Computers in biology and medicine最新文献_第5页

Transfer learning in spirometry: CNN models for automated flow-volume curve quality control in paediatric populations 肺活量测定中的迁移学习：用于儿科人群自动流量-容积曲线质量控制的 CNN 模型。

IF 7 2区医学 Q1 BIOLOGY

Computers in biology and medicine

Pub Date : 2024-11-13 DOI: 10.1016/j.compbiomed.2024.109341

Carla Martins , Henrique Barros , André Moreira

Problem

Current spirometers face challenges in evaluating acceptability criteria, often requiring manual visual inspection by trained specialists. Automating this process could improve diagnostic workflows and reduce variability in test assessments.

Aim

This study aimed to apply transfer learning to convolutional neural networks (CNNs) to automate the classification of spirometry flow-volume curves based on acceptability criteria.

Methods

A total of 5287 spirometry flow-volume curves were divided into three categories: (A) all criteria met, (B) early termination, and (C) non-acceptable results. Six CNN models (VGG16, InceptionV3, Xception, ResNet152V2, InceptionResNetV2, DenseNet121) were trained using a balanced dataset after data augmentation. The models' performance was evaluated on part of the original unbalanced dataset with accuracy, precision, recall, and F1-score metrics.

Results

VGG16 achieved the highest accuracy at 93.9 %, while ResNet152V2 had the lowest at 83.0 %. Non-acceptable curves (Group C) were the easiest to classify, with precision reaching at least 87.7 %. Early termination curves (Group B) were the most challenging, with precision ranging from 75.0 % to 90.3 %.

Conclusion

CNN models, particularly VGG16, show promise in automating spirometry quality control, potentially reducing the need for manual inspection by specialized technicians. This approach can streamline spirometry assessments, offering consistent, high-quality diagnostics even in non-specialized or low-resource environments.

问题：目前的肺活量计在评估可接受性标准方面面临挑战，通常需要训练有素的专家进行人工目测。目的：本研究旨在将迁移学习应用于卷积神经网络（CNN），根据可接受性标准对肺活量流量-容积曲线进行自动分类：共有 5287 条肺活量流速-容积曲线被分为三类：(A) 符合所有标准，(B) 提前终止，(C) 不可接受的结果。经过数据增强后，使用平衡数据集对六个 CNN 模型（VGG16、InceptionV3、Xception、ResNet152V2、InceptionResNetV2、DenseNet121）进行了训练。在部分原始非平衡数据集上对模型的性能进行了评估，评估指标包括准确度、精确度、召回率和 F1 分数：结果：VGG16 的准确率最高，为 93.9%，而 ResNet152V2 的准确率最低，为 83.0%。不可接受曲线（C 组）最容易分类，精确度至少达到 87.7%。早期终止曲线（B 组）最具挑战性，精确度从 75.0 % 到 90.3 % 不等：CNN模型，尤其是VGG16，有望实现肺活量质量控制的自动化，从而减少专业技术人员进行人工检查的需要。这种方法可以简化肺活量评估，即使在非专业或资源匮乏的环境中也能提供一致的高质量诊断。

{"title":"Transfer learning in spirometry: CNN models for automated flow-volume curve quality control in paediatric populations","authors":"Carla Martins , Henrique Barros , André Moreira","doi":"10.1016/j.compbiomed.2024.109341","DOIUrl":"10.1016/j.compbiomed.2024.109341","url":null,"abstract":"<div><h3>Problem</h3><div>Current spirometers face challenges in evaluating acceptability criteria, often requiring manual visual inspection by trained specialists. Automating this process could improve diagnostic workflows and reduce variability in test assessments.</div></div><div><h3>Aim</h3><div>This study aimed to apply transfer learning to convolutional neural networks (CNNs) to automate the classification of spirometry flow-volume curves based on acceptability criteria.</div></div><div><h3>Methods</h3><div>A total of 5287 spirometry flow-volume curves were divided into three categories: (A) all criteria met, (B) early termination, and (C) non-acceptable results. Six CNN models (VGG16, InceptionV3, Xception, ResNet152V2, InceptionResNetV2, DenseNet121) were trained using a balanced dataset after data augmentation. The models' performance was evaluated on part of the original unbalanced dataset with accuracy, precision, recall, and F1-score metrics.</div></div><div><h3>Results</h3><div>VGG16 achieved the highest accuracy at 93.9 %, while ResNet152V2 had the lowest at 83.0 %. Non-acceptable curves (Group C) were the easiest to classify, with precision reaching at least 87.7 %. Early termination curves (Group B) were the most challenging, with precision ranging from 75.0 % to 90.3 %.</div></div><div><h3>Conclusion</h3><div>CNN models, particularly VGG16, show promise in automating spirometry quality control, potentially reducing the need for manual inspection by specialized technicians. This approach can streamline spirometry assessments, offering consistent, high-quality diagnostics even in non-specialized or low-resource environments.</div></div>","PeriodicalId":10578,"journal":{"name":"Computers in biology and medicine","volume":"184 ","pages":"Article 109341"},"PeriodicalIF":7.0,"publicationDate":"2024-11-13","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142615857","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Evolutionary bioinformatics with veiled biological database for health care operations 进化生物信息学与用于医疗保健业务的隐藏生物数据库。

IF 7 2区医学 Q1 BIOLOGY

Computers in biology and medicine

Pub Date : 2024-11-13 DOI: 10.1016/j.compbiomed.2024.109418

Hariprasath Manoharan , S.A. Edalatpanah

The tremendous growth of biological data processing systems in the realm of health care applications has made real-time information accessible to everyone with no processing lags. Bioinformatics is even integrated into most wireless technology applications to account for all physical characteristics. The planned model focuses on evolutionary bioinformatics for medical sensor applications in health care. The optimization scenario is executed by combining genetic and ant colony optimization methods (GACO). In the proposed technique, the design concerns are implemented with appropriate transmitting and receiving modules, and individual bits are framed for extra bioinformatics data processing components. a design that completely minimizes all errors in the big data processing stage. Such a design completely lowers the overall error in the huge data processing state since all channels can be accessed in accordance with the framed bits. Furthermore, the quality of service is maximized because all channels carrying bioinformatics data are kept at high quality bits, increasing utility rates. The experiments were conducted using five scenarios to evaluate the effectiveness of the proposed design. The findings indicate that the proposed technique can handle bioinformatics data for healthcare in real time with a service quality of 95 %.

生物数据处理系统在医疗保健应用领域的迅猛发展，使每个人都能获得实时信息，而且没有处理滞后问题。生物信息学甚至被整合到大多数无线技术应用中，以考虑到所有物理特性。计划中的模型侧重于进化生物信息学在医疗保健领域的医疗传感器应用。优化方案通过结合遗传和蚁群优化方法（GACO）来执行。在所提出的技术中，设计关注点通过适当的发射和接收模块来实现，并为额外的生物信息学数据处理组件设置了单独的位。由于所有信道都可根据帧位进行访问，因此这种设计可完全降低海量数据处理状态下的总体误差。此外，由于所有携带生物信息学数据的信道都能保持高质量比特，从而提高了实用率，因此服务质量也得到了最大化。实验使用了五种场景来评估所提设计的有效性。实验结果表明，所提出的技术可以实时处理用于医疗保健的生物信息数据，服务质量达到 95%。

{"title":"Evolutionary bioinformatics with veiled biological database for health care operations","authors":"Hariprasath Manoharan , S.A. Edalatpanah","doi":"10.1016/j.compbiomed.2024.109418","DOIUrl":"10.1016/j.compbiomed.2024.109418","url":null,"abstract":"<div><div>The tremendous growth of biological data processing systems in the realm of health care applications has made real-time information accessible to everyone with no processing lags. Bioinformatics is even integrated into most wireless technology applications to account for all physical characteristics. The planned model focuses on evolutionary bioinformatics for medical sensor applications in health care. The optimization scenario is executed by combining genetic and ant colony optimization methods (GACO). In the proposed technique, the design concerns are implemented with appropriate transmitting and receiving modules, and individual bits are framed for extra bioinformatics data processing components. a design that completely minimizes all errors in the big data processing stage. Such a design completely lowers the overall error in the huge data processing state since all channels can be accessed in accordance with the framed bits. Furthermore, the quality of service is maximized because all channels carrying bioinformatics data are kept at high quality bits, increasing utility rates. The experiments were conducted using five scenarios to evaluate the effectiveness of the proposed design. The findings indicate that the proposed technique can handle bioinformatics data for healthcare in real time with a service quality of 95 %.</div></div>","PeriodicalId":10578,"journal":{"name":"Computers in biology and medicine","volume":"184 ","pages":"Article 109418"},"PeriodicalIF":7.0,"publicationDate":"2024-11-13","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142616403","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

SPE-YOLO: A deep learning model focusing on small pulmonary embolism detection SPE-YOLO：专注于小型肺栓塞检测的深度学习模型。

IF 7 2区医学 Q1 BIOLOGY

Computers in biology and medicine

Pub Date : 2024-11-13 DOI: 10.1016/j.compbiomed.2024.109402

Houde Wu , Qifei Xu , Xinliu He , Haijun Xu , Yun Wang , Li Guo

Objectives

By developing the deep learning model SPE-YOLO, the detection of small pulmonary embolism has been improved, leading to more accurate identification of this condition. This advancement aims to better serve medical diagnosis and treatment.

Methods

This retrospective study analyzed images of 142 patients from Tianjin Medical University General Hospital using YOLOv8 as the foundational framework. Firstly, a small detection head P2 was added to better capture and identify small targets. Secondly, the SEAttention mechanism was integrated into the model to enhance focus on crucial features and optimize detection accuracy. Lastly, the feature extraction process was refined by introducing ODConv convolution to capture more comprehensive contextual information, thereby enhancing the detection performance of small pulmonary embolisms. The model's practical application ability was evaluated using 2000 cases from the RSNA dataset as an external test set.

Results

In the Tianjin test set, our model achieved a precision of 84.20 % and an accuracy of 81.50 %. This represents an improvement of approximately 5 % and 4 % respectively compared to the original YOLOv8. F1 scores, recall rates and average accuracy have also increased by 4 %, 5 %, 6 %, respectively. In data from the external validation set of RSNA, SPE-YOLO exhibited its effectiveness with a sensitivity of 90.70 % and an accuracy of 86.45 %.

Conclusion

The SPE-YOLO algorithm demonstrates strong capability in identifying small pulmonary embolisms, offering clinicians a more accurate and efficient diagnostic tool. This advancement is expected to enhance the quality of pulmonary embolism diagnosis and support the progress of medical services.

目的：通过开发深度学习模型 SPE-YOLO，改进了对小肺栓塞的检测，从而更准确地识别这种疾病。这一进步旨在更好地服务于医疗诊断和治疗：这项回顾性研究以 YOLOv8 为基础框架，分析了天津医科大学总医院 142 名患者的图像。首先，增加了一个小型检测头 P2，以更好地捕捉和识别小目标。其次，将 SEAttention 机制集成到模型中，以加强对关键特征的关注，优化检测精度。最后，通过引入 ODConv 卷积对特征提取过程进行了改进，以获取更全面的上下文信息，从而提高对小肺栓塞的检测性能。以 RSNA 数据集的 2000 个病例作为外部测试集，对模型的实际应用能力进行了评估：在天津测试集中，我们的模型达到了 84.20 % 的精确度和 81.50 % 的准确度。与最初的 YOLOv8 相比，分别提高了约 5% 和 4%。F1 分数、召回率和平均准确率也分别提高了 4%、5% 和 6%。在 RSNA 外部验证集的数据中，SPE-YOLO 的灵敏度为 90.70%，准确率为 86.45%，显示了其有效性：SPE-YOLO算法在识别微小肺栓塞方面表现出很强的能力，为临床医生提供了更准确、更高效的诊断工具。这一进步有望提高肺栓塞诊断的质量，促进医疗服务的进步。

{"title":"SPE-YOLO: A deep learning model focusing on small pulmonary embolism detection","authors":"Houde Wu , Qifei Xu , Xinliu He , Haijun Xu , Yun Wang , Li Guo","doi":"10.1016/j.compbiomed.2024.109402","DOIUrl":"10.1016/j.compbiomed.2024.109402","url":null,"abstract":"<div><h3>Objectives</h3><div>By developing the deep learning model SPE-YOLO, the detection of small pulmonary embolism has been improved, leading to more accurate identification of this condition. This advancement aims to better serve medical diagnosis and treatment.</div></div><div><h3>Methods</h3><div>This retrospective study analyzed images of 142 patients from Tianjin Medical University General Hospital using YOLOv8 as the foundational framework. Firstly, a small detection head P2 was added to better capture and identify small targets. Secondly, the SEAttention mechanism was integrated into the model to enhance focus on crucial features and optimize detection accuracy. Lastly, the feature extraction process was refined by introducing ODConv convolution to capture more comprehensive contextual information, thereby enhancing the detection performance of small pulmonary embolisms. The model's practical application ability was evaluated using 2000 cases from the RSNA dataset as an external test set.</div></div><div><h3>Results</h3><div>In the Tianjin test set, our model achieved a precision of 84.20 % and an accuracy of 81.50 %. This represents an improvement of approximately 5 % and 4 % respectively compared to the original YOLOv8. F1 scores, recall rates and average accuracy have also increased by 4 %, 5 %, 6 %, respectively. In data from the external validation set of RSNA, SPE-YOLO exhibited its effectiveness with a sensitivity of 90.70 % and an accuracy of 86.45 %.</div></div><div><h3>Conclusion</h3><div>The SPE-YOLO algorithm demonstrates strong capability in identifying small pulmonary embolisms, offering clinicians a more accurate and efficient diagnostic tool. This advancement is expected to enhance the quality of pulmonary embolism diagnosis and support the progress of medical services.</div></div>","PeriodicalId":10578,"journal":{"name":"Computers in biology and medicine","volume":"184 ","pages":"Article 109402"},"PeriodicalIF":7.0,"publicationDate":"2024-11-13","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142615722","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Computational-driven discovery of AI-2 quorum sensing inhibitor targeting the 5′- methylthioadenosine/S-adenosylhomocysteine nucleosidase (MTAN) to combat drug-resistant Helicobacter pylori 通过计算发现针对 5'- 甲基硫代腺苷/S-腺苷高半胱氨酸核苷酸酶（MTAN）的 AI-2 法定量感应抑制剂，以抗击耐药性幽门螺旋杆菌。

IF 7 2区医学 Q1 BIOLOGY

Computers in biology and medicine

Pub Date : 2024-11-13 DOI: 10.1016/j.compbiomed.2024.109409

Manish Kumar , Avinash Karkada Ashok , Thejaswi Bhat , Krishnakumar Ballamoole , Pavan Gollapalli

MTAN is an attainable therapeutic target for H. pylori because it may minimize virulence production, limit resistance, and impair quorum sensing without affecting gut flora. Here, 457 compounds with anti-H. pylori activity were methodically analyzed, revealing a diverse array of chemical classes and unique compounds. Molecular docking studies identified three potential compounds with high binding affinities, Dehydrocostus lactone, keramamine B, and ZINC00013531409, each having binding affinity of −7.9, −9.2, and −8.3 kcal/mol, respectively. Molecular dynamics simulations of the ZINC00013531409-MTAN interactions in comparison with Apo-MTAN demonstrated stability and interactions of 300 ns, with key residues involved in protein-ligand binding illuminated. Analysis of hydrogen bonds (Ile52, Met174, and Arg194) and secondary structure variations further elucidated the binding interactions and conformational changes within the complex. Binding free energy calculations shed light on the energetics and interactions governing the complex formation of the ZINC00013531409-MTAN complex. PCA elucidated the dominant modes of motion, along with FEL revealed the energetically favorable states and then DCCM shed light on the correlated motions between residues. Overall, this study offers a detailed computational evaluation of ZINC00013531409 with anti-H. pylori activity, highlighting toxicity profile, conformational stability, and binding interactions, providing a foundation for further drug development efforts toward bacterial resistance.

MTAN是幽门螺杆菌的一个可实现的治疗靶点，因为它可以在不影响肠道菌群的情况下最大限度地减少毒力产生、限制抗药性并损害法定量感应。本文对 457 种具有抗幽门螺杆菌活性的化合物进行了方法学分析，发现了一系列不同的化学类别和独特的化合物。分子对接研究发现了三种具有高结合亲和力的潜在化合物，即去氢木香内酯、角胺 B 和 ZINC00013531409，它们的结合亲和力分别为 -7.9、-9.2 和 -8.3 kcal/mol。与 Apo-MTAN 相比，ZINC00013531409-MTAN 相互作用的分子动力学模拟显示了 300 ns 的稳定性和相互作用，并阐明了参与蛋白质配体结合的关键残基。对氢键（Ile52、Met174 和 Arg194）和二级结构变化的分析进一步阐明了复合物内部的结合相互作用和构象变化。结合自由能计算揭示了支配 ZINC00013531409-MTAN 复合物形成的能量和相互作用。PCA 阐明了主要的运动模式，FEL 揭示了能量上有利的状态，然后 DCCM 揭示了残基之间的相关运动。总之，本研究对具有抗幽门螺杆菌活性的 ZINC00013531409 进行了详细的计算评估，突出了其毒性特征、构象稳定性和结合相互作用，为进一步开发抗菌药物奠定了基础。

{"title":"Computational-driven discovery of AI-2 quorum sensing inhibitor targeting the 5′- methylthioadenosine/S-adenosylhomocysteine nucleosidase (MTAN) to combat drug-resistant Helicobacter pylori","authors":"Manish Kumar , Avinash Karkada Ashok , Thejaswi Bhat , Krishnakumar Ballamoole , Pavan Gollapalli","doi":"10.1016/j.compbiomed.2024.109409","DOIUrl":"10.1016/j.compbiomed.2024.109409","url":null,"abstract":"<div><div>MTAN is an attainable therapeutic target for <em>H. pylori</em> because it may minimize virulence production, limit resistance, and impair quorum sensing without affecting gut flora. Here, 457 compounds with anti-<em>H. pylori</em> activity were methodically analyzed, revealing a diverse array of chemical classes and unique compounds. Molecular docking studies identified three potential compounds with high binding affinities, Dehydrocostus lactone, keramamine B, and ZINC00013531409, each having binding affinity of −7.9, −9.2, and −8.3 kcal/mol, respectively. Molecular dynamics simulations of the ZINC00013531409-MTAN interactions in comparison with Apo-MTAN demonstrated stability and interactions of 300 ns, with key residues involved in protein-ligand binding illuminated. Analysis of hydrogen bonds (Ile52, Met174, and Arg194) and secondary structure variations further elucidated the binding interactions and conformational changes within the complex. Binding free energy calculations shed light on the energetics and interactions governing the complex formation of the ZINC00013531409-MTAN complex. PCA elucidated the dominant modes of motion, along with FEL revealed the energetically favorable states and then DCCM shed light on the correlated motions between residues. Overall, this study offers a detailed computational evaluation of ZINC00013531409 with anti-<em>H. pylori</em> activity, highlighting toxicity profile, conformational stability, and binding interactions, providing a foundation for further drug development efforts toward bacterial resistance.</div></div>","PeriodicalId":10578,"journal":{"name":"Computers in biology and medicine","volume":"184 ","pages":"Article 109409"},"PeriodicalIF":7.0,"publicationDate":"2024-11-13","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142616397","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Feature-centric registration of large deformed images using transformers and correlation distance 使用变换器和相关距离对大型变形图像进行以特征为中心的配准。

IF 7 2区医学 Q1 BIOLOGY

Computers in biology and medicine

Pub Date : 2024-11-12 DOI: 10.1016/j.compbiomed.2024.109356

Heeyeon Kim , Minkyung Lee , Bohyoung Kim , Yeong-Gil Shin , Minyoung Chung

In deformable medical image registration, both a robust backbone registration network and a suitable similarity metric are essential. This paper introduces a robust registration network combined with a feature-based loss function, specifically designed to handle large deformations and address the challenge of the absence of ground truth data. Tackling large deformations typically requires either expanding the receptive field or breaking down extensive deformations into smaller, more manageable ones. We address this challenge through two key network components: the coarse-to-fine estimation of the target displacement vector field (DVF) and the integration of the Transformer’s feature attention mechanism. To further enhance registration performance, we propose a novel feature correlation-based distance metric that leverages the symmetric properties of the correlation matrix to efficiently exploit feature correlations. Additionally, by utilizing the features extracted directly from the registration network, we eliminate the need for additional feature extraction networks. Experimental results demonstrate that our feature correlation-based loss function is particularly effective in achieving accurate registration in the absence of ground truth data. Our method has proven successful in both mono-modality abdomen CT registration and brain MRI atlas registration, leading to improvements in Dice similarity coefficient and other evaluation metrics.

在可变形医学影像配准中，稳健的骨干配准网络和合适的相似度度量至关重要。本文介绍了一种与基于特征的损失函数相结合的鲁棒性配准网络，专门用于处理大变形和解决缺乏地面实况数据的难题。处理大变形通常需要扩大感受野或将大变形分解成更小、更易于处理的变形。我们通过两个关键的网络组件来应对这一挑战：从粗到细的目标位移矢量场（DVF）估算和 Transformer 特征关注机制的整合。为了进一步提高配准性能，我们提出了一种基于特征相关性的新型距离度量，该度量利用相关矩阵的对称特性来有效利用特征相关性。此外，通过直接利用从注册网络中提取的特征，我们不再需要额外的特征提取网络。实验结果表明，在没有地面实况数据的情况下，我们基于特征相关性的损失函数在实现精确配准方面特别有效。事实证明，我们的方法在单模态腹部 CT 配准和脑部 MRI 图集配准中都取得了成功，提高了 Dice 相似性系数和其他评价指标。

{"title":"Feature-centric registration of large deformed images using transformers and correlation distance","authors":"Heeyeon Kim , Minkyung Lee , Bohyoung Kim , Yeong-Gil Shin , Minyoung Chung","doi":"10.1016/j.compbiomed.2024.109356","DOIUrl":"10.1016/j.compbiomed.2024.109356","url":null,"abstract":"<div><div>In deformable medical image registration, both a robust backbone registration network and a suitable similarity metric are essential. This paper introduces a robust registration network combined with a feature-based loss function, specifically designed to handle large deformations and address the challenge of the absence of ground truth data. Tackling large deformations typically requires either expanding the receptive field or breaking down extensive deformations into smaller, more manageable ones. We address this challenge through two key network components: the coarse-to-fine estimation of the target displacement vector field (DVF) and the integration of the Transformer’s feature attention mechanism. To further enhance registration performance, we propose a novel feature correlation-based distance metric that leverages the symmetric properties of the correlation matrix to efficiently exploit feature correlations. Additionally, by utilizing the features extracted directly from the registration network, we eliminate the need for additional feature extraction networks. Experimental results demonstrate that our feature correlation-based loss function is particularly effective in achieving accurate registration in the absence of ground truth data. Our method has proven successful in both mono-modality abdomen CT registration and brain MRI atlas registration, leading to improvements in Dice similarity coefficient and other evaluation metrics.</div></div>","PeriodicalId":10578,"journal":{"name":"Computers in biology and medicine","volume":"184 ","pages":"Article 109356"},"PeriodicalIF":7.0,"publicationDate":"2024-11-12","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142616412","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Exploring natural products potential: A similarity-based target prediction tool for natural products 探索天然产品的潜力：基于相似性的天然产品目标预测工具。

IF 7 2区医学 Q1 BIOLOGY

Computers in biology and medicine

Pub Date : 2024-11-12 DOI: 10.1016/j.compbiomed.2024.109351

Abeer Abdulhakeem Mansour Alhasbary , Nurul Hashimah Ahamed Hassain Malim , Siti Zuraidah Mohamad Zobir

Natural products are invaluable resources in drug discovery due to their substantial structural diversity. However, predicting their interactions with druggable protein targets remains a challenge, primarily due to the limited availability of bioactivity data. This study introduces CTAPred (Compound-Target Activity Prediction), an open-source command-line tool designed to predict potential protein targets for natural products. CTAPred employs a two-stage approach, combining fingerprinting and similarity-based search techniques to identify likely drug targets for these bioactive compounds. Despite its simplicity, the tool's performance is comparable to that of more complex methods, demonstrating proficiency in target retrieval for natural product compounds. Furthermore, this study explores the optimal number of reference compounds most similar to the query compound, aiming to refine target prediction accuracy. The findings demonstrated the superior performance of considering only the most similar reference compounds for target prediction. CTAPred is freely available at https://github.com/Alhasbary/CTAPred, offering a valuable resource for deciphering natural product-target associations and advancing drug discovery.

天然产物结构多样，是药物发现的宝贵资源。然而，主要由于生物活性数据的可用性有限，预测它们与可药用蛋白质靶点的相互作用仍然是一项挑战。本研究介绍了 CTAPred（化合物-靶标活性预测），这是一种开源命令行工具，旨在预测天然产物的潜在蛋白质靶标。CTAPred 采用两阶段方法，结合指纹识别和基于相似性的搜索技术，为这些生物活性化合物确定可能的药物靶点。尽管该工具非常简单，但其性能可与更复杂的方法相媲美，证明了其在天然产物化合物靶标检索方面的能力。此外，本研究还探讨了与查询化合物最相似的参考化合物的最佳数量，旨在提高靶标预测的准确性。研究结果表明，仅考虑最相似的参考化合物进行目标预测的性能更优越。CTAPred 可在 https://github.com/Alhasbary/CTAPred 免费获取，它为破译天然产物-靶标关联和推进药物发现提供了宝贵的资源。

{"title":"Exploring natural products potential: A similarity-based target prediction tool for natural products","authors":"Abeer Abdulhakeem Mansour Alhasbary , Nurul Hashimah Ahamed Hassain Malim , Siti Zuraidah Mohamad Zobir","doi":"10.1016/j.compbiomed.2024.109351","DOIUrl":"10.1016/j.compbiomed.2024.109351","url":null,"abstract":"<div><div>Natural products are invaluable resources in drug discovery due to their substantial structural diversity. However, predicting their interactions with druggable protein targets remains a challenge, primarily due to the limited availability of bioactivity data. This study introduces CTAPred (Compound-Target Activity Prediction), an open-source command-line tool designed to predict potential protein targets for natural products. CTAPred employs a two-stage approach, combining fingerprinting and similarity-based search techniques to identify likely drug targets for these bioactive compounds. Despite its simplicity, the tool's performance is comparable to that of more complex methods, demonstrating proficiency in target retrieval for natural product compounds. Furthermore, this study explores the optimal number of reference compounds most similar to the query compound, aiming to refine target prediction accuracy. The findings demonstrated the superior performance of considering only the most similar reference compounds for target prediction. CTAPred is freely available at <span><span>https://github.com/Alhasbary/CTAPred</span><svg><path></path></svg></span>, offering a valuable resource for deciphering natural product-target associations and advancing drug discovery.</div></div>","PeriodicalId":10578,"journal":{"name":"Computers in biology and medicine","volume":"184 ","pages":"Article 109351"},"PeriodicalIF":7.0,"publicationDate":"2024-11-12","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142616408","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Using advanced machine learning algorithms to predict academic major completion: A cross-sectional study 使用先进的机器学习算法预测学业专业完成情况：横断面研究

IF 7 2区医学 Q1 BIOLOGY

Computers in biology and medicine

Pub Date : 2024-11-12 DOI: 10.1016/j.compbiomed.2024.109372

Alireza Kordbagheri , Mohammadreza Kordbagheri , Natalie Tayim , Abdulnaser Fakhrou , Mohammadreza Davoudi

Background

Existing prediction methods for academic majors based on personality traits have notable gaps, including limited model complexity and generalizability.The current study aimed to utilize advanced Machine Learning (ML) algorithms with smoothing functions to predict academic majors completed based on personality subscales.

Methods

We used reports from 59,413 individuals to perform the current study. All advanced algorithms implemented in this article were based on R software (version 4.1.3, R Core Team, 2021). All model parameters were optimized based on resampling and cross-validation (CV). In addition, pseudo-R² as a robust metric has been used to compare the performance of models, which, unlike most studies, considers the quality of model-predicted probabilities.

Result

The results indicated that advanced ML models' performance on training and test data was superior to logistic regression. Pseudo-R² and AUC results showed that advanced models such as kNN, GBE, and RF had the highest scores based on test data compared to other models. The pseudo-R² values for the models used in this study varied across the test dataset; the lowest value belonged to the logistic regression algorithm at .022, and the highest value was recorded for the kNN algorithm at .099. The agreeableness subscale is the most influential component in predicting the completion of university education, followed by conscientiousness and emotional stability.

Conclusion

The potential of advanced methods to enhance the accuracy and validity of predictions is a promising development in our field. Their performance, particularly in handling large data sets with complex patterns, is a reason for optimism about the future of research in this area.

背景：本研究旨在利用先进的机器学习（ML）算法和平滑函数来预测基于人格分量表完成的学业专业：本研究使用了来自 59,413 人的报告。本文采用的所有高级算法均基于 R 软件（4.1.3 版，R 核心团队，2021 年）。所有模型参数都根据重采样和交叉验证（CV）进行了优化。此外，还使用了伪 R2 作为稳健指标来比较模型的性能，与大多数研究不同的是，伪 R2 考虑了模型预测概率的质量：结果：结果表明，高级 ML 模型在训练和测试数据上的表现优于逻辑回归。伪 R2 和 AUC 结果显示，与其他模型相比，kNN、GBE 和 RF 等高级模型在测试数据上的得分最高。本研究中使用的模型的伪 R2 值在整个测试数据集中各不相同；逻辑回归算法的伪 R2 值最低，为 0.022，而 kNN 算法的伪 R2 值最高，为 0.099。在预测完成大学教育方面，"合意性 "分量表的影响最大，其次是 "自觉性 "和 "情绪稳定性"：先进方法在提高预测的准确性和有效性方面的潜力是我们这个领域的一个很有前途的发展。这些方法的性能，尤其是在处理具有复杂模式的大型数据集方面的性能，是我们对这一领域的研究前景持乐观态度的原因。

{"title":"Using advanced machine learning algorithms to predict academic major completion: A cross-sectional study","authors":"Alireza Kordbagheri , Mohammadreza Kordbagheri , Natalie Tayim , Abdulnaser Fakhrou , Mohammadreza Davoudi","doi":"10.1016/j.compbiomed.2024.109372","DOIUrl":"10.1016/j.compbiomed.2024.109372","url":null,"abstract":"<div><h3>Background</h3><div>Existing prediction methods for academic majors based on personality traits have notable gaps, including limited model complexity and generalizability.The current study aimed to utilize advanced Machine Learning (ML) algorithms with smoothing functions to predict academic majors completed based on personality subscales.</div></div><div><h3>Methods</h3><div>We used reports from 59,413 individuals to perform the current study. All advanced algorithms implemented in this article were based on R software (version 4.1.3, R Core Team, 2021). All model parameters were optimized based on resampling and cross-validation (CV). In addition, pseudo-R<sup>2</sup> as a robust metric has been used to compare the performance of models, which, unlike most studies, considers the quality of model-predicted probabilities.</div></div><div><h3>Result</h3><div>The results indicated that advanced ML models' performance on training and test data was superior to logistic regression. Pseudo-R<sup>2</sup> and AUC results showed that advanced models such as kNN, GBE, and RF had the highest scores based on test data compared to other models. The pseudo-R<sup>2</sup> values for the models used in this study varied across the test dataset; the lowest value belonged to the logistic regression algorithm at .022, and the highest value was recorded for the kNN algorithm at .099. The agreeableness subscale is the most influential component in predicting the completion of university education, followed by conscientiousness and emotional stability.</div></div><div><h3>Conclusion</h3><div>The potential of advanced methods to enhance the accuracy and validity of predictions is a promising development in our field. Their performance, particularly in handling large data sets with complex patterns, is a reason for optimism about the future of research in this area.</div></div>","PeriodicalId":10578,"journal":{"name":"Computers in biology and medicine","volume":"184 ","pages":"Article 109372"},"PeriodicalIF":7.0,"publicationDate":"2024-11-12","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142615876","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

MCRANet: MTSL-based connectivity region attention network for PD-L1 status segmentation in H&E stained images MCRANet：基于 MTSL 的连接区域注意力网络，用于 H&E 染色图像中的 PD-L1 状态分割。

IF 7 2区医学 Q1 BIOLOGY

Computers in biology and medicine

Pub Date : 2024-11-12 DOI: 10.1016/j.compbiomed.2024.109357

Xixiang Deng , Jiayang Luo , Pan Huang , Peng He , Jiahao Li , Yanan Liu , Hualiang Xiao , Peng Feng

The quantitative analysis of Programmed death-ligand 1 (PD-L1) via Immunohistochemical (IHC) plays a crucial role in guiding immunotherapy. However, IHC faces challenges, including high costs, time consumption and result variability. Conversely, Hematoxylin-Eosin (H&E) staining offers cost-effectiveness, speed, and stable results. Nonetheless, H&E staining, which solely visualizes cellular morphological features, lacks clinical applicability in detecting biomarker expressions like PD-L1. Substituting H&E staining for IHC in determining PD-L1 status is a clinically significant and challenging task. Motivated by above observations, we propose a Multi-Task supervised learning (MTSL)-based connectivity region attention network (MCRANet) for PD-L1 status segmentation in H&E stained images. To reduce interference from non-tumor areas, the MTSL-based region attention is proposed to enhances the network's capability to distinguish between tumor and non-tumor regions. Consequently, this augmentation further improves the network's segmentation efficacy for PD-L1 positive and negative regions. Furthermore, the PD-L1 expression regions demonstrate interconnection throughout the tissue section. Leveraging this topological prior knowledge, we integrate a connectivity modeling module (CM module) within the MTSL-based region attention module (MRA module) to enhance the precision of MTSL-based region attention localization. This integration further improves the structural similarity between the segmentation results and the ground truth. Extensive visual and quantitative results demonstrate that our supervised-learning-guided MRA module produces more interpretable attention and the introduced CM module provides accurate positional attention to the MRA module. Compared to other state-of-the-art networks, MCRANet exhibits superior segmentation performance with a dice similarity coefficient (DSC) of 79.6 % on the lung squamous cell carcinoma (LUSC) PD-L1 status dataset.

通过免疫组织化学（IHC）对程序性死亡配体 1（PD-L1）进行定量分析在指导免疫疗法方面发挥着至关重要的作用。然而，IHC 面临着成本高、耗时长、结果易变等挑战。相反，血红素-伊红（H&E）染色则具有成本效益高、速度快、结果稳定等优点。然而，H&E 染色只能观察细胞形态特征，在检测 PD-L1 等生物标记表达方面缺乏临床适用性。用 IHC 代替 H&E 染色来确定 PD-L1 状态是一项具有临床意义和挑战性的任务。受上述观察结果的启发，我们提出了一种基于多任务监督学习（MTSL）的连接区域注意网络（MCRANet），用于 H&E 染色图像中的 PD-L1 状态分割。为了减少非肿瘤区域的干扰，我们提出了基于 MTSL 的区域注意力，以增强网络区分肿瘤和非肿瘤区域的能力。因此，这一增强功能进一步提高了网络对 PD-L1 阳性和阴性区域的分割效率。此外，PD-L1 表达区域在整个组织切片中显示出相互联系。利用拓扑先验知识，我们在基于 MTSL 的区域关注模块（MRA 模块）中集成了连接建模模块（CM 模块），以提高基于 MTSL 的区域关注定位的精确度。这种整合进一步提高了分割结果与地面实况之间的结构相似性。广泛的视觉和定量结果表明，我们以监督学习为指导的 MRA 模块产生了更多可解释的注意力，而引入的 CM 模块则为 MRA 模块提供了精确的定位注意力。与其他最先进的网络相比，MCRANet 在肺鳞状细胞癌（LUSC）PD-L1 状态数据集上表现出卓越的分割性能，骰子相似系数（DSC）达到 79.6%。

{"title":"MCRANet: MTSL-based connectivity region attention network for PD-L1 status segmentation in H&E stained images","authors":"Xixiang Deng , Jiayang Luo , Pan Huang , Peng He , Jiahao Li , Yanan Liu , Hualiang Xiao , Peng Feng","doi":"10.1016/j.compbiomed.2024.109357","DOIUrl":"10.1016/j.compbiomed.2024.109357","url":null,"abstract":"<div><div>The quantitative analysis of Programmed death-ligand 1 (PD-L1) via Immunohistochemical (IHC) plays a crucial role in guiding immunotherapy. However, IHC faces challenges, including high costs, time consumption and result variability. Conversely, Hematoxylin-Eosin (H&E) staining offers cost-effectiveness, speed, and stable results. Nonetheless, H&E staining, which solely visualizes cellular morphological features, lacks clinical applicability in detecting biomarker expressions like PD-L1. Substituting H&E staining for IHC in determining PD-L1 status is a clinically significant and challenging task. Motivated by above observations, we propose a Multi-Task supervised learning (MTSL)-based connectivity region attention network (MCRANet) for PD-L1 status segmentation in H&E stained images. To reduce interference from non-tumor areas, the MTSL-based region attention is proposed to enhances the network's capability to distinguish between tumor and non-tumor regions. Consequently, this augmentation further improves the network's segmentation efficacy for PD-L1 positive and negative regions. Furthermore, the PD-L1 expression regions demonstrate interconnection throughout the tissue section. Leveraging this topological prior knowledge, we integrate a connectivity modeling module (CM module) within the MTSL-based region attention module (MRA module) to enhance the precision of MTSL-based region attention localization. This integration further improves the structural similarity between the segmentation results and the ground truth. Extensive visual and quantitative results demonstrate that our supervised-learning-guided MRA module produces more interpretable attention and the introduced CM module provides accurate positional attention to the MRA module. Compared to other state-of-the-art networks, MCRANet exhibits superior segmentation performance with a dice similarity coefficient (DSC) of 79.6 % on the lung squamous cell carcinoma (LUSC) PD-L1 status dataset.</div></div>","PeriodicalId":10578,"journal":{"name":"Computers in biology and medicine","volume":"184 ","pages":"Article 109357"},"PeriodicalIF":7.0,"publicationDate":"2024-11-12","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142615232","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

The impact of PTEN status on glioblastoma multiforme: A glial cell type-specific study identifies unique prognostic markers PTEN 状态对多形性胶质母细胞瘤的影响：胶质细胞类型特异性研究确定了独特的预后标志物。

IF 7 2区医学 Q1 BIOLOGY

Computers in biology and medicine

Pub Date : 2024-11-12 DOI: 10.1016/j.compbiomed.2024.109395

Sai Krishna A.V.S. , Swati Sinha , Manchanahalli R. Satyanarayana Rao , Sainitin Donakonda

Glioblastoma multiforme (GBM) is the most invasive form of brain tumor, accounting for 5 % of the cases per 100,000 people in various countries. The phosphatase and tensin homolog deleted from chromosome 10 (PTEN) is a well-known tumor suppressor, and its alteration leads to a deleterious effect on GBM progression. The molecular mechanism of tumorigenesis in glial cell types, driven by PTEN status, is yet to be elucidated. In this study, we analyzed publicly available single-cell transcriptome profiles of PTEN wild-type (WT) and NULL GBM patients. We compared them with normal brain data to uncover many unique gene sets influenced by PTEN status. The co-expression network analysis of differentially expressed genes (DEGs) between normal brain and PTEN (WT and NULL) identified highly interconnected genes. The weighted gene co-expression network analysis (WGCNA), based on the DESeq2 algorithm, identified glial cell-type-specific modules in PTEN status-dependent bulk RNA expression profiles. We overlapped network module gene sets from single-cell and bulk transcriptome profiles, and shared genes were considered for further analysis. The hallmark pathway enrichment analysis of the genes unique to PTEN-WT and NULL revealed various tumor growth-related pathways across the glial cell types. Further characterization of PTEN-WT and PTEN-NULL networks belonging to the single-cell and bulk RNA datasets revealed that PTEN status influences the network modules in astrocytes, microglia, and oligodendrocyte precursor cells. An integrated influence value algorithm identified hub genes for each glial cell type. The prognostic analysis identified clinically relevant hub genes specific to the cell type in PTEN-WT: GLIPR2 (astrocytes), CFH, IL32, MXRA5 (microglia), and PTEN-NULL: ID1 (astrocytes) and LAT2 (microglia). Our glial cell type-level transcriptome analysis unearthed unique molecular pathways and prognostic markers in PTEN status-dependent GBM patients.

多形性胶质母细胞瘤（GBM）是侵袭性最强的脑肿瘤，在各国每 10 万人中占 5%。从 10 号染色体上删除的磷酸酶和天丝同源物（PTEN）是一种众所周知的肿瘤抑制因子，它的改变会对 GBM 的进展产生有害影响。PTEN状态驱动胶质细胞类型肿瘤发生的分子机制尚待阐明。在本研究中，我们分析了可公开获得的 PTEN 野生型（WT）和 NULL GBM 患者的单细胞转录组图谱。我们将它们与正常大脑数据进行了比较，发现了许多受 PTEN 状态影响的独特基因组。对正常大脑与 PTEN（WT 和 NULL）之间差异表达基因（DEGs）的共表达网络分析发现了高度相互关联的基因。基于 DESeq2 算法的加权基因共表达网络分析（WGCNA）在 PTEN 状态依赖的大量 RNA 表达谱中发现了胶质细胞类型特异性模块。我们将单细胞和大量转录组图谱中的网络模块基因组重叠，并考虑对共享基因进行进一步分析。对PTEN-WT和NULL特有基因的标志性通路富集分析显示了神经胶质细胞类型中与肿瘤生长相关的各种通路。对属于单细胞和大量RNA数据集的PTEN-WT和PTEN-NULL网络的进一步表征显示，PTEN状态影响了星形胶质细胞、小胶质细胞和少突胶质细胞前体细胞的网络模块。综合影响值算法确定了每种神经胶质细胞类型的枢纽基因。预后分析确定了 PTEN-WT 细胞类型特有的临床相关中心基因：GLIPR2（星形胶质细胞）、CFH、IL32、MXRA5（小胶质细胞），以及 PTEN-NULL：ID1（星形胶质细胞）和 LAT2（小胶质细胞）。我们的神经胶质细胞类型水平转录组分析发现了 PTEN 状态依赖型 GBM 患者的独特分子通路和预后标志物。

{"title":"The impact of PTEN status on glioblastoma multiforme: A glial cell type-specific study identifies unique prognostic markers","authors":"Sai Krishna A.V.S. , Swati Sinha , Manchanahalli R. Satyanarayana Rao , Sainitin Donakonda","doi":"10.1016/j.compbiomed.2024.109395","DOIUrl":"10.1016/j.compbiomed.2024.109395","url":null,"abstract":"<div><div>Glioblastoma multiforme (GBM) is the most invasive form of brain tumor, accounting for 5 % of the cases per 100,000 people in various countries. The phosphatase and tensin homolog deleted from chromosome 10 (PTEN) is a well-known tumor suppressor, and its alteration leads to a deleterious effect on GBM progression. The molecular mechanism of tumorigenesis in glial cell types, driven by PTEN status, is yet to be elucidated. In this study, we analyzed publicly available single-cell transcriptome profiles of PTEN wild-type (WT) and NULL GBM patients. We compared them with normal brain data to uncover many unique gene sets influenced by PTEN status. The co-expression network analysis of differentially expressed genes (DEGs) between normal brain and PTEN (WT and NULL) identified highly interconnected genes. The weighted gene co-expression network analysis (WGCNA), based on the DESeq2 algorithm, identified glial cell-type-specific modules in PTEN status-dependent bulk RNA expression profiles. We overlapped network module gene sets from single-cell and bulk transcriptome profiles, and shared genes were considered for further analysis. The hallmark pathway enrichment analysis of the genes unique to PTEN-WT and NULL revealed various tumor growth-related pathways across the glial cell types. Further characterization of PTEN-WT and PTEN-NULL networks belonging to the single-cell and bulk RNA datasets revealed that PTEN status influences the network modules in astrocytes, microglia, and oligodendrocyte precursor cells. An integrated influence value algorithm identified hub genes for each glial cell type. The prognostic analysis identified clinically relevant hub genes specific to the cell type in PTEN-WT: <em>GLIPR2</em> (astrocytes), <em>CFH</em>, <em>IL32</em>, <em>MXRA5</em> (microglia), and PTEN-NULL: <em>ID1</em> (astrocytes) and <em>LAT2</em> (microglia). Our glial cell type-level transcriptome analysis unearthed unique molecular pathways and prognostic markers in PTEN status-dependent GBM patients.</div></div>","PeriodicalId":10578,"journal":{"name":"Computers in biology and medicine","volume":"184 ","pages":"Article 109395"},"PeriodicalIF":7.0,"publicationDate":"2024-11-12","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142615780","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

SARS-CoV-2 main protease (M-pro) mutational profiling: An insight into mutation coldspots SARS-CoV-2 主要蛋白酶（M-pro）突变分析：洞察突变冷门

IF 7 2区医学 Q1 BIOLOGY

Computers in biology and medicine

Pub Date : 2024-11-12 DOI: 10.1016/j.compbiomed.2024.109344

Pol Garcia-Segura, Ariadna Llop-Peiró, Nil Novau-Ferré, Júlia Mestres-Truyol, Bryan Saldivar-Espinoza, Gerard Pujadas, Santiago Garcia-Vallvé

SARS-CoV-2 and the COVID-19 pandemic have marked a milestone in the history of scientific research worldwide. To ensure that treatments are successful in the mid-long term, it is crucial to characterize SARS-CoV-2 mutations, as they might lead to viral resistance. Data from >5,700,000 SARS-CoV-2 genomes available at GISAID was used to report SARS-CoV-2 mutations. Given the pivotal role of its main protease (M-pro) in virus replication, a detailed analysis of SARS-CoV-2 M-pro mutations was conducted, with particular attention to mutation-resistant residues or mutation coldspots, defined as those residues that have mutated in five or fewer genomes. 32 mutation coldspots were identified, most of which mediate interprotomer interactions or funneling interaction networks from the substrate-binding site towards the dimerization surface and vice versa. Besides, mutation coldspots were virtually conserved in all main proteases from other CoVs. Our results provide valuable information about key residues to M-pro structure that could be useful in rational target-directed drug design and establish a solid groundwork based on mutation analyses for the inhibition of M-pro dimerization, with a potential applicability to future coronavirus outbreaks.

SARS-CoV-2 和 COVID-19 大流行标志着世界科学研究史上的一个里程碑。为了确保治疗在中长期内取得成功，确定 SARS-CoV-2 基因突变的特征至关重要，因为这些突变可能导致病毒耐药性。我们利用 GISAID 现有的超过 5,700,000 个 SARS-CoV-2 基因组的数据来报告 SARS-CoV-2 基因突变。鉴于其主要蛋白酶（M-pro）在病毒复制中的关键作用，我们对 SARS-CoV-2 M-pro 变异进行了详细分析，并特别关注抗突变残基或突变冷点，即在五个或更少的基因组中发生突变的残基。共发现了 32 个突变冷点，其中大部分介导了原体间的相互作用或从底物结合位点到二聚化表面的漏斗状相互作用网络，反之亦然。此外，突变冷点在其他 CoVs 的所有主要蛋白酶中几乎都是保守的。我们的研究结果提供了有关M-pro结构关键残基的宝贵信息，这些信息可能有助于合理的靶向药物设计，并为基于突变分析的M-pro二聚化抑制奠定了坚实的基础，有可能适用于未来的冠状病毒爆发。

{"title":"SARS-CoV-2 main protease (M-pro) mutational profiling: An insight into mutation coldspots","authors":"Pol Garcia-Segura, Ariadna Llop-Peiró, Nil Novau-Ferré, Júlia Mestres-Truyol, Bryan Saldivar-Espinoza, Gerard Pujadas, Santiago Garcia-Vallvé","doi":"10.1016/j.compbiomed.2024.109344","DOIUrl":"10.1016/j.compbiomed.2024.109344","url":null,"abstract":"<div><div>SARS-CoV-2 and the COVID-19 pandemic have marked a milestone in the history of scientific research worldwide. To ensure that treatments are successful in the mid-long term, it is crucial to characterize SARS-CoV-2 mutations, as they might lead to viral resistance. Data from >5,700,000 SARS-CoV-2 genomes available at GISAID was used to report SARS-CoV-2 mutations. Given the pivotal role of its main protease (M-pro) in virus replication, a detailed analysis of SARS-CoV-2 M-pro mutations was conducted, with particular attention to mutation-resistant residues or mutation coldspots, defined as those residues that have mutated in five or fewer genomes. 32 mutation coldspots were identified, most of which mediate interprotomer interactions or funneling interaction networks from the substrate-binding site towards the dimerization surface and <em>vice versa</em>. Besides, mutation coldspots were virtually conserved in all main proteases from other CoVs. Our results provide valuable information about key residues to M-pro structure that could be useful in rational target-directed drug design and establish a solid groundwork based on mutation analyses for the inhibition of M-pro dimerization, with a potential applicability to future coronavirus outbreaks.</div></div>","PeriodicalId":10578,"journal":{"name":"Computers in biology and medicine","volume":"184 ","pages":"Article 109344"},"PeriodicalIF":7.0,"publicationDate":"2024-11-12","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142615502","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0