resulting in the loss of one amino acid in the translated protein. Structural modeling revealed disruption in the protein’s 3D conformation. Physicochemical analysis showed minor changes in stability and hydropathy between wild-type and mutant forms. Docking simulations indicated that Ivacaftor binds favorably to the mutant CFTR (–7.45 kcal/mol), forming stabilizing π-H and H-π interactions, particularly with TRP 51, VAL 194, and ASP 195.

Conclusions

The integration of genetic, structural, and docking analyses supports the role of Ivacaftor as an effective modulator of F508del-CFTR function. The identified interaction hotspots offer a structural basis for optimizing CFTR-targeted therapeutics and support further exploration of personalized drug design in CF treatment.

背景：囊性纤维化（CF）是一种危及生命的常染色体隐性遗传病，主要由CFTR基因突变引起，该基因突变破坏了氯离子在上皮膜上的转运。最常见的致病性变异位508 （F508del）的苯丙氨酸缺失，导致蛋白质错误折叠，通道门控缺陷，CFTR蛋白降解加速。尽管lvacaftor是fda批准的CFTR增强剂，可提高通道打开概率，但其与F508del突变的CFTR的确切相互作用模式在分子水平上仍不完全清楚。目的：本研究旨在阐明F508del突变的遗传和结构后果，并利用集成的计算机工作流来表征与突变CFTR结合的lvacator。通过结合序列分析、理化谱分析、同源性建模和分子对接，本研究试图(i)确定F508del缺失引起的构象改变，（ii）绘制与lvacaftor结合相关的关键残基，以及（iii）提供结构原理，以指导cftr靶向治疗的优化和针对携带F508del突变的CF患者的个性化治疗方法。方法：采用NCBI、BLASTx、ExPASy等工具对CFTR基因及蛋白序列进行分析。通过SWISS-Model构建野生型和f508del突变型CFTR蛋白的结构模型，计算理化性质。利用MOE（2024.06）对Ivacaftor与突变体CFTR进行分子对接，分析配体-受体相互作用。结果：F508del突变被证实在第10外显子上导致3个核苷酸（CTT）缺失，导致翻译蛋白中缺失1个氨基酸。结构模型显示蛋白质的三维构象被破坏。理化分析表明，野生型和突变型在稳定性和亲水性方面有微小的变化。对接模拟表明，Ivacaftor与突变体CFTR结合良好（-7.45 kcal/mol），形成稳定的π-H和H-π相互作用，特别是与TRP 51、VAL 194和ASP 195。结论：综合遗传、结构和对接分析，支持Ivacaftor作为F508del-CFTR功能的有效调节剂的作用。确定的相互作用热点为优化cftr靶向治疗提供了结构基础，并支持进一步探索CF治疗的个性化药物设计。

{"title":"Insilico analysis of the pathogenic deletion mutations (CFTR-F508del) associated with cystic fibrosis disease","authors":"Maryam Dhary Kamel","doi":"10.1016/j.compbiolchem.2025.108865","DOIUrl":"10.1016/j.compbiolchem.2025.108865","url":null,"abstract":"<div><h3>Background</h3><div>Cystic fibrosis(CF) is a life-threatening autosomal recessive disorder primaily caused by mutations in the <em><strong>CFTR</strong></em> gene that disrupt chloride ion transport across epithelial membranes. The deletion of phenylalanine at position 508 (F508del) which is the most pathogenic variant frequent, leading to misfolding of protein, defective channel gating, and accelerated degradtion of the <em><strong>CFTR</strong></em> protein. Although lvacaftor is an FDA-approved <em><strong>CFTR</strong></em> potentiator that improves channel opening probability, its exact interaction pattern with the F508del - mutated <em><strong>CFTR</strong></em> remains incompletely understood at the molecular level.</div></div><div><h3>Objectives</h3><div>The present study aimed to elucidate the genetic and structural consequences of the F508del mutation and to characterize lvacaftor binding th the mutant <em><strong>CFTR</strong></em> using an integrative in silico workflow. By combining sequence analysis, physicochemical profiling, homolgy modeling, and molecular docking, the study sought to (i) identify conformational alterations caused by the F508del deletion, (ii) map key residues involved in lvacaftor binding, and (iii) provide a structural rationale that could guide the optimization of <em><strong>CFTR</strong></em>-targeted therapies and personalized treatment approaches for CF patients carrying the F508del mutation.</div></div><div><h3>Methods</h3><div>The <em><strong>CFTR</strong></em> gene and protein sequences were analyzed using NCBI, BLASTx, and ExPASy tools. Structural models of wild-type and F508del-mutant <em><strong>CFTR</strong></em> proteins were constructed via SWISS-Model, and physicochemical properties were computed. Molecular docking of Ivacaftor with the mutant CFTR was conducted using MOE (2024.06), and ligand–receptor interactions were analyzed.</div></div><div><h3>Results</h3><div>The F508del mutation was confirmed to cause a three-nucleotide (CTT) deletion in exon 10, resulting in the loss of one amino acid in the translated protein. Structural modeling revealed disruption in the protein’s 3D conformation. Physicochemical analysis showed minor changes in stability and hydropathy between wild-type and mutant forms. Docking simulations indicated that Ivacaftor binds favorably to the mutant CFTR (–7.45 kcal/mol), forming stabilizing π-H and H-π interactions, particularly with TRP 51, VAL 194, and ASP 195.</div></div><div><h3>Conclusions</h3><div>The integration of genetic, structural, and docking analyses supports the role of Ivacaftor as an effective modulator of F508del-<em><strong>CFTR</strong></em> function. The identified interaction hotspots offer a structural basis for optimizing <em><strong>CFTR</strong></em>-targeted therapeutics and support further exploration of personalized drug design in CF treatment.</div></div>","PeriodicalId":10616,"journal":{"name":"Computational Biology and Chemistry","volume":"121 ","pages":"Article 108865"},"PeriodicalIF":3.1,"publicationDate":"2026-04-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145835585","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

SSHF-DTI: Leveraging structural similarity and hierarchical features through a fusion network for drug-target interaction prediction SSHF-DTI：通过融合网络利用结构相似性和层次特征进行药物-靶标相互作用预测

IF 3.1 4区生物学 Q2 BIOLOGY

Computational Biology and Chemistry

Pub Date : 2026-04-01 Epub Date: 2025-12-20 DOI: 10.1016/j.compbiolchem.2025.108853

Yuxiao Zhang, Chengping Zhao

Predicting drug-target interactions (DTI) and binding affinities (DTA) is essential for drug discovery, but experimental methods remain costly and time-consuming. While deep learning approaches have improved prediction performance, many existing models rely on single-source data and lack integration of cross-domain features, limiting their generalization. This study aims to develop a more robust and generalizable model for DTI and DTA prediction.

We propose SSHF-DTI, a model that integrates structurally similar information and multi-source substructure features to effectively preserve chemically meaningful molecular fragments and capture hierarchical feature dependencies. Structural similarity is incorporated through data enrichment based on Tanimoto coefficient evaluation of Morgan fingerprint similarity. This hybrid architecture combines transformer and convolutional components and further achieves hierarchical feature fusion, thereby significantly optimizing model performance improvement.

Compared with baseline methods, SSHF-DTI achieved improvements of 0.031 and 0.147 in ROC-AUC and PR-AUC respectively on the Davis dataset. It also demonstrated strong generalization in drug-drug interaction (DDI) tasks and showed high sensitivity in distinguishing fine-grained molecular structural features affecting binding affinity.

SSHF-DTI provides a powerful, generalizable framework for DTI/DTA/DDI prediction, capable of capturing complex hierarchical feature interactions. It shows promise for supporting drug discovery and virtual screening applications.

预测药物-靶标相互作用（DTI）和结合亲和力（DTA）对于药物发现至关重要，但实验方法仍然昂贵且耗时。虽然深度学习方法提高了预测性能，但许多现有模型依赖于单源数据，缺乏跨域特征的集成，限制了它们的泛化。本研究旨在为DTI和DTA预测建立一个更稳健和可推广的模型。我们提出了SSHF-DTI模型，该模型集成了结构相似信息和多源子结构特征，以有效地保留化学上有意义的分子片段并捕获分层特征依赖关系。基于谷本系数评价的摩根指纹相似度，通过数据充实，纳入结构相似度。该混合架构结合了变压器和卷积组件，进一步实现了层次化特征融合，从而显著优化了模型的性能。与基线方法相比，SSHF-DTI在Davis数据集上的ROC-AUC和PR-AUC分别提高了0.031和0.147。它在药物-药物相互作用（DDI）任务中也表现出很强的通用性，并在区分影响结合亲和力的细粒分子结构特征方面表现出高灵敏度。SSHF-DTI为DTI/DTA/DDI预测提供了一个强大的、可推广的框架，能够捕获复杂的分层特征交互。它显示了支持药物发现和虚拟筛选应用的前景。

{"title":"SSHF-DTI: Leveraging structural similarity and hierarchical features through a fusion network for drug-target interaction prediction","authors":"Yuxiao Zhang, Chengping Zhao","doi":"10.1016/j.compbiolchem.2025.108853","DOIUrl":"10.1016/j.compbiolchem.2025.108853","url":null,"abstract":"<div><div>Predicting drug-target interactions (DTI) and binding affinities (DTA) is essential for drug discovery, but experimental methods remain costly and time-consuming. While deep learning approaches have improved prediction performance, many existing models rely on single-source data and lack integration of cross-domain features, limiting their generalization. This study aims to develop a more robust and generalizable model for DTI and DTA prediction.</div><div>We propose SSHF-DTI, a model that integrates structurally similar information and multi-source substructure features to effectively preserve chemically meaningful molecular fragments and capture hierarchical feature dependencies. Structural similarity is incorporated through data enrichment based on Tanimoto coefficient evaluation of Morgan fingerprint similarity. This hybrid architecture combines transformer and convolutional components and further achieves hierarchical feature fusion, thereby significantly optimizing model performance improvement.</div><div>Compared with baseline methods, SSHF-DTI achieved improvements of 0.031 and 0.147 in ROC-AUC and PR-AUC respectively on the Davis dataset. It also demonstrated strong generalization in drug-drug interaction (DDI) tasks and showed high sensitivity in distinguishing fine-grained molecular structural features affecting binding affinity.</div><div>SSHF-DTI provides a powerful, generalizable framework for DTI/DTA/DDI prediction, capable of capturing complex hierarchical feature interactions. It shows promise for supporting drug discovery and virtual screening applications.</div></div>","PeriodicalId":10616,"journal":{"name":"Computational Biology and Chemistry","volume":"121 ","pages":"Article 108853"},"PeriodicalIF":3.1,"publicationDate":"2026-04-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145836668","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

DeepHFFT-m7G: A dual-channel self-attention and hybrid feature fusion framework for RNA m7G modification identification DeepHFFT-m7G：一种用于RNA m7G修饰识别的双通道自关注和混合特征融合框架

IF 3.1 4区生物学 Q2 BIOLOGY

Computational Biology and Chemistry

Pub Date : 2026-04-01 Epub Date: 2025-11-26 DOI: 10.1016/j.compbiolchem.2025.108812

Yongxian Fan, Zeheng Wu, Guicong Sun

N7-Methylguanosine (m7G) is a prevalent RNA modification that has attracted significant attention in recent RNA functional research. Multiple studies have shown that m7G modifications play a crucial role in the initiation and progression of various human diseases. Although multiple deep learning methods have been employed to predict m7G modification sites, their accuracy remains suboptimal. In this work, we propose a novel method called DeepHFFT-m7G, which is based on hybrid feature fusion and a dual-channel self-attention network. This approach aims to efficiently identify m7G methylation sites in RNA sequences. First, we integrate four classical RNA sequence features with embedding vectors based on RNA2Vec. Next, a multi-branch convolutional neural network (CNN) constructs a dual-channel feature extraction module, capturing local sequence features. A Transformer encoding module is then introduced to extract global features. Finally, the sequence embedding is transferred to a multi-layer perceptron (MLP) to achieve efficient m7G site prediction. Independent testing demonstrates that DeepHFFT-m7G achieves AUROC, accuracy, MCC, and specificity scores of 97.53%, 96.92%, 93.93%, and 97.63%, respectively, significantly outperforming existing state-of-the-art (SOTA) methods. Furthermore, comparative experiments and visualization analyses further validate the superiority and robust generalization ability of DeepHFFT-m7G.

n7 -甲基鸟苷（m7G）是一种普遍存在的RNA修饰，近年来在RNA功能研究中引起了广泛关注。多项研究表明，m7G修饰在各种人类疾病的发生和发展中起着至关重要的作用。尽管多种深度学习方法已被用于预测m7G修饰位点，但其准确性仍然不是最佳的。在这项工作中，我们提出了一种名为DeepHFFT-m7G的新方法，该方法基于混合特征融合和双通道自关注网络。该方法旨在有效地识别RNA序列中的m7G甲基化位点。首先，将四种经典RNA序列特征与基于RNA2Vec的嵌入向量进行整合。其次，多分支卷积神经网络（CNN）构建双通道特征提取模块，捕获局部序列特征；然后引入Transformer编码模块来提取全局特征。最后，将序列嵌入转移到多层感知器（MLP）中，以实现高效的m7G站点预测。独立测试表明，DeepHFFT-m7G的AUROC、准确度、MCC和特异性评分分别为97.53%、96.92%、93.93%和97.63%，显著优于现有的最先进（SOTA）方法。对比实验和可视化分析进一步验证了DeepHFFT-m7G算法的优越性和鲁棒泛化能力。

{"title":"DeepHFFT-m7G: A dual-channel self-attention and hybrid feature fusion framework for RNA m7G modification identification","authors":"Yongxian Fan, Zeheng Wu, Guicong Sun","doi":"10.1016/j.compbiolchem.2025.108812","DOIUrl":"10.1016/j.compbiolchem.2025.108812","url":null,"abstract":"<div><div>N7-Methylguanosine (m7G) is a prevalent RNA modification that has attracted significant attention in recent RNA functional research. Multiple studies have shown that m7G modifications play a crucial role in the initiation and progression of various human diseases. Although multiple deep learning methods have been employed to predict m7G modification sites, their accuracy remains suboptimal. In this work, we propose a novel method called DeepHFFT-m7G, which is based on hybrid feature fusion and a dual-channel self-attention network. This approach aims to efficiently identify m7G methylation sites in RNA sequences. First, we integrate four classical RNA sequence features with embedding vectors based on RNA2Vec. Next, a multi-branch convolutional neural network (CNN) constructs a dual-channel feature extraction module, capturing local sequence features. A Transformer encoding module is then introduced to extract global features. Finally, the sequence embedding is transferred to a multi-layer perceptron (MLP) to achieve efficient m7G site prediction. Independent testing demonstrates that DeepHFFT-m7G achieves AUROC, accuracy, MCC, and specificity scores of 97.53%, 96.92%, 93.93%, and 97.63%, respectively, significantly outperforming existing state-of-the-art (SOTA) methods. Furthermore, comparative experiments and visualization analyses further validate the superiority and robust generalization ability of DeepHFFT-m7G.</div></div>","PeriodicalId":10616,"journal":{"name":"Computational Biology and Chemistry","volume":"121 ","pages":"Article 108812"},"PeriodicalIF":3.1,"publicationDate":"2026-04-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145617430","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

TEWS: Transformer-empowered weakly supervised prediction of immune score and genetic mutations in liver cancer from whole slide image TEWS：从整个幻灯片图像中对肝癌的免疫评分和基因突变进行变压器授权弱监督预测

IF 3.1 4区生物学 Q2 BIOLOGY

Computational Biology and Chemistry

Pub Date : 2026-04-01 Epub Date: 2025-12-17 DOI: 10.1016/j.compbiolchem.2025.108837

Bin Wang , Xianchuan Chen , Ying Li , Huiqing Wang , Keke Wang , Xiaoyan Li

Whole Slide Imaging (WSI) plays a crucial role in predicting immune scores by providing detailed cellular and tissue-level insights, thereby enhancing pathological diagnosis accuracy and biomarker detection. Additionally, WSI contributes essential information for personalized treatment decisions. However, the application of deep learning models to WSI is hindered by two major challenges. First, the substantial data requirements, combined with the high cost and labor-intensive nature of sample annotation, limit the availability of well-labeled datasets. Second, the computational demands of these models pose practical constraints for many healthcare institutions. To address these challenges, we propose a weakly supervised deep learning model based on the transformer architecture. Specifically, we design a novel network incorporating the Swin Transformer, which, unlike traditional convolutional networks, emphasizes global feature extraction. This improves the accuracy of pseudo-label assignment during feature embedding. Additionally, we integrate a gated attention pooling mechanism and employ multi-instance learning (MIL), enabling, for the first time, immune score prediction directly from WSIs. Our model was evaluated through 5-fold cross-validation and achieved an area under the curve (AUC) of 0.88 for immune score prediction in liver cancer. Furthermore, it demonstrated strong predictive performance for genetic mutations, achieving an AUC of 0.99 for CUB and Sushi multiple domains 1 (CSMD1) mutations and 0.86 for Tumor Protein 53 (TP53) mutations in liver cancer. These results highlight the potential of transformer-based weakly supervised learning in computational pathology.

全切片成像（WSI）通过提供详细的细胞和组织水平的见解，从而提高病理诊断的准确性和生物标志物检测，在预测免疫评分方面起着至关重要的作用。此外，WSI为个性化治疗决策提供了必要的信息。然而，深度学习模型在WSI中的应用受到两个主要挑战的阻碍。首先，大量的数据需求，加上样本标注的高成本和劳动密集型性质，限制了标记良好的数据集的可用性。其次，这些模型的计算需求对许多医疗机构构成了实际限制。为了解决这些挑战，我们提出了一个基于变压器架构的弱监督深度学习模型。具体来说，我们设计了一个包含Swin变压器的新型网络，与传统的卷积网络不同，它强调全局特征提取。这提高了特征嵌入过程中伪标签分配的准确性。此外，我们集成了一种门控注意力池机制，并采用了多实例学习（MIL），首次实现了直接从wsi中预测免疫评分。我们的模型通过5倍交叉验证进行评估，实现了肝癌免疫评分预测的曲线下面积（AUC）为0.88。此外，它对基因突变具有很强的预测能力，对肝癌中的CUB和Sushi多结构域1 （CSMD1）突变的AUC为0.99，对肿瘤蛋白53 （TP53）突变的AUC为0.86。这些结果突出了基于变压器的弱监督学习在计算病理学中的潜力。

{"title":"TEWS: Transformer-empowered weakly supervised prediction of immune score and genetic mutations in liver cancer from whole slide image","authors":"Bin Wang , Xianchuan Chen , Ying Li , Huiqing Wang , Keke Wang , Xiaoyan Li","doi":"10.1016/j.compbiolchem.2025.108837","DOIUrl":"10.1016/j.compbiolchem.2025.108837","url":null,"abstract":"<div><div>Whole Slide Imaging (WSI) plays a crucial role in predicting immune scores by providing detailed cellular and tissue-level insights, thereby enhancing pathological diagnosis accuracy and biomarker detection. Additionally, WSI contributes essential information for personalized treatment decisions. However, the application of deep learning models to WSI is hindered by two major challenges. First, the substantial data requirements, combined with the high cost and labor-intensive nature of sample annotation, limit the availability of well-labeled datasets. Second, the computational demands of these models pose practical constraints for many healthcare institutions. To address these challenges, we propose a weakly supervised deep learning model based on the transformer architecture. Specifically, we design a novel network incorporating the Swin Transformer, which, unlike traditional convolutional networks, emphasizes global feature extraction. This improves the accuracy of pseudo-label assignment during feature embedding. Additionally, we integrate a gated attention pooling mechanism and employ multi-instance learning (MIL), enabling, for the first time, immune score prediction directly from WSIs. Our model was evaluated through 5-fold cross-validation and achieved an area under the curve (AUC) of 0.88 for immune score prediction in liver cancer. Furthermore, it demonstrated strong predictive performance for genetic mutations, achieving an AUC of 0.99 for CUB and Sushi multiple domains 1 (CSMD1) mutations and 0.86 for Tumor Protein 53 (TP53) mutations in liver cancer. These results highlight the potential of transformer-based weakly supervised learning in computational pathology.</div></div>","PeriodicalId":10616,"journal":{"name":"Computational Biology and Chemistry","volume":"121 ","pages":"Article 108837"},"PeriodicalIF":3.1,"publicationDate":"2026-04-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145836510","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

Efficient drug–target affinity prediction via interaction features and parallel CNN–BiLSTM with attention 基于相互作用特征和并行CNN-BiLSTM的有效药物-靶点亲和力预测

IF 3.1 4区生物学 Q2 BIOLOGY

Computational Biology and Chemistry

Pub Date : 2026-04-01 Epub Date: 2025-12-27 DOI: 10.1016/j.compbiolchem.2025.108862

Jiffriya Mohamed Abdul Cader , M.A. Hakim Newton , Abdul Sattar

Drug–Target Affinity (DTA) prediction is critical for reducing failure rates in drug discovery, but existing deep learning methods often trade efficiency for accuracy. Existing CNN–LSTM methods for DTA have convolutional neural networks (CNNs) and long short-term memory (LSTM) in series. CNNs capture local patterns, while LSTMs model long-range dependencies. In series CNN–LSTM architectures, sequential dependencies are only modelled after convolutional compression, leading to loss of raw order information and limiting long-range interaction capture. Existing graph neural networks (GNNs) for DTA, on the other hand, capture structural interactions more explicitly but require large numbers of parameters, long training times, and high computational resources. To address the challenges, we propose EDTA (Efficient Deep Learning for Drug–Target Affinity prediction), a lighter architecture that combines CNNs and bidirectional LSTM (BiLSTM) in parallel and also uses attention mechanism to simultaneously capture both local structural patterns and global sequential dependencies. This design ensures that important interactions are exploited without the computational overhead. On benchmark datasets, EDTA achieves

r_{m}^{2}

values of 0.783 (Davis) and 0.787 (KIBA), outperforming state-of-the-art DTA methods while using fewer parameters, less memory, and up to five-fold faster inference. A virtual screening experiment on the Database of Useful Decoys: Enhanced (DUD-E) dataset further confirms its effectiveness in distinguishing binders from decoys. By emphasizing both efficiency and strong

r_{m}^{2}

performance, EDTA demonstrates that accurate DTA prediction can be achieved without sacrificing scalability or sustainability, offering a more practical solution for modern drug discovery.

药物靶标亲和力（DTA）预测对于降低药物发现的失败率至关重要，但现有的深度学习方法往往以效率为代价。现有的DTA CNN-LSTM方法将卷积神经网络（cnn）和长短期记忆（LSTM）串联在一起。cnn捕获局部模式，而lstm建模远程依赖关系。在串行CNN-LSTM体系结构中，顺序依赖关系仅在卷积压缩后建模，导致原始顺序信息丢失，限制了远程交互捕获。另一方面，现有的用于DTA的图神经网络（gnn）更明确地捕获结构相互作用，但需要大量的参数、较长的训练时间和高计算资源。为了解决这些挑战，我们提出了EDTA（高效深度学习药物-目标亲和力预测），这是一种轻量级的架构，将cnn和双向LSTM （BiLSTM）并行结合，并使用注意力机制同时捕获局部结构模式和全局顺序依赖关系。这种设计确保了在没有计算开销的情况下利用重要的交互。在基准数据集上，EDTA的rm2值为0.783 （Davis）和0.787 (KIBA)，优于最先进的DTA方法，同时使用更少的参数、更少的内存和高达五倍的推理速度。在有用诱饵数据库：增强（DUD-E）数据集上进行的虚拟筛选实验进一步证实了该方法在区分粘合剂和诱饵方面的有效性。通过强调效率和强大的rm2性能，EDTA表明可以在不牺牲可扩展性或可持续性的情况下实现准确的DTA预测，为现代药物发现提供更实用的解决方案。

{"title":"Efficient drug–target affinity prediction via interaction features and parallel CNN–BiLSTM with attention","authors":"Jiffriya Mohamed Abdul Cader , M.A. Hakim Newton , Abdul Sattar","doi":"10.1016/j.compbiolchem.2025.108862","DOIUrl":"10.1016/j.compbiolchem.2025.108862","url":null,"abstract":"<div><div>Drug–Target Affinity (DTA) prediction is critical for reducing failure rates in drug discovery, but existing deep learning methods often trade efficiency for accuracy. Existing CNN–LSTM methods for DTA have convolutional neural networks (CNNs) and long short-term memory (LSTM) in series. CNNs capture local patterns, while LSTMs model long-range dependencies. In series CNN–LSTM architectures, sequential dependencies are only modelled after convolutional compression, leading to loss of raw order information and limiting long-range interaction capture. Existing graph neural networks (GNNs) for DTA, on the other hand, capture structural interactions more explicitly but require large numbers of parameters, long training times, and high computational resources. To address the challenges, we propose EDTA (Efficient Deep Learning for Drug–Target Affinity prediction), a lighter architecture that combines CNNs and bidirectional LSTM (BiLSTM) in parallel and also uses attention mechanism to simultaneously capture both local structural patterns and global sequential dependencies. This design ensures that important interactions are exploited without the computational overhead. On benchmark datasets, EDTA achieves <span><math><msubsup><mrow><mi>r</mi></mrow><mrow><mi>m</mi></mrow><mrow><mn>2</mn></mrow></msubsup></math></span> values of 0.783 (Davis) and 0.787 (KIBA), outperforming state-of-the-art DTA methods while using fewer parameters, less memory, and up to five-fold faster inference. A virtual screening experiment on the Database of Useful Decoys: Enhanced (DUD-E) dataset further confirms its effectiveness in distinguishing binders from decoys. By emphasizing both efficiency and strong <span><math><msubsup><mrow><mi>r</mi></mrow><mrow><mi>m</mi></mrow><mrow><mn>2</mn></mrow></msubsup></math></span> performance, EDTA demonstrates that accurate DTA prediction can be achieved without sacrificing scalability or sustainability, offering a more practical solution for modern drug discovery.</div></div>","PeriodicalId":10616,"journal":{"name":"Computational Biology and Chemistry","volume":"121 ","pages":"Article 108862"},"PeriodicalIF":3.1,"publicationDate":"2026-04-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145880342","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

A 16S rRNA-based meta-analysis of gut microbiota in diabetic nephropathy using QIIME2 and publicly available NGS datasets 使用QIIME2和公开可用的NGS数据集对糖尿病肾病患者肠道微生物群进行基于16S rrna的荟萃分析

IF 3.1 4区生物学 Q2 BIOLOGY

Computational Biology and Chemistry

Pub Date : 2026-04-01 Epub Date: 2026-01-02 DOI: 10.1016/j.compbiolchem.2026.108876

Chahat Chopra , Deepak Kukkar , Harsimran Kaur

Gut microbial profiles can differ significantly between diabetic nephropathy (DN), diabetic patients, and healthy controls (HCs). The exact microbial taxa involved in DN progression is yet to be fully characterized. Therefore, this study aims to compare the gut microbiota od DN patients with diabetic and healthy individuals. Accordingly, this study executes a pioneering metanalytical view using the publicly available datasets (National centre for biotechnology information) to evaluate DN associated variation in gut-microbiota diversity. We hypothesize that the DN patients should have a smaller number of beneficial microbes along with a greater fraction of pathogenic microbial composition relative to the other two groups. Specifically, this report utilizes quantitative insights into microbial ecology 2 (QIIME2) platform to identify an association between gut microbiota composition and DN advancement. This novelty provides distinctive nature to our work in comparison to the broader diabetes microbiome studies. Our study enables comprehensive taxonomic profiling, differential abundance testing, and alpha and beta diversity analyses across multiple studies. A total of six studies were included, comprising 684 samples from both DN patients and HCs. Post quality control check, these samples were processed using QIIME2 platform for taxonomic profiling and diversity analysis to characterize microbial dysbiosis in DN patients in relation to other two groups. Alpha diversity indicates insignificant trend towards reduction of microbial diversity as observed using the Kruskal–Wallis test, Shannon index, Observed features (richness), and Faith’s PD analysis (p > 0.05). Additionally, beta diversity analyses revealed a trend toward microbial richness in DN compared to diabetic individuals and HCs, though differences were statically insignificant (P > 0.05). Taxonomic profiling showed a depletion of beneficial genera (e.g., Faecalibacterium, Roseburia, and Bifidobacterium) with false discovery rate (FDR)-adjusted p < 0.05. Contrarily, pathogenic and pro-inflammatory taxa including Escherichia-Shigella, Enterococcus, and Klebsiella showed higher abundance in the DN group (FDR-adjusted p < 0.05). These compositional shifts highlight pronounced gut dysbiosis during the transition from diabetes to DN, suggesting a potential association between gut-kidney axis.

糖尿病肾病（DN）、糖尿病患者和健康对照（hc）之间的肠道微生物谱可能存在显著差异。参与DN进展的确切微生物类群尚未得到充分表征。因此，本研究旨在比较糖尿病和健康人DN患者的肠道微生物群。因此，本研究使用公开可用的数据集（国家生物技术信息中心）执行开创性的元分析观点，以评估肠道微生物群多样性与DN相关的变化。我们假设，与其他两组相比，DN患者的有益微生物数量较少，而致病微生物组成的比例较大。具体而言，本报告利用微生物生态学（QIIME2）平台的定量见解来确定肠道微生物群组成与DN进展之间的关联。与更广泛的糖尿病微生物组研究相比，这种新颖性为我们的工作提供了独特的性质。我们的研究能够在多个研究中进行全面的分类分析、差异丰度测试和α和β多样性分析。共纳入6项研究，包括来自DN患者和hc患者的684份样本。质量控制检查后，这些样本使用QIIME2平台进行分类分析和多样性分析，以表征DN患者与其他两组患者的微生物生态失调。通过Kruskal-Wallis检验、Shannon指数、observed features（丰富度）和Faith’s PD分析（p >； 0.05），Alpha多样性表明微生物多样性减少的趋势不显著。此外，β -多样性分析显示，与糖尿病个体和hcc相比，DN中微生物丰富度呈上升趋势，但差异不显著（P >； 0.05）。分类学分析显示有益菌（如Faecalibacterium， Roseburia和Bifidobacterium）减少，假发现率（FDR）调整后p <； 0.05。相反，致病性和促炎分类群，包括埃希氏志贺氏菌、肠球菌和克雷伯氏菌，在DN组中丰度更高（经fdr校正p <； 0.05）。这些成分的变化突出了从糖尿病到DN转变过程中明显的肠道生态失调，表明肠肾轴之间存在潜在的关联。

{"title":"A 16S rRNA-based meta-analysis of gut microbiota in diabetic nephropathy using QIIME2 and publicly available NGS datasets","authors":"Chahat Chopra , Deepak Kukkar , Harsimran Kaur","doi":"10.1016/j.compbiolchem.2026.108876","DOIUrl":"10.1016/j.compbiolchem.2026.108876","url":null,"abstract":"<div><div>Gut microbial profiles can differ significantly between diabetic nephropathy (DN), diabetic patients, and healthy controls (HCs). The exact microbial taxa involved in DN progression is yet to be fully characterized. Therefore, this study aims to compare the gut microbiota od DN patients with diabetic and healthy individuals. Accordingly, this study executes a pioneering metanalytical view using the publicly available datasets (National centre for biotechnology information) to evaluate DN associated variation in gut-microbiota diversity. We hypothesize that the DN patients should have a smaller number of beneficial microbes along with a greater fraction of pathogenic microbial composition relative to the other two groups. Specifically, this report utilizes quantitative insights into microbial ecology 2 (QIIME2) platform to identify an association between gut microbiota composition and DN advancement. This novelty provides distinctive nature to our work in comparison to the broader diabetes microbiome studies. Our study enables comprehensive taxonomic profiling, differential abundance testing, and alpha and beta diversity analyses across multiple studies. A total of six studies were included, comprising 684 samples from both DN patients and HCs. Post quality control check, these samples were processed using QIIME2 platform for taxonomic profiling and diversity analysis to characterize microbial dysbiosis in DN patients in relation to other two groups. Alpha diversity indicates insignificant trend towards reduction of microbial diversity as observed using the Kruskal–Wallis test, Shannon index, Observed features (richness), and Faith’s PD analysis (p > 0.05). Additionally, beta diversity analyses revealed a trend toward microbial richness in DN compared to diabetic individuals and HCs, though differences were statically insignificant (P > 0.05). Taxonomic profiling showed a depletion of beneficial genera (e.g., <em>Faecalibacterium</em>, <em>Roseburia</em>, and <em>Bifidobacterium</em>) with false discovery rate (FDR)-adjusted <em>p</em> < 0.05. Contrarily, <em>p</em>athogenic and pro-inflammatory taxa including <em>Escherichia-Shigella</em>, <em>Enterococcus</em>, and <em>Klebsiella</em> showed higher abundance in the DN group (FDR-adjusted <em>p</em> < 0.05). These compositional shifts highlight pronounced gut dysbiosis during the transition from diabetes to DN, suggesting a potential association between gut-kidney axis.</div></div>","PeriodicalId":10616,"journal":{"name":"Computational Biology and Chemistry","volume":"121 ","pages":"Article 108876"},"PeriodicalIF":3.1,"publicationDate":"2026-04-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145880348","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

Network pharmacology combined with molecular docking and experimental validation reveals the anti-pharyngitis mechanism of Humei Qingyan granule 网络药理学结合分子对接和实验验证揭示了虎眉清炎颗粒的抗咽炎作用机制。

IF 3.1 4区生物学 Q2 BIOLOGY

Computational Biology and Chemistry

Pub Date : 2026-04-01 Epub Date: 2025-11-26 DOI: 10.1016/j.compbiolchem.2025.108799

Yalan Wu , Jian Liu , Changwei Yu , Sisi Zhang , Min Zhang

This study explored the molecular mechanisms of Humei Qingyan granule (HMQY) in treating pharyngitis via network pharmacology, molecular docking, and in vivo experiments. Public databases were used to collect HMQY’s main ingredients, their targets, and pharyngitis-related targets. The protein–protein interaction (PPI) network was constructed using the STRING 11.0 database, with hub targets pinpointed via Cytoscape. Gene Ontology (GO) and Kyoto Encyclopedia of Genes and Genomes (KEGG) enrichment analyses were employed to predict biological processes and pathway, followed by molecular docking of key components and targets. Finally, an acute pharyngitis rat model was used to validate the predicted targets and pathways, through enzyme-linked immunosorbent assay (ELISA), quantitative real-time polymerase chain reaction (qRT-PCR), and western blot assays. Results revealed active compounds like kaempferol, quercetin, and physovenine, with hub targets including AKT1, TP53, VEGFA, CASP3, and IL-1β. KEGG analysis highlighted the PI3K-Akt pathway and inflammation-cancer crosstalk as critical. Molecular docking showed strong binding between AKT1 and HMQY’s core components. In vivo experiments confirmed HMQY effectively alleviated pharyngeal pathological damage in rats with acute pharyngitis, and molecular biology experiments further revealed that HMQY treated pharyngitis by regulating the PI3K-Akt pathway, as shown by the inhibition of the protein expression of PI3K, p-Akt, and p-NF-κB, and a significant reduction in the level of IL-1β. In conclusion, this study preliminarily validated the active ingredients, key targets, and potential pathways of HMQY in the treatment of pharyngitis. The findings suggest HMQY exerts its therapeutic effects on pharyngitis primarily by inhibiting the PI3K-Akt/NF-κB signaling pathway.

本研究通过网络药理学、分子对接、体内实验等方法探讨虎眉清炎颗粒治疗咽炎的分子机制。利用公共数据库收集HMQY的主要成分、靶点及咽炎相关靶点。使用STRING 11.0数据库构建蛋白-蛋白相互作用（PPI）网络，并通过Cytoscape确定枢纽靶点。利用基因本体（GO）和京都基因与基因组百科全书（KEGG）富集分析预测生物过程和途径，然后进行关键组分和靶点的分子对接。最后，采用急性咽炎大鼠模型，通过酶联免疫吸附试验（ELISA）、定量实时聚合酶链反应（qRT-PCR）和western blot检测验证预测的靶点和途径。结果发现山奈酚、槲皮素和physovenine等活性化合物，其中心靶点包括AKT1、TP53、VEGFA、CASP3和IL-1β。KEGG分析强调PI3K-Akt通路和炎症-癌症串扰至关重要。分子对接显示，AKT1与HMQY的核心成分结合较强。体内实验证实HMQY可有效缓解急性咽炎大鼠咽部病理损伤，分子生物学实验进一步揭示HMQY通过调节PI3K- akt通路治疗咽炎，表现为抑制PI3K、p-Akt、p-NF-κB蛋白表达，显著降低IL-1β水平。综上所述，本研究初步验证了HMQY治疗咽炎的有效成分、关键靶点及潜在通路。研究结果提示，HMQY主要通过抑制PI3K-Akt/NF-κB信号通路发挥其治疗咽炎的作用。

{"title":"Network pharmacology combined with molecular docking and experimental validation reveals the anti-pharyngitis mechanism of Humei Qingyan granule","authors":"Yalan Wu , Jian Liu , Changwei Yu , Sisi Zhang , Min Zhang","doi":"10.1016/j.compbiolchem.2025.108799","DOIUrl":"10.1016/j.compbiolchem.2025.108799","url":null,"abstract":"<div><div>This study explored the molecular mechanisms of Humei Qingyan granule (HMQY) in treating pharyngitis via network pharmacology, molecular docking, and in vivo experiments. Public databases were used to collect HMQY’s main ingredients, their targets, and pharyngitis-related targets. The protein–protein interaction (PPI) network was constructed using the STRING 11.0 database, with hub targets pinpointed via Cytoscape. Gene Ontology (GO) and Kyoto Encyclopedia of Genes and Genomes (KEGG) enrichment analyses were employed to predict biological processes and pathway, followed by molecular docking of key components and targets. Finally, an acute pharyngitis rat model was used to validate the predicted targets and pathways, through enzyme-linked immunosorbent assay (ELISA), quantitative real-time polymerase chain reaction (qRT-PCR), and western blot assays. Results revealed active compounds like kaempferol, quercetin, and physovenine, with hub targets including AKT1, TP53, VEGFA, CASP3, and IL-1β. KEGG analysis highlighted the PI3K-Akt pathway and inflammation-cancer crosstalk as critical. Molecular docking showed strong binding between AKT1 and HMQY’s core components. In vivo experiments confirmed HMQY effectively alleviated pharyngeal pathological damage in rats with acute pharyngitis, and molecular biology experiments further revealed that HMQY treated pharyngitis by regulating the PI3K-Akt pathway, as shown by the inhibition of the protein expression of PI3K, p-Akt, and p-NF-κB, and a significant reduction in the level of IL-1β. In conclusion, this study preliminarily validated the active ingredients, key targets, and potential pathways of HMQY in the treatment of pharyngitis. The findings suggest HMQY exerts its therapeutic effects on pharyngitis primarily by inhibiting the PI3K-Akt/NF-κB signaling pathway.</div></div>","PeriodicalId":10616,"journal":{"name":"Computational Biology and Chemistry","volume":"121 ","pages":"Article 108799"},"PeriodicalIF":3.1,"publicationDate":"2026-04-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145717074","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

Mechanistic insights into marine-derived PDE6D inhibitors disrupting prenyl-binding to modulate leukemia-associated RAS trafficking 海洋来源的PDE6D抑制剂破坏戊酰结合以调节白血病相关RAS运输的机制见解。

IF 3.1 4区生物学 Q2 BIOLOGY

Computational Biology and Chemistry

Pub Date : 2026-04-01 Epub Date: 2025-12-03 DOI: 10.1016/j.compbiolchem.2025.108821

Perwez Alam , Ali Akhtar , Sarfaraz Ahmed , Zafrul Hasan

Leukemia is a significant clinical issue, and some of its subtypes are characterized by dysregulated prenylation-dependent signaling that contributes to disease onset and therapeutic resistance. Intracellular trafficking of farnesylated cargos of oncogenic signaling involves prenyl-binding chaperone PDE6D. Herein, we report a systematic computational platform toward the identification of structurally new marine fungal metabolites as prospective inhibitors of PDE6D. Our pipeline integrates high-throughput virtual screens, machine learning-guided QSAR models of molecules, density functional theory (DFT)-based molecular optimization, redocking, and long-timescale molecular dynamics (MD) investigations. Three lead molecules (CMNPD26276, CMNPD27347, and CMNPD29044) were identified using this pipeline based on their calculated inhibitory potency (pIC₅₀), desired electronic properties, and dynamic conformational stability within the binding crevice of PDE6D. Among these molecules, CMNPD29044 registered the maximum binding free energy and formed sustained interactions with and toward crucial binding residues of the binding of the PDE6D cargos. Additional MM-GBSA calculation and free energy landscape investigation confirmed a strong and energetically favorable binding conformation. Principal component analysis of these ligands suggested that they stabilize individual conformational states of the PDE6D and favor a possible interaction with a functionally essential site. This work suggests the therapeutic potential of targeting the prenyl-binding chaperone PDE6D as a novel non-enzymatic approach to leukemia treatment and identifies the strength of combining machine learning, quantum chemical, and atomistic simulation methods for the early identification of candidate drugs. Proposed marine candidates show promising leads toward experimental confirmation and toward SAR development for the treatment of PDE6D-targeted leukemia.

白血病是一个重要的临床问题，它的一些亚型的特征是失调的前置酰化依赖信号，这有助于疾病的发病和治疗耐药性。肿瘤信号的法酰化货物的细胞内运输涉及preyl结合伴侣PDE6D。在此，我们报告了一个系统的计算平台，用于鉴定结构上新的海洋真菌代谢物作为PDE6D的潜在抑制剂。我们的产品线集成了高通量虚拟屏幕、机器学习引导的QSAR分子模型、基于密度泛函数理论（DFT）的分子优化、再对接和长时间分子动力学（MD）研究。三个先导分子（CMNPD26276， CMNPD27347和CMNPD29044）根据其计算的抑制效能（pIC₅₀），所需的电子性质和PDE6D结合缝隙内的动态构象稳定性使用该管道进行鉴定。在这些分子中，CMNPD29044的结合自由能最大，并与PDE6D载物结合的关键结合残基形成持续相互作用。额外的MM-GBSA计算和自由能景观调查证实了一个强大的和能量有利的结合构象。这些配体的主成分分析表明，它们稳定了PDE6D的单个构象状态，并有利于与功能必需位点的可能相互作用。这项工作表明，靶向丙烯基结合伴侣PDE6D作为一种新的非酶治疗白血病的方法具有治疗潜力，并确定了机器学习、量子化学和原子模拟方法相结合的优势，可以早期识别候选药物。拟议的海洋候选物在实验确认和SAR开发方面显示出有希望的结果，可用于治疗pde6d靶向白血病。

{"title":"Mechanistic insights into marine-derived PDE6D inhibitors disrupting prenyl-binding to modulate leukemia-associated RAS trafficking","authors":"Perwez Alam , Ali Akhtar , Sarfaraz Ahmed , Zafrul Hasan","doi":"10.1016/j.compbiolchem.2025.108821","DOIUrl":"10.1016/j.compbiolchem.2025.108821","url":null,"abstract":"<div><div>Leukemia is a significant clinical issue, and some of its subtypes are characterized by dysregulated prenylation-dependent signaling that contributes to disease onset and therapeutic resistance. Intracellular trafficking of farnesylated cargos of oncogenic signaling involves prenyl-binding chaperone PDE6D. Herein, we report a systematic computational platform toward the identification of structurally new marine fungal metabolites as prospective inhibitors of PDE6D. Our pipeline integrates high-throughput virtual screens, machine learning-guided QSAR models of molecules, density functional theory (DFT)-based molecular optimization, redocking, and long-timescale molecular dynamics (MD) investigations. Three lead molecules (CMNPD26276, CMNPD27347, and CMNPD29044) were identified using this pipeline based on their calculated inhibitory potency (pIC₅₀), desired electronic properties, and dynamic conformational stability within the binding crevice of PDE6D. Among these molecules, CMNPD29044 registered the maximum binding free energy and formed sustained interactions with and toward crucial binding residues of the binding of the PDE6D cargos. Additional MM-GBSA calculation and free energy landscape investigation confirmed a strong and energetically favorable binding conformation. Principal component analysis of these ligands suggested that they stabilize individual conformational states of the PDE6D and favor a possible interaction with a functionally essential site. This work suggests the therapeutic potential of targeting the prenyl-binding chaperone PDE6D as a novel non-enzymatic approach to leukemia treatment and identifies the strength of combining machine learning, quantum chemical, and atomistic simulation methods for the early identification of candidate drugs. Proposed marine candidates show promising leads toward experimental confirmation and toward SAR development for the treatment of PDE6D-targeted leukemia.</div></div>","PeriodicalId":10616,"journal":{"name":"Computational Biology and Chemistry","volume":"121 ","pages":"Article 108821"},"PeriodicalIF":3.1,"publicationDate":"2026-04-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145727858","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

Genome-wide in-silico characterization and expression profiles of serpin gene family in foxtail millet (Setaria italica) 谷子（Setaria italica） serpin基因家族的全基因组分析及表达谱

IF 3.1 4区生物学 Q2 BIOLOGY

Computational Biology and Chemistry

Pub Date : 2026-04-01 Epub Date: 2025-11-26 DOI: 10.1016/j.compbiolchem.2025.108811

Areeba Batool , Shazia Rehman , Sadia Malik , Saqlain Haider , Banzeer Ahsan Abbasi , Shamim Akhtar

In plants, serpins form a superfamily of serine and cysteine protease inhibitors involved in stress tolerance and defense, with promising applications in pest management and agricultural biotechnology. This study identified and characterized 18 serpin (SiSRP) genes in the foxtail millet (Setaria italica) genome, distributed across different chromosomes. Analyses of gene structure, conserved motifs, and phylogeny indicated duplication events under strong purifying selection, driving the expansion and functional diversity of the serpin family. Prediction of cis-elements within promoter of SiSRP genes intimated that these serpin proteins are likely involved in plant’s adaptive responses to abiotic stresses. Synteny with sorghum, rice, maize, and barley suggested that these genes originated before the divergence of these species, showing a close relationship between SiSRPs and sorghum genes. Expression profiles from RNA-seq data revealed functional diversification of these genes in plant development and defense. Multiple sequence alignment highlighted the LR gene (SiSRP9–2) as a promising candidate for pest control, disease resistance, and regulation of programmed cell death. This work provides the first genome-wide analysis of serpin genes in foxtail millet, offering valuable insights for enhancing crop stress tolerance. However, experimental validation of identified stress-related genes is required for dissecting the functional roles of serpins in crops like foxtail millet.

在植物中，蛇形蛋白是丝氨酸和半胱氨酸蛋白酶抑制剂的一个超家族，与抗逆性和防御有关，在害虫管理和农业生物技术方面具有广阔的应用前景。本研究鉴定并鉴定了谷子（Setaria italica）基因组中分布在不同染色体上的18个蛇形蛋白（SiSRP）基因。基因结构、保守基序和系统发育分析表明，在强烈的净化选择下发生了重复事件，推动了serpin家族的扩展和功能多样性。对SiSRP基因启动子内顺式元件的预测表明，这些丝氨酸蛋白可能参与了植物对非生物胁迫的适应性反应。与高粱、水稻、玉米和大麦的同源性表明，这些基因起源于这些物种分化之前，表明sisrp与高粱基因关系密切。RNA-seq数据的表达谱揭示了这些基因在植物发育和防御中的功能多样化。多重序列比对表明，LR基因（SiSRP9-2）是一个有希望的害虫防治、抗病和程序性细胞死亡调控的候选基因。本研究首次对谷子蛇形蛋白基因进行了全基因组分析，为提高作物的抗逆性提供了有价值的见解。然而，为了剖析蛇形蛋白在谷子等作物中的功能作用，需要对已识别的应激相关基因进行实验验证。

{"title":"Genome-wide in-silico characterization and expression profiles of serpin gene family in foxtail millet (Setaria italica)","authors":"Areeba Batool , Shazia Rehman , Sadia Malik , Saqlain Haider , Banzeer Ahsan Abbasi , Shamim Akhtar","doi":"10.1016/j.compbiolchem.2025.108811","DOIUrl":"10.1016/j.compbiolchem.2025.108811","url":null,"abstract":"<div><div>In plants, serpins form a superfamily of serine and cysteine protease inhibitors involved in stress tolerance and defense, with promising applications in pest management and agricultural biotechnology. This study identified and characterized 18 serpin (SiSRP) genes in the foxtail millet (<em>Setaria italica</em>) genome, distributed across different chromosomes. Analyses of gene structure, conserved motifs, and phylogeny indicated duplication events under strong purifying selection, driving the expansion and functional diversity of the serpin family. Prediction of cis-elements within promoter of SiSRP genes intimated that these serpin proteins are likely involved in plant’s adaptive responses to abiotic stresses. Synteny with sorghum, rice, maize, and barley suggested that these genes originated before the divergence of these species, showing a close relationship between SiSRPs and sorghum genes. Expression profiles from RNA-seq data revealed functional diversification of these genes in plant development and defense. Multiple sequence alignment highlighted the LR gene (<em>SiSRP9–2</em>) as a promising candidate for pest control, disease resistance, and regulation of programmed cell death. This work provides the first genome-wide analysis of serpin genes in foxtail millet, offering valuable insights for enhancing crop stress tolerance. However, experimental validation of identified stress-related genes is required for dissecting the functional roles of serpins in crops like foxtail millet.</div></div>","PeriodicalId":10616,"journal":{"name":"Computational Biology and Chemistry","volume":"121 ","pages":"Article 108811"},"PeriodicalIF":3.1,"publicationDate":"2026-04-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145617431","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

Blast cell segmentation and leukemia classification using hybrid Deep Kronecker WideResNet using blood smear images 利用混合Deep Kronecker WideResNet对血液涂片图像进行胚细胞分割和白血病分类

IF 3.1 4区生物学 Q2 BIOLOGY

Computational Biology and Chemistry

Pub Date : 2026-04-01 Epub Date: 2025-11-25 DOI: 10.1016/j.compbiolchem.2025.108800

Mylapalli Ramesh , Naga Mallikharjunarao Billa , Venkata Kishore Kumar Rejeti , Cherukuri Phanindra

Acute Lymphoblastic Leukemia (ALL) is a dangerous form of leukemia, which disturbs the bone marrow and irregular White Blood Cells (WBC) for the persons of all age groups. Early diagnosis of leukemia is necessary to provide appropriate care and cure the patients. Still, the detection of ALL from Peripheral Blood Smear (PBS) images is complicated due to the size and shape of the cells. This variability creates more difficulty for the segmentation and classification of Leukemia. To solve these issues, the Deep Kronecker Wide Residual Network (DKWRN) is developed for classifying Leukemia. The blood smear images are preprocessed using adaptive Gaussian filtering; hence, the noise from the image is diminished. After that, RefineNet effectively segments the Blast cell from the image. Then, augmentation such as flipping, resizing, and rotation enhances the dimension of the image. Following augmentation, essential features like Binary Pattern of Phase Congruency (BPPC) with Discrete Cosine Transform (DCT), and statistical features are extracted. Finally, the DKWRN classifies the Leukemia into early Pre-B, Pre-B, Pro-B, and Hematogones. Here, the proposed DKWRN is the integration of WideResNet (WRN) and Deep Kronecker Network (DKN). Furthermore, the proposed model offered optimal accuracy, True Negative Rate (TNR), recall, precision, and F1-score of 92.12 %, 91.40 %, 90.36 %, 91.56 %, and 90.96 %, respectively. Furthermore, the proposed DKWRN model demonstrated notable enhancements in classification accuracy when compared to existing techniques. Specifically, its performance exceeded that of Bayesian-based optimized convolutional neural network (CNN) for ALL detection (BO-ALLCNN) by 5.91 %, Support vector machine (SVM) by 4.63 %, ResNet-18 combined with SVM by 4.04 %, Residual Convolutional Neural Network (ResNet-152) by 2.27 %, DKN by 1.51 %, and WRN by 0.55 %, respectively.

急性淋巴细胞白血病（Acute Lymphoblastic Leukemia， ALL）是一种危险的白血病，它会扰乱所有年龄组的人的骨髓和不规则的白细胞（WBC）。白血病的早期诊断对于提供适当的护理和治疗是必要的。然而，由于细胞的大小和形状，从外周血涂片（PBS）图像中检测ALL是复杂的。这种可变性给白血病的分割和分类带来了更多的困难。为了解决这些问题，提出了基于深度Kronecker Wide Residual Network （DKWRN）的白血病分类方法。采用自适应高斯滤波对血液涂片图像进行预处理；因此，图像中的噪声被减弱了。之后，RefineNet有效地将Blast细胞从图像中分割出来。然后，诸如翻转、调整大小和旋转等增强操作可以增强图像的尺寸。在增强后，提取了离散余弦变换（DCT）相同二值模式（BPPC）和统计特征等基本特征。最后，DKWRN将白血病分为早期Pre-B、Pre-B、Pro-B和造血细胞。本文提出的DKWRN是WideResNet （WRN）和Deep Kronecker Network （DKN）的集成。此外，该模型的最优准确率、真阴性率（TNR）、召回率、准确率和f1评分分别为92.12 %、91.40 %、90.36 %、91.56 %和90.96 %。此外，与现有技术相比，所提出的DKWRN模型在分类精度方面有显著提高。具体而言，其在ALL检测方面的性能分别超过基于贝叶斯的优化卷积神经网络（CNN）（BO-ALLCNN） 5.91 %、支持向量机（SVM） 4.63 %、ResNet-18与SVM结合的4.04 %、残差卷积神经网络（ResNet-152） 2.27 %、DKN 1.51 %、WRN 0.55 %。

{"title":"Blast cell segmentation and leukemia classification using hybrid Deep Kronecker WideResNet using blood smear images","authors":"Mylapalli Ramesh , Naga Mallikharjunarao Billa , Venkata Kishore Kumar Rejeti , Cherukuri Phanindra","doi":"10.1016/j.compbiolchem.2025.108800","DOIUrl":"10.1016/j.compbiolchem.2025.108800","url":null,"abstract":"<div><div>Acute Lymphoblastic Leukemia (ALL) is a dangerous form of leukemia, which disturbs the bone marrow and irregular White Blood Cells (WBC) for the persons of all age groups. Early diagnosis of leukemia is necessary to provide appropriate care and cure the patients. Still, the detection of ALL from Peripheral Blood Smear (PBS) images is complicated due to the size and shape of the cells. This variability creates more difficulty for the segmentation and classification of Leukemia. To solve these issues, the Deep Kronecker Wide Residual Network (DKWRN) is developed for classifying Leukemia. The blood smear images are preprocessed using adaptive Gaussian filtering; hence, the noise from the image is diminished. After that, RefineNet effectively segments the Blast cell from the image. Then, augmentation such as flipping, resizing, and rotation enhances the dimension of the image. Following augmentation, essential features like Binary Pattern of Phase Congruency (BPPC) with Discrete Cosine Transform (DCT), and statistical features are extracted. Finally, the DKWRN classifies the Leukemia into early Pre-B, Pre-B, Pro-B, and Hematogones. Here, the proposed DKWRN is the integration of WideResNet (WRN) and Deep Kronecker Network (DKN). Furthermore, the proposed model offered optimal accuracy, True Negative Rate (TNR), recall, precision, and F1-score of 92.12 %, 91.40 %, 90.36 %, 91.56 %, and 90.96 %, respectively. Furthermore, the proposed DKWRN model demonstrated notable enhancements in classification accuracy when compared to existing techniques. Specifically, its performance exceeded that of Bayesian-based optimized convolutional neural network (CNN) for ALL detection (BO-ALLCNN) by 5.91 %, Support vector machine (SVM) by 4.63 %, ResNet-18 combined with SVM by 4.04 %, Residual Convolutional Neural Network (ResNet-152) by 2.27 %, DKN by 1.51 %, and WRN by 0.55 %, respectively.</div></div>","PeriodicalId":10616,"journal":{"name":"Computational Biology and Chemistry","volume":"121 ","pages":"Article 108800"},"PeriodicalIF":3.1,"publicationDate":"2026-04-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145682573","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