Pub Date : 2025-05-15DOI: 10.1016/j.ymeth.2025.05.005
Ming Chen , Haike Li , Yunhan Pan , Yinglong Dai , Xiujuan Lei , Yi Pan
The prediction of mechanisms within drug-target interactions (DTIs) can boost the drug discovery process, which has traditionally relied on time-consuming and expensive laboratory experiments. Despite much more attention has been paid to predicting DTIs, but few studies focused on their activating/inhibiting mechanisms. In this work, we model DTIs on signed heterogeneous networks, through categorizing activating/inhibiting DTIs into signed links, and accordingly introducing the coherence/incoherence between drugs on a common target to construct signed drug-drug links. We propose a multi-filter based signed heterogeneous graph convolutional network (MFSHGCN) for drugs and targets embedding, via employing dual filters on both the signed drug-drug sub-graph and the signed DTI sub-graph to converge the spectral information from positive and negative edges. We further put forward an end-to-end framework to predict activation and inhibition within DTIs. The comparison results demonstrate the introduction of coherence/incoherence of drug pairs and the design of our multi-filter system can effectively improve the prediction metrics, even without relying on rich node information and interactions from drug pairs or target pairs. Case studies on breast cancer and lung cancer confirm the model's feasibility.
{"title":"Multi-filter based signed heterogeneous graph convolutional networks for predicting activating/inhibiting drug-target interactions","authors":"Ming Chen , Haike Li , Yunhan Pan , Yinglong Dai , Xiujuan Lei , Yi Pan","doi":"10.1016/j.ymeth.2025.05.005","DOIUrl":"10.1016/j.ymeth.2025.05.005","url":null,"abstract":"<div><div>The prediction of mechanisms within drug-target interactions (DTIs) can boost the drug discovery process, which has traditionally relied on time-consuming and expensive laboratory experiments. Despite much more attention has been paid to predicting DTIs, but few studies focused on their activating/inhibiting mechanisms. In this work, we model DTIs on signed heterogeneous networks, through categorizing activating/inhibiting DTIs into signed links, and accordingly introducing the coherence/incoherence between drugs on a common target to construct signed drug-drug links. We propose a multi-filter based signed heterogeneous graph convolutional network (MFSHGCN) for drugs and targets embedding, via employing dual filters on both the signed drug-drug sub-graph and the signed DTI sub-graph to converge the spectral information from positive and negative edges. We further put forward an end-to-end framework to predict activation and inhibition within DTIs. The comparison results demonstrate the introduction of coherence/incoherence of drug pairs and the design of our multi-filter system can effectively improve the prediction metrics, even without relying on rich node information and interactions from drug pairs or target pairs. Case studies on breast cancer and lung cancer confirm the model's feasibility.</div></div>","PeriodicalId":390,"journal":{"name":"Methods","volume":"241 ","pages":"Pages 51-58"},"PeriodicalIF":4.2,"publicationDate":"2025-05-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144092458","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"生物学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2025-05-13DOI: 10.1016/j.ymeth.2025.05.004
Bert Vandendriessche , Jolien Huyghebaert , Kirsten Van Rossem , Tycho Canter Cremers , Kevin De Man , Ewa Sieliwonczyk , Hanne Boen , Dogan Akdeniz , Laura Rabaut , Jolien Schippers , Peter Ponsaerts , R. Frank Kooy , Bart Loeys , Dorien Schepers , Maaike Alaerts
Precise gene editing with conventional CRISPR/Cas9 is often constrained by low knock-in (KI) efficiencies (≈ 2–20 %) in human induced pluripotent stem cells (hiPSCs) and human embryonic stem cells (hESCs). This limitation typically necessitates labour-intensive manual isolation and genotyping of hundreds of colonies to identify correctly edited cells. Fluorescence- or antibiotic-based enrichment methods facilitate the identification process but can compromise cell viability and genomic integrity. Here, we present a footprint-free editing strategy that combines low-density seeding with next-generation sequencing (NGS) to rapidly identify cell populations containing precisely modified clones. By optimising the transfection workflow and adhering to CRISPR/Cas9 KI design principles, we achieved high average editing efficiencies of 64 % in hiPSCs (introducing a Brugada syndrome-associated variant) and 51 % in hESCs (introducing a neurodevelopmental disorder (NDD)-associated variant). Furthermore, under suboptimal CRISPR design conditions, this approach successfully identified hESC clones carrying a second NDD-associated variant, despite average KI efficiencies below 1 %. Importantly, genomic integrity was preserved throughout subcloning rounds, as confirmed by Sanger sequencing and single nucleotide polymorphism (SNP) array analysis. Hence, this NGS-based enrichment strategy reliably identifies desired KI clones under both optimal and challenging conditions, reducing the need for extensive colony screening and offering an effective alternative to fluorescence- and antibiotic-based selection methods.
{"title":"An NGS-based approach for precise and footprint-free CRISPR-based gene editing in human stem cells","authors":"Bert Vandendriessche , Jolien Huyghebaert , Kirsten Van Rossem , Tycho Canter Cremers , Kevin De Man , Ewa Sieliwonczyk , Hanne Boen , Dogan Akdeniz , Laura Rabaut , Jolien Schippers , Peter Ponsaerts , R. Frank Kooy , Bart Loeys , Dorien Schepers , Maaike Alaerts","doi":"10.1016/j.ymeth.2025.05.004","DOIUrl":"10.1016/j.ymeth.2025.05.004","url":null,"abstract":"<div><div>Precise gene editing with conventional CRISPR/Cas9 is often constrained by low <em>knock-in</em> (KI) efficiencies (≈ 2–20 %) in human induced pluripotent stem cells (hiPSCs) and human embryonic stem cells (hESCs). This limitation typically necessitates labour-intensive manual isolation and genotyping of hundreds of colonies to identify correctly edited cells. Fluorescence- or antibiotic-based enrichment methods facilitate the identification process but can compromise cell viability and genomic integrity. Here, we present a footprint-free editing strategy that combines low-density seeding with next-generation sequencing (NGS) to rapidly identify cell populations containing precisely modified clones. By optimising the transfection workflow and adhering to CRISPR/Cas9 KI design principles, we achieved high average editing efficiencies of 64 % in hiPSCs (introducing a Brugada syndrome-associated variant) and 51 % in hESCs (introducing a neurodevelopmental disorder (NDD)-associated variant). Furthermore, under suboptimal CRISPR design conditions, this approach successfully identified hESC clones carrying a second NDD-associated variant, despite average KI efficiencies below 1 %. Importantly, genomic integrity was preserved throughout subcloning rounds, as confirmed by Sanger sequencing and single nucleotide polymorphism (SNP) array analysis. Hence, this NGS-based enrichment strategy reliably identifies desired KI clones under both optimal and challenging conditions, reducing the need for extensive colony screening and offering an effective alternative to fluorescence- and antibiotic-based selection methods.</div></div>","PeriodicalId":390,"journal":{"name":"Methods","volume":"241 ","pages":"Pages 33-42"},"PeriodicalIF":4.2,"publicationDate":"2025-05-13","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143947684","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"生物学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2025-05-09DOI: 10.1016/j.ymeth.2025.05.003
Peifu Han , Jianmin Wang , Dayan Liu , Lin Liu , Tao Song
Static graphs play a pivotal role in modeling and analyzing biological and biomedical data. However, many real-world scenarios—such as disease progression and drug pharmacokinetic processes—exhibit dynamic behaviors. Consequently, static graph methods often struggle to robustly address new environments characterized by complex and previously unseen relationship changes. Here, we propose a method for constructing temporal knowledge inference agents tailored to disease pathways, enabling effective relation reasoning beyond their training environment under complex shifts. To achieve this, we developed an imitation learning framework using liquid neural networks, a class of continuous-time neural models inspired by the brain function that are causal and adaptable to changing conditions. Our findings indicate that liquid agents can distill the essential tasks from knowledge graph inputs while accounting temporal evolution, thereby enabling the transfer of temporal skills to novel time nodes. Compared to state-of-the-art deep reinforcement learning agents, experiments demonstrate that temporal robustness in decision-making emerges uniquely in liquid networks.
{"title":"Robust temporal knowledge inference via pathway snapshots with liquid neural network","authors":"Peifu Han , Jianmin Wang , Dayan Liu , Lin Liu , Tao Song","doi":"10.1016/j.ymeth.2025.05.003","DOIUrl":"10.1016/j.ymeth.2025.05.003","url":null,"abstract":"<div><div>Static graphs play a pivotal role in modeling and analyzing biological and biomedical data. However, many real-world scenarios—such as disease progression and drug pharmacokinetic processes—exhibit dynamic behaviors. Consequently, static graph methods often struggle to robustly address new environments characterized by complex and previously unseen relationship changes. Here, we propose a method for constructing temporal knowledge inference agents tailored to disease pathways, enabling effective relation reasoning beyond their training environment under complex shifts. To achieve this, we developed an imitation learning framework using liquid neural networks, a class of continuous-time neural models inspired by the brain function that are causal and adaptable to changing conditions. Our findings indicate that liquid agents can distill the essential tasks from knowledge graph inputs while accounting temporal evolution, thereby enabling the transfer of temporal skills to novel time nodes. Compared to state-of-the-art deep reinforcement learning agents, experiments demonstrate that temporal robustness in decision-making emerges uniquely in liquid networks.</div></div>","PeriodicalId":390,"journal":{"name":"Methods","volume":"241 ","pages":"Pages 24-32"},"PeriodicalIF":4.2,"publicationDate":"2025-05-09","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143942678","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"生物学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2025-05-08DOI: 10.1016/j.ymeth.2025.05.002
Ricardo Jose Branco Leote , Caroline G. Sanz , Victor C. Diculescu , Madalina Maria Barsan
Overexpression of pyruvate kinase (PyK) is linked to many kinds of malignant tumors, representing therefore one of the most promising therapeutic targets for cancer treatment. Inhibition of PyK slows down tumor growth or causes tumor cell death, minimizing cancer cell proliferation, and understanding inhibitor mechanism of action can significantly improve cancer therapy. The present work describes the use of an amperometric bienzymatic biosensor, based on PyK and pyruvate oxidase (PyOx), in enzyme inhibition studies of four kinase inhibitors, CPG77675, Nilotinib, Ruxolitinib, Cerdulatinib. Their inhibition mechanism is studied and discussed in detail, with a thorough evaluation of their enzyme-inhibitor complex binding constants (Ki) and the inhibitor concentration required for 50% inhibition (IC50), employing standard inhibition procedure graphical methods. The biosensor is successfully applied for the quantification of the inhibitors by fixed potential amperometry, with excellent detection limit values in the pM range. It is the first detection method reported for the anticancer drugs CPG77675 and Cerdulatinib. The electrochemical assay based on the biosensor brings several advantages over the available assay kits for high-throughput screening (HTS) of kinase inhibitors, namely: low cost, easy operability and robustness demonstrated by biosensor high reproducibility and both operational and storage stability, offering an opportunity to discover new inhibitors and optimize their therapeutic index.
{"title":"Electrochemical assay for the quantification of anticancer drugs and their inhibition mechanism","authors":"Ricardo Jose Branco Leote , Caroline G. Sanz , Victor C. Diculescu , Madalina Maria Barsan","doi":"10.1016/j.ymeth.2025.05.002","DOIUrl":"10.1016/j.ymeth.2025.05.002","url":null,"abstract":"<div><div>Overexpression of pyruvate kinase (PyK) is linked to many kinds of malignant tumors, representing therefore one of the most promising therapeutic targets for cancer treatment. Inhibition of PyK slows down tumor growth or causes tumor cell death, minimizing cancer cell proliferation, and understanding inhibitor mechanism of action can significantly improve cancer therapy. The present work describes the use of an amperometric bienzymatic biosensor, based on PyK and pyruvate oxidase (PyOx), in enzyme inhibition studies of four kinase inhibitors, CPG77675, Nilotinib, Ruxolitinib, Cerdulatinib. Their inhibition mechanism is studied and discussed in detail, with a thorough evaluation of their enzyme-inhibitor complex binding constants (<em>K<sub>i</sub></em>) and the inhibitor concentration required for 50% inhibition <em>(IC<sub>50</sub></em>), employing standard inhibition procedure graphical methods. The biosensor is successfully applied for the quantification of the inhibitors by fixed potential amperometry, with excellent detection limit values in the pM range. It is the first detection method reported for the anticancer drugs CPG77675 and Cerdulatinib. The electrochemical assay based on the biosensor brings several advantages over the available assay kits for high-throughput screening (HTS) of kinase inhibitors, namely: low cost, easy operability and robustness demonstrated by biosensor high reproducibility and both operational and storage stability, offering an opportunity to discover new inhibitors and optimize their therapeutic index.</div></div>","PeriodicalId":390,"journal":{"name":"Methods","volume":"241 ","pages":"Pages 13-23"},"PeriodicalIF":4.2,"publicationDate":"2025-05-08","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143942677","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"生物学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2025-05-03DOI: 10.1016/j.ymeth.2025.05.001
Isabel Quintanilla , Benura Azeroglu , Md Abdul Kader Sagar , Travis H. Stracker , Eros Lazzerini Denchi , Gianluca Pegoraro
Telomere elongation is essential for the proliferation of cancer cells. Telomere length control is achieved either by the activation of the telomerase enzyme, or by the recombination-based Alternative Lengthening of Telomeres (ALT) pathway. ALT is active in about 10–15% of human cancers, but its molecular underpinnings remain poorly understood, preventing the discovery of potential novel therapeutic targets. Pooled CRISPR-based functional genomic screens enable the unbiased discovery of molecular factors involved in cancer biology. Recently, Optical Pooled Screens (OPS) have significantly extended the capabilities of pooled functional genomics screens to enable sensitive imaging-based readouts at the single cell level and large scale. To gain a better understanding of the ALT pathway, we developed a novel OPS assay that employs telomeric native DNA FISH (nFISH) as an optical quantitative readout to measure ALT activity. The assay uses standard OPS protocols for library preparation and sequencing. As a critical element, an optimized nFISH protocol is performed before in situ sequencing to maximize the assay performance. We show that the modified nFISH protocol faithfully detects changes in ALT activity upon CRISPR knock-out (KO) of the FANCM and BLM genes, which were previously implicated in ALT. Overall, the OPS-nFISH assay is a reliable method that can provide deep insights into the ALT pathway in a high-throughput format.
端粒延长是癌细胞增殖的必要条件。端粒长度控制是通过端粒酶的激活或基于重组的端粒选择性延长（ALT）途径实现的。ALT在大约10-15%的人类癌症中有活性，但其分子基础仍然知之甚少，阻碍了潜在的新治疗靶点的发现。汇集基于crispr的功能基因组筛选使参与癌症生物学的分子因素的公正发现成为可能。最近，光学池屏幕（OPS）显著扩展了池功能基因组学屏幕的能力，使单细胞水平和大规模的基于成像的敏感读数成为可能。为了更好地了解ALT途径，我们开发了一种新的OPS检测方法，该方法使用端粒原生DNA FISH （nFISH）作为光学定量读数来测量ALT活性。该分析使用标准OPS方案进行文库制备和测序。作为一个关键因素，优化的nFISH方案在原位测序之前执行，以最大限度地提高分析性能。我们发现，修改后的nFISH方案忠实地检测到FANCM和BLM基因的CRISPR敲除（KO）时ALT活性的变化，这些基因先前与ALT有关。总的来说，OPS-nFISH试验是一种可靠的方法，可以以高通量的形式深入了解ALT途径。
{"title":"Optical pooled screening for the discovery of regulators of the alternative lengthening of telomeres pathway","authors":"Isabel Quintanilla , Benura Azeroglu , Md Abdul Kader Sagar , Travis H. Stracker , Eros Lazzerini Denchi , Gianluca Pegoraro","doi":"10.1016/j.ymeth.2025.05.001","DOIUrl":"10.1016/j.ymeth.2025.05.001","url":null,"abstract":"<div><div>Telomere elongation is essential for the proliferation of cancer cells. Telomere length control is achieved either by the activation of the telomerase enzyme, or by the recombination-based Alternative Lengthening of Telomeres (ALT) pathway. ALT is active in about 10–15% of human cancers, but its molecular underpinnings remain poorly understood, preventing the discovery of potential novel therapeutic targets. Pooled CRISPR-based functional genomic screens enable the unbiased discovery of molecular factors involved in cancer biology. Recently, Optical Pooled Screens (OPS) have significantly extended the capabilities of pooled functional genomics screens to enable sensitive imaging-based readouts at the single cell level and large scale. To gain a better understanding of the ALT pathway, we developed a novel OPS assay that employs telomeric native DNA FISH (nFISH) as an optical quantitative readout to measure ALT activity. The assay uses standard OPS protocols for library preparation and sequencing. As a critical element, an optimized nFISH protocol is performed before in situ sequencing to maximize the assay performance. We show that the modified nFISH protocol faithfully detects changes in ALT activity upon CRISPR knock-out (KO) of the <em>FANCM</em> and <em>BLM</em> genes, which were previously implicated in ALT. Overall, the OPS-nFISH assay is a reliable method that can provide deep insights into the ALT pathway in a high-throughput format.</div></div>","PeriodicalId":390,"journal":{"name":"Methods","volume":"241 ","pages":"Pages 1-12"},"PeriodicalIF":4.2,"publicationDate":"2025-05-03","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143923043","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"生物学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2025-04-29DOI: 10.1016/j.ymeth.2025.04.018
Jacob Tizhe Liberty , Sabri Bromage , Endurance Peter , Olivia C. Ihedioha , Fatemah B. Alsalman , Tochukwu Samuel Odogwu
Foodborne pathogens represent a significant challenge to global food safety, causing widespread illnesses and economic losses. The growing complexity of food supply chains and the emergence of antimicrobial resistance necessitate rapid, sensitive, and portable diagnostic tools. CRISPR technology has emerged as a transformative solution, offering unparalleled precision and adaptability in pathogen detection. This review explores CRISPR’s role in addressing critical gaps in traditional and modern diagnostic methods, emphasizing its advantages in sensitivity, specificity, and scalability. CRISPR-based diagnostics, such as Cas12 and Cas13 systems, enable rapid detection of bacterial and viral pathogens, as well as toxins and chemical hazards, directly in food matrices. Their integration with isothermal amplification techniques and portable biosensors enhances field applicability, making them ideal for decentralized and real-time testing. Additionally, CRISPR’s potential extends beyond food safety, contributing to public health efforts by monitoring antimicrobial resistance and supporting One Health frameworks. Despite these advancements, challenges remain, including issues with performance in complex food matrices, scalability, and regulatory barriers. This review highlights future directions, including AI integration for assay optimization, the development of universal CRISPR platforms, and the adoption of sustainable diagnostic solutions. By tackling these challenges, CRISPR has the potential to redefine global food safety standards and create a more resilient food system. Collaborative research and innovation will be critical to fully unlocking its transformative potential in food safety and public health.
{"title":"CRISPR revolution: Unleashing precision pathogen detection to safeguard public health and food safety","authors":"Jacob Tizhe Liberty , Sabri Bromage , Endurance Peter , Olivia C. Ihedioha , Fatemah B. Alsalman , Tochukwu Samuel Odogwu","doi":"10.1016/j.ymeth.2025.04.018","DOIUrl":"10.1016/j.ymeth.2025.04.018","url":null,"abstract":"<div><div>Foodborne pathogens represent a significant challenge to global food safety, causing widespread illnesses and economic losses. The growing complexity of food supply chains and the emergence of antimicrobial resistance necessitate rapid, sensitive, and portable diagnostic tools. CRISPR technology has emerged as a transformative solution, offering unparalleled precision and adaptability in pathogen detection. This review explores CRISPR’s role in addressing critical gaps in traditional and modern diagnostic methods, emphasizing its advantages in sensitivity, specificity, and scalability. CRISPR-based diagnostics, such as Cas12 and Cas13 systems, enable rapid detection of bacterial and viral pathogens, as well as toxins and chemical hazards, directly in food matrices. Their integration with isothermal amplification techniques and portable biosensors enhances field applicability, making them ideal for decentralized and real-time testing. Additionally, CRISPR’s potential extends beyond food safety, contributing to public health efforts by monitoring antimicrobial resistance and supporting One Health frameworks. Despite these advancements, challenges remain, including issues with performance in complex food matrices, scalability, and regulatory barriers. This review highlights future directions, including AI integration for assay optimization, the development of universal CRISPR platforms, and the adoption of sustainable diagnostic solutions. By tackling these challenges, CRISPR has the potential to redefine global food safety standards and create a more resilient food system. Collaborative research and innovation will be critical to fully unlocking its transformative potential in food safety and public health.</div></div>","PeriodicalId":390,"journal":{"name":"Methods","volume":"240 ","pages":"Pages 180-194"},"PeriodicalIF":4.2,"publicationDate":"2025-04-29","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143907880","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"生物学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2025-04-28DOI: 10.1016/j.ymeth.2025.04.016
Asim Zaman , Mazen M. Yassin , Irfan Mehmud , Anbo Cao , Jiaxi Lu , Haseeb Hassan , Yan Kang
Brain lesion segmentation is challenging in medical image analysis, aiming to delineate lesion regions precisely. Deep learning (DL) techniques have recently demonstrated promising results across various computer vision tasks, including semantic segmentation, object detection, and image classification. This paper offers an overview of recent DL algorithms for brain tumor and stroke segmentation, drawing on literature from 2021 to 2024. It highlights the strengths, limitations, current research challenges, and unexplored areas in imaging-based brain lesion classification based on insights from over 250 recent review papers. Techniques addressing difficulties like class imbalance and multi-modalities are presented. Optimization methods for improving performance regarding computational and structural complexity and processing speed are discussed. These include lightweight neural networks, multilayer architectures, and computationally efficient, highly accurate network designs. The paper also reviews generic and latest frameworks of different brain lesion detection techniques and highlights publicly available benchmark datasets and their issues. Furthermore, open research areas, application prospects, and future directions for DL-based brain lesion classification are discussed. Future directions include integrating neural architecture search methods with domain knowledge, predicting patient survival levels, and learning to separate brain lesions using patient statistics. To ensure patient privacy, future research is anticipated to explore privacy-preserving learning frameworks. Overall, the presented suggestions serve as a guideline for researchers and system designers involved in brain lesion detection and stroke segmentation tasks.
{"title":"Challenges, optimization strategies, and future horizons of advanced deep learning approaches for brain lesion segmentation","authors":"Asim Zaman , Mazen M. Yassin , Irfan Mehmud , Anbo Cao , Jiaxi Lu , Haseeb Hassan , Yan Kang","doi":"10.1016/j.ymeth.2025.04.016","DOIUrl":"10.1016/j.ymeth.2025.04.016","url":null,"abstract":"<div><div>Brain lesion segmentation is challenging in medical image analysis, aiming to delineate lesion regions precisely. Deep learning (DL) techniques have recently demonstrated promising results across various computer vision tasks, including semantic segmentation, object detection, and image classification. This paper offers an overview of recent DL algorithms for brain tumor and stroke segmentation, drawing on literature from 2021 to 2024. It highlights the strengths, limitations, current research challenges, and unexplored areas in imaging-based brain lesion classification based on insights from over 250 recent review papers. Techniques addressing difficulties like class imbalance and multi-modalities are presented. Optimization methods for improving performance regarding computational and structural complexity and processing speed are discussed. These include lightweight neural networks, multilayer architectures, and computationally efficient, highly accurate network designs. The paper also reviews generic and latest frameworks of different brain lesion detection techniques and highlights publicly available benchmark datasets and their issues. Furthermore, open research areas, application prospects, and future directions for DL-based brain lesion classification are discussed. Future directions include integrating neural architecture search methods with domain knowledge, predicting patient survival levels, and learning to separate brain lesions using patient statistics. To ensure patient privacy, future research is anticipated to explore privacy-preserving learning frameworks. Overall, the presented suggestions serve as a guideline for researchers and system designers involved in brain lesion detection and stroke segmentation tasks.</div></div>","PeriodicalId":390,"journal":{"name":"Methods","volume":"239 ","pages":"Pages 140-168"},"PeriodicalIF":4.2,"publicationDate":"2025-04-28","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143900110","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"生物学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2025-04-28DOI: 10.1016/j.ymeth.2025.04.017
Dhruba Jyoti Sarkar , Ramij Raja , V. Santhana Kumar , Soumyadeb Bhattacharyya , Souvik Pal , Subhankar Mukherjee , Basanta Kumar Das
Tetracycline is extensively used in aquaculture as a therapeutic agent that needs to be monitored due to food safety concerns. Aptasensing has been revealed as a suitable diagnostic platform for tetracycline sensing in food matrix due to its quick, low cost and robust nature. But, the colorimetric aptasensing of tetracycline employing the peroxidase activity of gold nanoparticles (AuNPs) to 3,3,5,5-tetramethylbenzidine (TMB) was unsuitable until now owing to the aptamer-specific alkaline binding buffer. The present study developed a method with an optimized reaction protocol diminishing the inhibitory effect of binding buffer on the sensor probe (AuNPs-aptamer + TMB + H2O2). The overall peroxidase activity of the sensor probe was only inhibited by tetracycline through selective adsorption on the AuNPs-aptamer complex. The peroxidase inhibition percentage in the test range of 0.01 to 0.5 mg L-1 tetracycline gave a logarithmic response (R2, 0.99) with a detection limit of 0.017 mg L-1 which is less than the prescribed limit (0.1 mg L-1) set by EU and FSSAI. The developed sensing system in fish muscle showed high recovery (111–115 %) with great potential for rapid detection of tetracycline in fish muscle.
{"title":"Breaking barrier of binding buffer in colorimetric aptasensing of tetracycline in food fish using peroxidase mimic gold NanoZyme","authors":"Dhruba Jyoti Sarkar , Ramij Raja , V. Santhana Kumar , Soumyadeb Bhattacharyya , Souvik Pal , Subhankar Mukherjee , Basanta Kumar Das","doi":"10.1016/j.ymeth.2025.04.017","DOIUrl":"10.1016/j.ymeth.2025.04.017","url":null,"abstract":"<div><div>Tetracycline is extensively used in aquaculture as a therapeutic agent that needs to be monitored due to food safety concerns. Aptasensing has been revealed as a suitable diagnostic platform for tetracycline sensing in food matrix due to its quick, low cost and robust nature. But, the colorimetric aptasensing of tetracycline employing the peroxidase activity of gold nanoparticles (AuNPs) to 3,3,5,5-tetramethylbenzidine (TMB) was unsuitable until now owing to the aptamer-specific alkaline binding buffer. The present study developed a method with an optimized reaction protocol diminishing the inhibitory effect of binding buffer on the sensor probe (AuNPs-aptamer + TMB + H<sub>2</sub>O<sub>2</sub>). The overall peroxidase activity of the sensor probe was only inhibited by tetracycline through selective adsorption on the AuNPs-aptamer complex. The peroxidase inhibition percentage in the test range of 0.01 to 0.5 mg L<sup>-1</sup> tetracycline gave a logarithmic response (R<sup>2</sup>, 0.99) with a detection limit of 0.017 mg L<sup>-1</sup> which is less than the prescribed limit (0.1 mg L<sup>-1</sup>) set by EU and FSSAI. The developed sensing system in fish muscle showed high recovery (111–115 %) with great potential for rapid detection of tetracycline in fish muscle.</div></div>","PeriodicalId":390,"journal":{"name":"Methods","volume":"240 ","pages":"Pages 145-153"},"PeriodicalIF":4.2,"publicationDate":"2025-04-28","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143891850","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"生物学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2025-04-25DOI: 10.1016/j.ymeth.2025.04.012
Di Yu , Xinyu Yang , Yifan Shang , Sisi Yuan , Yuansheng Liu , Yiping Liu
Drug-target interaction (DTI) prediction is critical in drug repositioning and discovery. In current metapath-based prediction methods, attention mechanisms are often used to differentiate the importance of various neighbors, enhancing the model's expressiveness. However, in biological networks with small-scale imbalanced data, attention mechanisms are prone to interference from noise and missing data, leading to instability in weight learning, reduced efficiency, and an increased risk of overfitting. To address these issues, we propose the use of average aggregation to mitigate noise, simplify model complexity, and improve stability. Specifically, we introduce a simplified mean aggregation method for DTI prediction. This approach uses average aggregation, effectively reducing noise interference, lowering model complexity, and preventing overfitting, making it especially suitable for current biological networks. Extensive testing on three heterogeneous biological datasets shows that SNADTI outperforms 12 leading methods across two evaluation metrics, significantly reducing training time and validating its effectiveness in DTI prediction. Complexity analysis reveals that our method offers a substantial computational speed advantage over other methods on the same dataset, highlighting its enhanced efficiency. Experimental results demonstrate that SNADTI excels in prediction accuracy, stability, and reproducibility, confirming its practicality and effectiveness in DTI prediction.
{"title":"Drug-target interaction prediction based on metapaths and simplified neighbor aggregation","authors":"Di Yu , Xinyu Yang , Yifan Shang , Sisi Yuan , Yuansheng Liu , Yiping Liu","doi":"10.1016/j.ymeth.2025.04.012","DOIUrl":"10.1016/j.ymeth.2025.04.012","url":null,"abstract":"<div><div>Drug-target interaction (DTI) prediction is critical in drug repositioning and discovery. In current metapath-based prediction methods, attention mechanisms are often used to differentiate the importance of various neighbors, enhancing the model's expressiveness. However, in biological networks with small-scale imbalanced data, attention mechanisms are prone to interference from noise and missing data, leading to instability in weight learning, reduced efficiency, and an increased risk of overfitting. To address these issues, we propose the use of average aggregation to mitigate noise, simplify model complexity, and improve stability. Specifically, we introduce a simplified mean aggregation method for DTI prediction. This approach uses average aggregation, effectively reducing noise interference, lowering model complexity, and preventing overfitting, making it especially suitable for current biological networks. Extensive testing on three heterogeneous biological datasets shows that SNADTI outperforms 12 leading methods across two evaluation metrics, significantly reducing training time and validating its effectiveness in DTI prediction. Complexity analysis reveals that our method offers a substantial computational speed advantage over other methods on the same dataset, highlighting its enhanced efficiency. Experimental results demonstrate that SNADTI excels in prediction accuracy, stability, and reproducibility, confirming its practicality and effectiveness in DTI prediction.</div></div>","PeriodicalId":390,"journal":{"name":"Methods","volume":"240 ","pages":"Pages 154-164"},"PeriodicalIF":4.2,"publicationDate":"2025-04-25","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143899247","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"生物学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2025-04-24DOI: 10.1016/j.ymeth.2025.04.009
Zhen Li , Juyuan Huang , Xinxin Liu , Peng Xu , Xinwen Shen , Chu Pan , Wei Zhang , Wenbin Liu , Henry Han
Predicting drug-target interactions (DTIs) accurately is essential in the field of drug discovery. Recently, artificial intelligence (AI) technologies, especially graph convolutional networks (GCNs), have been developed to tackle this challenge. However, as the number of GCN layers increases, models may lose critical information due to excessive smoothing. Moreover, these methods often lack interpretability and are dependent on specific datasets, which limits their generalizability. Consequently, this study introduces a novel method, KRN-DTI, which employs interpretable GCN technology to predict DTIs based on a drug-target heterogeneous network. The method uses GCN technology to identify potential DTIs by leveraging known interactions and dynamically adjusting the weights, thereby enhancing the model's interpretability. Additionally, residual connection technology is employed to integrate GNN outputs, mitigating the over-smoothing issue. Furthermore, the model's interpretability is enhanced by adaptively adjusting weights using Kolmogorov–Arnold Networks (KAN) and attention mechanisms. Experimental results show that KRN-DTI outperforms several advanced computational methods on the benchmark dataset. Case studies further highlight the effectiveness of KRN-DTI in predicting potential DTIs, showcasing its potential for real-world applications in drug discovery. Our code and data are publicly accessible at: https://github.com/lizhen5000/KRN-DTI.git.
{"title":"KRN-DTI: Towards accurate drug-target interaction prediction with Kolmogorov-Arnold and residual networks","authors":"Zhen Li , Juyuan Huang , Xinxin Liu , Peng Xu , Xinwen Shen , Chu Pan , Wei Zhang , Wenbin Liu , Henry Han","doi":"10.1016/j.ymeth.2025.04.009","DOIUrl":"10.1016/j.ymeth.2025.04.009","url":null,"abstract":"<div><div>Predicting drug-target interactions (DTIs) accurately is essential in the field of drug discovery. Recently, artificial intelligence (AI) technologies, especially graph convolutional networks (GCNs), have been developed to tackle this challenge. However, as the number of GCN layers increases, models may lose critical information due to excessive smoothing. Moreover, these methods often lack interpretability and are dependent on specific datasets, which limits their generalizability. Consequently, this study introduces a novel method, KRN-DTI, which employs interpretable GCN technology to predict DTIs based on a drug-target heterogeneous network. The method uses GCN technology to identify potential DTIs by leveraging known interactions and dynamically adjusting the weights, thereby enhancing the model's interpretability. Additionally, residual connection technology is employed to integrate GNN outputs, mitigating the over-smoothing issue. Furthermore, the model's interpretability is enhanced by adaptively adjusting weights using Kolmogorov–Arnold Networks (KAN) and attention mechanisms. Experimental results show that KRN-DTI outperforms several advanced computational methods on the benchmark dataset. Case studies further highlight the effectiveness of KRN-DTI in predicting potential DTIs, showcasing its potential for real-world applications in drug discovery. Our code and data are publicly accessible at: <span><span>https://github.com/lizhen5000/KRN-DTI.git</span><svg><path></path></svg></span>.</div></div>","PeriodicalId":390,"journal":{"name":"Methods","volume":"240 ","pages":"Pages 137-144"},"PeriodicalIF":4.2,"publicationDate":"2025-04-24","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143887765","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"生物学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
扫码关注我们
求助内容：
应助结果提醒方式：
京公网安备 11010802042870号