Pub Date : 2025-03-04DOI: 10.1016/j.ymeth.2025.03.001
Sisi Yuan , Chen Zhao , Lin Liu , Guifei Zhou
Accurate analysis of molecular structures and the rapid generation of valid molecules remain significant challenges in De Novo drug design. In this study, we propose the Multinomial Generated Diffusion Model (MGDM) for molecular generation. This model leverages a multinomial diffusion framework to process discrete data, with a focus on learning the multinomial distribution inherent in the dataset. During the generation process, the model progressively denoises molecules, transitioning from a uniform noise distribution to ultimately produce valid molecular structures. Initially, we generate molecules unconditionally to expand the compound library. In the next phase, we focus on generating molecules with specific properties to assess the model’s capacity for conditional generation. For this, we implement a classifier-free guidance strategy, which directs the diffusion model’s task without the need for training separate classifier models. To validate the effectiveness of our framework, we conducted experiments using the Molecular Sets (MOSES) dataset. The results demonstrate that, compared to several state-of-the-art methods, MGDM generates valid molecules while achieving superior or comparable performance in terms of novelty and diversity.
{"title":"MGDM: Molecular generation using a multinomial diffusion model","authors":"Sisi Yuan , Chen Zhao , Lin Liu , Guifei Zhou","doi":"10.1016/j.ymeth.2025.03.001","DOIUrl":"10.1016/j.ymeth.2025.03.001","url":null,"abstract":"<div><div>Accurate analysis of molecular structures and the rapid generation of valid molecules remain significant challenges in De Novo drug design. In this study, we propose the <u>M</u>ultinomial <u>G</u>enerated <u>D</u>iffusion <u>M</u>odel (MGDM) for molecular generation. This model leverages a multinomial diffusion framework to process discrete data, with a focus on learning the multinomial distribution inherent in the dataset. During the generation process, the model progressively denoises molecules, transitioning from a uniform noise distribution to ultimately produce valid molecular structures. Initially, we generate molecules unconditionally to expand the compound library. In the next phase, we focus on generating molecules with specific properties to assess the model’s capacity for conditional generation. For this, we implement a classifier-free guidance strategy, which directs the diffusion model’s task without the need for training separate classifier models. To validate the effectiveness of our framework, we conducted experiments using the Molecular Sets (MOSES) dataset. The results demonstrate that, compared to several state-of-the-art methods, MGDM generates valid molecules while achieving superior or comparable performance in terms of novelty and diversity.</div></div>","PeriodicalId":390,"journal":{"name":"Methods","volume":"239 ","pages":"Pages 1-9"},"PeriodicalIF":4.2,"publicationDate":"2025-03-04","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143571840","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 : 2025-02-27DOI: 10.1016/j.ymeth.2025.02.009
Aneta Węgierek-Ciuk , Maria Baczewska , Katarzyna Gałczyńska , Paula Ortega , Francisco Javier de la Mata , Małgorzata Kujawińska , Michał Arabski
Resistance of cancer cells to chemotherapy is one of the major causes of treatment failure and poor patient survival. Reduced cellular response to drugs can result from their genetic diversity, acquired mutations of drug targets, epigenetic modifications and many others. Metallodendrimers, in particular, ruthenium dendrimers of the first and second generation are promising novel anticancer drug carriers, as their usage can result in increased drug concentration in tumour tissue and reduced toxicity in healthy tissues. However the conventional, biological methods do not provide sufficient knowledge about cytotoxicity of these compounds. Therefore in the paper we propose an efficient, multimodal methodology for cytotoxicity studies at cellular level. It combines the conventional flow cytometry method (Annexin V-FITC assay), which provides global/statistical information about a cell culture, and digital holographic microscopy (DHM) allowing for continuous quantitative monitoring of cells behaviour and identifying cells with non-standard behaviour. The results reveal that tested ruthenium metallodendrimers have a strong impact on apoptotic and necrotic death of both human alveolar epithelial cell line A549 and Chinese hamster ovary cell line CHO-K1. We also have shown that DHM enables detection of the individual, drug resistance cells, through real-time monitoring of a single cell. We believe that the methodology proposed is a necessary supplement to conventional approach for studying drug cytotoxicity, which will help, in the near future, to overcome the problem of cellular resistance to anticancer therapies.
{"title":"Enhanced methodology for analysis cytotoxicity of ruthenium dendrimers","authors":"Aneta Węgierek-Ciuk , Maria Baczewska , Katarzyna Gałczyńska , Paula Ortega , Francisco Javier de la Mata , Małgorzata Kujawińska , Michał Arabski","doi":"10.1016/j.ymeth.2025.02.009","DOIUrl":"10.1016/j.ymeth.2025.02.009","url":null,"abstract":"<div><div>Resistance of cancer cells to chemotherapy is one of the major causes of treatment failure and poor patient survival. Reduced cellular response to drugs can result from their genetic diversity, acquired mutations of drug targets, epigenetic modifications and many others. Metallodendrimers, in particular, ruthenium dendrimers of the first and second generation are promising novel anticancer drug carriers, as their usage can result in increased drug concentration in tumour tissue and reduced toxicity in healthy tissues. However the conventional, biological methods do not provide sufficient knowledge about cytotoxicity of these compounds. Therefore in the paper we propose an efficient, multimodal methodology for cytotoxicity studies at cellular level. It combines the conventional flow cytometry method (Annexin V-FITC assay), which provides global/statistical information about a cell culture, and digital holographic microscopy (DHM) allowing for continuous quantitative monitoring of cells behaviour and identifying cells with non-standard behaviour. The results reveal that tested ruthenium metallodendrimers have a strong impact on apoptotic and necrotic death of both human alveolar epithelial cell line A549 and Chinese hamster ovary cell line CHO-K1. We also have shown that DHM enables detection of the individual, drug resistance cells, through real-time monitoring of a single cell. We believe that the methodology proposed is a necessary supplement to conventional approach for studying drug cytotoxicity, which will help, in the near future, to overcome the problem of cellular resistance to anticancer therapies.</div></div>","PeriodicalId":390,"journal":{"name":"Methods","volume":"238 ","pages":"Pages 1-10"},"PeriodicalIF":4.2,"publicationDate":"2025-02-27","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143536354","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}
Bacterial vesicles (BVs) are membrane-bound extracellular vesicles (EV) released from bacteria. They are known to play crucial role in bacterial communication, host-pathogen interactions, transfer of virulence factors, contribute to immune modulation and are the key players in microbial pathogenesis and survival in the host. Despite their significance, isolation and investigating BVs from human samples remains challenging, necessitating an easy, reliable and reproducible protocol. BVs have been limited due to methodological difficulties in isolating them from host-derived EVs, and the existing knowledge primarily relies on bacteria cultured under controlled laboratory conditions. This study presents a method, where we can identify the enriched BVs and characterizing them from plasma and stool samples of healthy individuals. Blood and fecal samples were collected, processed to density gradient ultracentrifugation to isolate and enrich BVs. Morphological characterization was performed using transmission electron microscopy (TEM), nanoparticle tracking analysis (NTA). Further, molecular markers OmpA (BV marker) was used to differentiate from host EVs (Alix as marker) using Western blot. Further the BV fraction was analyzed for LPS and LTA using ELISA. To understand functional relevance, BVs proteomics was performed from BV enriched plasma and stool using mass spectrometry from healthy individuals. The enriched BVs were also co-cultured with healthy peripheral blood mononuclear cells, labelled with Pkh26 dye and analysed at different time points for mRNA expression of candidate genes involved in immune regulation by qRT-PCR. Both TEM and NTA confirmed the presence of BVs, with sizes ranging from 25 nm to 250 nm. The western blot analysis revealed the fractions 6–9 are enriched with host EVs with the presence of Alix and fractions 10–13 contains BVs with the presence of OmpA. Interestingly, the proteomic analysis identified 439 proteins associated with plasma-derived BVs and 327 in stool-derived BVs, with 300 common to both. The Gene ontology and KEGG pathway analysis revealed the majority of proteins associated were immune regulation, cell activation, binding, and catalytic activity. Next, the functional assays indicated BVs were uptaken by PBMCs within 10 mins and it upregulated Toll-like receptor 2 (TLR-2) expression within 30 min. Hence, study establishes a reliable method to identify enriched BV population from human samples. Revealed the proteins associated with BVs in healthy individuals and their role in immune regulation. These findings may provide a platform to investigate BVs potential for diagnostic and therapeutic applications in various diseases.
{"title":"Comprehensive method for isolation and functional characterization of bacterial vesicles from human biological samples","authors":"Swati Thangariyal , Sakshi Bhat , Ashmit Mittal , P. Debishree Subudhi , Preeti Negi , Chhagan Bihari , Shiv Kumar Sarin , Sukriti Baweja","doi":"10.1016/j.ymeth.2025.02.012","DOIUrl":"10.1016/j.ymeth.2025.02.012","url":null,"abstract":"<div><div>Bacterial vesicles (BVs) are membrane-bound extracellular vesicles (EV) released from bacteria. They are known to play crucial role in bacterial communication, host-pathogen interactions, transfer of virulence factors, contribute to immune modulation and are the key players in microbial pathogenesis and survival in the host. Despite their significance, isolation and investigating BVs from human samples remains challenging, necessitating an easy, reliable and reproducible protocol. BVs have been limited due to methodological difficulties in isolating them from host-derived EVs, and the existing knowledge primarily relies on bacteria cultured under controlled laboratory conditions. This study presents a method, where we can identify the enriched BVs and characterizing them from plasma and stool samples of healthy individuals. Blood and fecal samples were collected, processed to density gradient ultracentrifugation to isolate and enrich BVs. Morphological characterization was performed using transmission electron microscopy (TEM), nanoparticle tracking analysis (NTA). Further, molecular markers OmpA (BV marker) was used to differentiate from host EVs (Alix as marker) using Western blot. Further the BV fraction was analyzed for LPS and LTA using ELISA. To understand functional relevance, BVs proteomics was performed from BV enriched plasma and stool using mass spectrometry from healthy individuals. The enriched BVs were also co-cultured with healthy peripheral blood mononuclear cells, labelled with Pkh26 dye and analysed at different time points for mRNA expression of candidate genes involved in immune regulation by qRT-PCR. Both TEM and NTA confirmed the presence of BVs, with sizes ranging from 25 nm to 250 nm. The western blot analysis revealed the fractions 6–9 are enriched with host EVs with the presence of Alix and fractions 10–13 contains BVs with the presence of OmpA. Interestingly, the proteomic analysis identified 439 proteins associated with plasma-derived BVs and 327 in stool-derived BVs, with 300 common to both. The Gene ontology and KEGG pathway analysis revealed the majority of proteins associated were immune regulation, cell activation, binding, and catalytic activity. Next, the functional assays indicated BVs were uptaken by PBMCs within 10 mins and it upregulated Toll-like receptor 2 (TLR-2) expression within 30 min. Hence, study establishes a reliable method to identify enriched BV population from human samples. Revealed the proteins associated with BVs in healthy individuals and their role in immune regulation. These findings may provide a platform to investigate BVs potential for diagnostic and therapeutic applications in various diseases.</div></div>","PeriodicalId":390,"journal":{"name":"Methods","volume":"237 ","pages":"Pages 1-8"},"PeriodicalIF":4.2,"publicationDate":"2025-02-27","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143526863","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 : 2025-02-27DOI: 10.1016/j.ymeth.2025.02.013
S. Michurina , Y. Goltseva , E. Ratner , K. Dergilev , E. Shestakova , I. Minniakhmetov , S. Rumyantsev , I. Stafeev , M. Shestakova , Ye. Parfyonova
Lipid droplets (LDs) are dynamic organelles that are present in almost all cell types, with a particularly high prevalence in adipocytes. The phenotype of LDs in these cells reflects their maturity, metabolic activity and function. Although LDs quantification in adipocytes is significant for understanding the origins of obesity and associated complications, it remains challenging and requires the implementation of computer science innovations.
This article outlines a practical workflow for application of Segment.ai neural network from the commercial software NIS-Elements and the open-source StarDist Jupyter notebook from the ZeroCostDL4Mic platform for the analysis of LDs number and morphology. To generate a training dataset, 3T3-L1 cells were differentiated into adipocytes and stained with lipophilic dye BODIPY493/503. Subsequently, confocal live cell images were acquired, annotated and used for training. As an example task, deep learning models were tested on their ability to detect LDs enlargement on images of adipocytes with inhibited lipolysis.
We demonstrated that both Segment.ai and StarDist models are capable of accurately recognising LDs on microphotographs, thereby significantly accelerating the processing of imaging data. The advantage of the Segment.ai model is its integration into NIS-Elements General Analysis 3, which performs quantitative and statistical data interpretation. Alternatively, StarDist is a more accessible and transparent tool, enabling precise model evaluation. In conclusion, both created approaches have the potential to accelerate the exploration of LDs dynamics, thus paving the way for further insights into how these organelles regulate energy homeostasis and contribute to the development of metabolic abnormalities.
{"title":"Artificial intelligence–enabled lipid droplets quantification: Comparative analysis of NIS-elements Segment.ai and ZeroCostDL4Mic StarDist networks","authors":"S. Michurina , Y. Goltseva , E. Ratner , K. Dergilev , E. Shestakova , I. Minniakhmetov , S. Rumyantsev , I. Stafeev , M. Shestakova , Ye. Parfyonova","doi":"10.1016/j.ymeth.2025.02.013","DOIUrl":"10.1016/j.ymeth.2025.02.013","url":null,"abstract":"<div><div>Lipid droplets (LDs) are dynamic organelles that are present in almost all cell types, with a particularly high prevalence in adipocytes. The phenotype of LDs in these cells reflects their maturity, metabolic activity and function. Although LDs quantification in adipocytes is significant for understanding the origins of obesity and associated complications, it remains challenging and requires the implementation of computer science innovations.</div><div>This article outlines a practical workflow for application of <em>Segment.ai</em> neural network from the commercial software NIS-Elements and the open-source <em>StarDist</em> Jupyter notebook from the ZeroCostDL4Mic platform for the analysis of LDs number and morphology. To generate a training dataset, 3T3-L1 cells were differentiated into adipocytes and stained with lipophilic dye BODIPY493/503. Subsequently, confocal live cell images were acquired, annotated and used for training. As an example task, deep learning models were tested on their ability to detect LDs enlargement on images of adipocytes with inhibited lipolysis.</div><div>We demonstrated that both <em>Segment.ai</em> and <em>StarDist</em> models are capable of accurately recognising LDs on microphotographs, thereby significantly accelerating the processing of imaging data. The advantage of the <em>Segment.ai</em> model is its integration into NIS-Elements General Analysis 3, which performs quantitative and statistical data interpretation. Alternatively, <em>StarDist</em> is a more accessible and transparent tool, enabling precise model evaluation. In conclusion, both created approaches have the potential to accelerate the exploration of LDs dynamics, thus paving the way for further insights into how these organelles regulate energy homeostasis and contribute to the development of metabolic abnormalities.</div></div>","PeriodicalId":390,"journal":{"name":"Methods","volume":"237 ","pages":"Pages 9-18"},"PeriodicalIF":4.2,"publicationDate":"2025-02-27","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143526864","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 : 2025-02-26DOI: 10.1016/j.ymeth.2025.02.011
Zahra Zareshahrabadi , Sara Shenavari , Forough Karami , Mohammad Hashem Hashempur , Mohammad Khorram , Ali Arabimonfard , Mahboobeh Jafari , Ali Mohammad Tamaddon , Gholamhossein Yousefi , Kamiar Zomorodian
Wound infections are challenging to manage, requiring innovative wound dressing systems with prescribed properties. Active wound dressings should provide a moist environment, protect from secondary infections, and remove wound exudate to accelerate tissue regeneration. Hydrogels are encouraging matrices for bioactive compound encapsulation in pharmaceutical applications. The aim of the present study is to design chitosan/gelatin /polyvinyl alcohol-based xerogel film containing nystatin for wound dressing applications with antifungal properties. The xerogel film is developed using a film-casting method and evaluated for its chemical and physical characteristics using FTIR, SEM, AFM, and tensile analysis. Water barrier properties of the film, such as the moisture content (17.84 ± 3.62 %), water solubility (44.50 ± 5.55 %), gel fraction (55.50 ± 5.55 %), and water vapour transmission rate (1912.25 ± 248.12 g m−2 per day), suggest a humid microenvironment suitable for wound. The xerogel film, characterized by its robust mechanical strength, substantial swelling capacity (∼120–400 %) across various pH levels, and acceptable bio-adhesive properties, reveals as a potential antifungal wound dressing material. In vitro toxicity assessments confirm its biocompatibility towards both RBCs and NIH-3 T3 fibroblast cells. The findings confirm that the film formulation has strong antifungal properties, with a minimum inhibitory concentration of 2–8 μL/mL against Candida species, as well as outstanding antibiofilm effectiveness (∼85 %) and a significant reduction of the fungal colony count (∼100 %). Moreover, its controlled drug release capabilities, along with antifungal properties, offer it as an appealing dressing for the localized treatment of superficial fungal infections. As a result, this xerogel film can be used as a versatile platform for advanced wound care and therapeutic applications.
{"title":"Chitosan-based xerogel film incorporating Nystatin: Synthesis, structural Analysis, and biological evaluation","authors":"Zahra Zareshahrabadi , Sara Shenavari , Forough Karami , Mohammad Hashem Hashempur , Mohammad Khorram , Ali Arabimonfard , Mahboobeh Jafari , Ali Mohammad Tamaddon , Gholamhossein Yousefi , Kamiar Zomorodian","doi":"10.1016/j.ymeth.2025.02.011","DOIUrl":"10.1016/j.ymeth.2025.02.011","url":null,"abstract":"<div><div>Wound infections are challenging to manage, requiring innovative wound dressing systems with prescribed properties. Active wound dressings should provide a moist environment, protect from secondary infections, and remove wound exudate to accelerate tissue regeneration. Hydrogels are encouraging matrices for bioactive compound encapsulation in pharmaceutical applications. The aim of the present study is to design chitosan/gelatin /polyvinyl alcohol-based xerogel film containing nystatin for wound dressing applications with antifungal properties. The xerogel film is developed using a film-casting method and evaluated for its chemical and physical characteristics using FTIR, SEM, AFM, and tensile analysis. Water barrier properties of the film, such as the moisture content (17.84 ± 3.62 %), water solubility (44.50 ± 5.55 %), gel fraction (55.50 ± 5.55 %), and water vapour transmission rate (1912.25 ± 248.12 g m<sup>−2</sup> per day), suggest a humid microenvironment suitable for wound. The xerogel film, characterized by its robust mechanical strength, substantial swelling capacity (∼120–400 %) across various pH levels, and acceptable bio-adhesive properties, reveals as a potential antifungal wound dressing material. <em>In vitro</em> toxicity assessments confirm its biocompatibility towards both RBCs and NIH-3 T3 fibroblast cells. The findings confirm that the film formulation has strong antifungal properties, with a minimum inhibitory concentration of 2–8 μL/mL against <em>Candida</em> species, as well as outstanding antibiofilm effectiveness (∼85 %) and a significant reduction of the fungal colony count (∼100 %). Moreover, its controlled drug release capabilities, along with antifungal properties, offer it as an appealing dressing for the localized treatment of superficial fungal infections. As a result, this xerogel film can be used as a versatile platform for advanced wound care and therapeutic applications.</div></div>","PeriodicalId":390,"journal":{"name":"Methods","volume":"237 ","pages":"Pages 19-33"},"PeriodicalIF":4.2,"publicationDate":"2025-02-26","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143530820","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 : 2025-02-26DOI: 10.1016/j.ymeth.2025.02.008
Clara Sanchez , Morgane Nadal , Céline Cansell , Sarah Laroui , Xavier Descombes , Carole Rovère , Éric Debreuve
Obesity is associated with brain inflammation, glial reactivity, and immune cells infiltration. Studies in rodents have shown that glial reactivity occurs within 24 h of high-fat diet (HFD) consumption, long before obesity development, and takes place mainly in the hypothalamus (HT), a crucial brain structure for controlling body weight. Understanding more precisely the kinetics of glial activation of two major brain cells (astrocytes and microglia) and their impact on eating behavior could prevent obesity and offer new prospects for therapeutic treatments. To understand the mechanisms pertaining to obesity-related neuroinflammation, we developed a fully automated algorithm, NutriMorph. Although some algorithms were developed in the past decade to detect and segment cells, they are highly specific, not fully automatic, and do not provide the desired morphological analysis. Our algorithm copes with these issues and performs the analysis of cells images (here, microglia of the hypothalamic arcuate nucleus), and the morphological clustering of these cells through statistical analysis and machine learning. Using the k-Means algorithm, it clusters the microglia of the control condition (healthy mice) and the different states of neuroinflammation induced by high-fat diets (obese mice) into subpopulations. This paper is an extension and re-analysis of a first published paper showing that microglial reactivity can already be seen after few hours of high-fat diet (Cansell et al., 2021 [5]). Thanks to NutriMorph algorithm, we unravel the presence of different hypothalamic microglial subpopulations (based on morphology) subject to proportion changes in response to already few hours of high-fat diet in mice.
{"title":"Computational detection, characterization, and clustering of microglial cells in a mouse model of fat-induced postprandial hypothalamic inflammation","authors":"Clara Sanchez , Morgane Nadal , Céline Cansell , Sarah Laroui , Xavier Descombes , Carole Rovère , Éric Debreuve","doi":"10.1016/j.ymeth.2025.02.008","DOIUrl":"10.1016/j.ymeth.2025.02.008","url":null,"abstract":"<div><div>Obesity is associated with brain inflammation, glial reactivity, and immune cells infiltration. Studies in rodents have shown that glial reactivity occurs within 24 h of high-fat diet (HFD) consumption, long before obesity development, and takes place mainly in the hypothalamus (HT), a crucial brain structure for controlling body weight. Understanding more precisely the kinetics of glial activation of two major brain cells (astrocytes and microglia) and their impact on eating behavior could prevent obesity and offer new prospects for therapeutic treatments. To understand the mechanisms pertaining to obesity-related neuroinflammation, we developed a fully automated algorithm, NutriMorph. Although some algorithms were developed in the past decade to detect and segment cells, they are highly specific, not fully automatic, and do not provide the desired morphological analysis. Our algorithm copes with these issues and performs the analysis of cells images (here, microglia of the hypothalamic arcuate nucleus), and the morphological clustering of these cells through statistical analysis and machine learning. Using the k-Means algorithm, it clusters the microglia of the control condition (healthy mice) and the different states of neuroinflammation induced by high-fat diets (obese mice) into subpopulations. This paper is an extension and re-analysis of a first published paper showing that microglial reactivity can already be seen after few hours of high-fat diet (Cansell et al., 2021 [5]). Thanks to NutriMorph algorithm, we unravel the presence of different hypothalamic microglial subpopulations (based on morphology) subject to proportion changes in response to already few hours of high-fat diet in mice.</div></div>","PeriodicalId":390,"journal":{"name":"Methods","volume":"236 ","pages":"Pages 28-38"},"PeriodicalIF":4.2,"publicationDate":"2025-02-26","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143530823","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"生物学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2025-02-26DOI: 10.1016/j.ymeth.2025.02.010
Bing-Xue Du , Haoyang Yu , Bei Zhu , Yahui Long , Min Wu , Jian-Yu Shi
It is a vital step to identify the enzyme turnover number (kcat) in synthetic biology and early-stage drug discovery. Recently, deep learning methods have achieved inspiring process to predict kcat with the development of multi-species enzyme-substrate pairs turnover number data. However, the performance of existing approaches still heavily depends on the effectiveness of feature extraction for enzymes and substrates, as well as the optimal fusion of these two types of features. Furthermore, it is essential to identify the key molecular substructures that significantly impact kcat prediction. To address these issues, we develop a novel end-to-end dual-representation interpretability framework GELKcat by harnessing graph transformers for substrate molecular encoding and CNNs for enzyme word2vec embeddings. We further integrate substrate and enzyme features using the adaptive gate network, which assigns optimal weights to capture the most suitable feature combinations. The comparison with several state-of-the-art methods demonstrates the superiority of our GELKcat and the ablation studies further illuminate the invaluable roles of three main components. Furthermore, case studies illustrate the interpretability of GELKcat by identifying the key functional groups in a substrate, which are significantly associated with turnover number. It is anticipated that this work can bridge current gaps in enzyme-substrate representation, which can give some guidance for drug discovery and synthetic biology.
{"title":"A novel interpretability framework for enzyme turnover number prediction boosted by pre-trained enzyme embeddings and adaptive gate network","authors":"Bing-Xue Du , Haoyang Yu , Bei Zhu , Yahui Long , Min Wu , Jian-Yu Shi","doi":"10.1016/j.ymeth.2025.02.010","DOIUrl":"10.1016/j.ymeth.2025.02.010","url":null,"abstract":"<div><div>It is a vital step to identify the enzyme turnover number (kcat) in synthetic biology and early-stage drug discovery. Recently, deep learning methods have achieved inspiring process to predict kcat with the development of multi-species enzyme-substrate pairs turnover number data. However, the performance of existing approaches still heavily depends on the effectiveness of feature extraction for enzymes and substrates, as well as the optimal fusion of these two types of features. Furthermore, it is essential to identify the key molecular substructures that significantly impact kcat prediction. To address these issues, we develop a novel end-to-end dual-representation interpretability framework GELKcat by harnessing graph transformers for substrate molecular encoding and CNNs for enzyme word2vec embeddings. We further integrate substrate and enzyme features using the adaptive gate network, which assigns optimal weights to capture the most suitable feature combinations. The comparison with several state-of-the-art methods demonstrates the superiority of our GELKcat and the ablation studies further illuminate the invaluable roles of three main components. Furthermore, case studies illustrate the interpretability of GELKcat by identifying the key functional groups in a substrate, which are significantly associated with turnover number. It is anticipated that this work can bridge current gaps in enzyme-substrate representation, which can give some guidance for drug discovery and synthetic biology.</div></div>","PeriodicalId":390,"journal":{"name":"Methods","volume":"237 ","pages":"Pages 45-52"},"PeriodicalIF":4.2,"publicationDate":"2025-02-26","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143531164","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 : 2025-02-22DOI: 10.1016/j.ymeth.2025.02.007
Elmira I. Yakupova , Polina A. Abramicheva , Vadim V. Rogachevsky , Elena A. Shishkova , Alexey D. Bocharnikov , Egor Y. Plotnikov , Ivan M. Vikhlyantsev
The sarcomeric giant protein titin affects the passive elasticity of the heart muscle and is crucial for proper cardiac function, including diastolic relaxation of the left ventricle. A useful common method for studying titin is electrophoretic analysis which can be used to examine the distribution of its isoforms in the heart. There are 5 titin parameters that can be analyzed: the N2BA/N2B isoforms ratio, the T2/T1 bands ratio, Cronos isoform content, NT isoform content, the total titin-to-myosin heavy chain (TT/MHC) ratio. These parameters can only be assessed through electrophoresis of giant proteins. It is known that these parameters are related to various biomolecular processes in muscle cells, such as providing of elastic properties, turnover, contraction, and maintaining a highly ordered sarcomere structure. In this review, we discuss the diagnostic potential of electrophoretic visualization of cardiac titin changes in various human heart diseases and animal models of physiological adaptations or pathologies.
{"title":"Cardiac titin isoforms: Practice in interpreting results of electrophoretic analysis","authors":"Elmira I. Yakupova , Polina A. Abramicheva , Vadim V. Rogachevsky , Elena A. Shishkova , Alexey D. Bocharnikov , Egor Y. Plotnikov , Ivan M. Vikhlyantsev","doi":"10.1016/j.ymeth.2025.02.007","DOIUrl":"10.1016/j.ymeth.2025.02.007","url":null,"abstract":"<div><div>The sarcomeric giant protein titin affects the passive elasticity of the heart muscle and is crucial for proper cardiac function, including diastolic relaxation of the left ventricle. A useful common method for studying titin is electrophoretic analysis which can be used to examine the distribution of its isoforms in the heart. There are 5 titin parameters that can be analyzed: the N2BA/N2B isoforms ratio, the T2/T1 bands ratio, Cronos isoform content, NT isoform content, the total titin-to-myosin heavy chain (TT/MHC) ratio. These parameters can only be assessed through electrophoresis of giant proteins. It is known that these parameters are related to various biomolecular processes in muscle cells, such as providing of elastic properties, turnover, contraction, and maintaining a highly ordered sarcomere structure. In this review, we discuss the diagnostic potential of electrophoretic visualization of cardiac titin changes in various human heart diseases and animal models of physiological adaptations or pathologies.</div></div>","PeriodicalId":390,"journal":{"name":"Methods","volume":"236 ","pages":"Pages 17-25"},"PeriodicalIF":4.2,"publicationDate":"2025-02-22","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143490308","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 : 2025-02-21DOI: 10.1016/j.ymeth.2025.02.006
Zachary Ells , Vinicius Ludwig , Adam B. Weiner , Koichiro Kimura , Andrea Farolfi , Karim Chamie , Joseph Shirk , Nicholas M. Donin , Robert Reiter , Johannes Czernin , Jeremie Calais , Magnus Dahlbom
Introduction
Surgeons commonly use cross sectional images to plan and prepare for surgical procedures. However, cognitively translating 2D images to surgical settings can be difficult and lead to sub-optimal resections. Lymph node dissection can be challenging due to the inability to locate small metastatic lesions, and their proximity to at-risk organ(s). 3D volume rendered (3D-VR) patient specific images can help to address these challenges. We created patient-specific 3D-VR images using freely available open-source programs.
Methods
This study included patients part of the clinical trial NCT04857502. Patients received a PET/CT prior to radioguided surgery. 3D Slicer was used to segment anatomy of interest (organs and tumor lesion(s)). After segmentation, the data was exported as an .OBJ file with an accompanying .MTL file. Manipulation of the .MTL file to restore model properties to the .OBJ file, were completed and both files were uploaded into Autodesk Viewer. Surgeons then received an email link to access the finished 3D-VR model on their smartphone or laptop for peri-operative preparation and/or guidance.
Results
The method was used in a series of 14 patients with prostate cancer undergoing pelvic lymph node dissection with PSMA-radioguided robotic surgery using pre-operative PSMA PET/CT images acquired on average 103 ± 69 days prior to resection. The creation of the 3D-VR models was successfully conducted in all 14 cases. In all cases, the lesions identified on the pre-operative PET/CT imaging 3D-VR models were successfully removed during surgery.
Conclusion
We created patient-specific anatomical 3D-VR models that the surgeons can use for pre-surgical planning and intraoperative tumor localization, by applying free, open-source software that could be used in any procedure requiring careful and strategical planning.
{"title":"A free method for patient-specific 3D-VR anatomical modeling for presurgical planning using DICOM images and open-source software","authors":"Zachary Ells , Vinicius Ludwig , Adam B. Weiner , Koichiro Kimura , Andrea Farolfi , Karim Chamie , Joseph Shirk , Nicholas M. Donin , Robert Reiter , Johannes Czernin , Jeremie Calais , Magnus Dahlbom","doi":"10.1016/j.ymeth.2025.02.006","DOIUrl":"10.1016/j.ymeth.2025.02.006","url":null,"abstract":"<div><h3>Introduction</h3><div>Surgeons commonly use cross sectional images to plan and prepare for surgical procedures. However, cognitively translating 2D images to surgical settings can be difficult and lead to sub-optimal resections. Lymph node dissection can be challenging due to the inability to locate small metastatic lesions, and their proximity to at-risk organ(s). 3D volume rendered (3D-VR) patient specific images can help to address these challenges. We created patient-specific 3D-VR images using freely available open-source programs.</div></div><div><h3>Methods</h3><div>This study included patients part of the clinical trial NCT04857502. Patients received a PET/CT prior to radioguided surgery. 3D Slicer was used to segment anatomy of interest (organs and tumor lesion(s)). After segmentation, the data was exported as an .OBJ file with an accompanying .MTL file. Manipulation of the .MTL file to restore model properties to the .OBJ file, were completed and both files were uploaded into Autodesk Viewer. Surgeons then received an email link to access the finished 3D-VR model on their smartphone or laptop for peri-operative preparation and/or guidance.</div></div><div><h3>Results</h3><div>The method was used in a series of 14 patients with prostate cancer undergoing pelvic lymph node dissection with PSMA-radioguided robotic surgery using pre-operative PSMA PET/CT images acquired on average 103 ± 69 days prior to resection. The creation of the 3D-VR models was successfully conducted in all 14 cases. In all cases, the lesions identified on the pre-operative PET/CT imaging 3D-VR models were successfully removed during surgery.</div></div><div><h3>Conclusion</h3><div>We created patient-specific anatomical 3D-VR models that the surgeons can use for pre-surgical planning and intraoperative tumor localization, by applying free, open-source software that could be used in any procedure requiring careful and strategical planning.</div></div>","PeriodicalId":390,"journal":{"name":"Methods","volume":"236 ","pages":"Pages 10-16"},"PeriodicalIF":4.2,"publicationDate":"2025-02-21","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143473927","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 : 2025-02-20DOI: 10.1016/j.ymeth.2025.02.005
Haiying Wang (Guest Editor), Xiaohua (Tony) Hu (Guest Editor)
{"title":"AI and machine learning in bioinformatics and biomedicine","authors":"Haiying Wang (Guest Editor), Xiaohua (Tony) Hu (Guest Editor)","doi":"10.1016/j.ymeth.2025.02.005","DOIUrl":"10.1016/j.ymeth.2025.02.005","url":null,"abstract":"","PeriodicalId":390,"journal":{"name":"Methods","volume":"236 ","pages":"Pages 26-27"},"PeriodicalIF":4.2,"publicationDate":"2025-02-20","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143476048","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