[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"doc-detail-119363-en":3,"doc-seo-119363-105":29,"detail-sidebar-cat-0-en-105":89},{"code":4,"msg":5,"data":6},0,"success",{"doc_id":7,"user_id":8,"nickname":9,"user_avatar":10,"doc_module":4,"category_id":11,"category_name":12,"doc_title":13,"doc_description":14,"doc_content":15,"file_id":16,"file_url":17,"file_type":18,"file_size":19,"view_count":4,"is_deleted":4,"is_public":20,"is_downloadable":20,"audit_status":20,"page_count":20,"language":21,"language_code":22,"site_id":23,"html_lang":22,"table_of_contents":24,"faqs":25,"seo_title":26,"seo_description":14,"update_tm":27,"read_time":28},119363,1099514067415,"Rowan","https://ap-avatar.wpscdn.com/avatar/100002539d78ffe74a7?x-image-process=image/resize,m_fixed,w_180,h_180&k=1779092875211072502",8,"Research & Report","ARSENAL - Antimicrobial ReSistance prEdictioN by a mAchine Learning method","ARSENAL presents a machine learning approach to predict antibiotic resistance levels from genomic data, addressing limits of standard antibiotic susceptibility testing that relies on slow (24–72 h) cultures and only cultivable bacteria. The method targets a large dataset of newly sequenced Streptococcus pneumoniae strains with MIC measurements for multiple beta-lactams. Feature selection reduces input from ~1,000,000 SNPs to ~1,000 using variance filtering, SNP clustering by orthogroups and inter-gene correlations, and optimized multivariate selection to improve MIC prediction.","ARSENAL: Antimicrobial ReSistance prEdictioN  \nby a mAchine Learning method  \nSimankov Nikolay 1,2,4,5,* , Ulysse Guyet 1,2 , Léa Bientz 3 , Véronique Dubois 3 , Alexis Groppi 1,2 , and Macha Nikolski 1,2  \n1 Univ. Bordeaux, CNRS, IBGC, UMR 5095, Bordeaux, 33077, France, 2 Univ. Bordeaux, Centre de Bioinformatique de Bordeaux (CBiB), Bordeaux, 33076, France, 3 MFP, CNRS 5234, Université de Bordeaux, Bordeaux,  \n4 Laboratory of Plant Pathology – TERRA-Gembloux Agro-BioTech – University of Liège (ULiège) -5030 Gembloux, Belgium, 5 Statistics, Computer Science and Modeling applied to bioengineering (SIMa)– TERRA – University of Liège (ULiège) 5030 Gembloux, Belgium.  \n*This Communication is supported by the Walloon Region as part of a FRIA grant.  \n\n| \u003Cbr>Background \u003Cbr>➢\u003Cbr>➢\u003Cbr>➢\u003Cbr>➢ | Antimicrobial resistance (AMR) is a growing health threat responsible for an estimated 700 000 deaths per year and this number is projected to reach 10 million by 2050 .\u003Cbr>Appropriate antibiotic therapy improves patient healing outcomes and is a key factor in preventing the emergence of antibiotic resistance.\u003Cbr>Antibiotic susceptibility testing (AST) from bacterial culture is the current clinical practice for assessing drug resistance by the determination of the minimum inhibitory concentration (MIC) corresponding to the lowest concentration of a specific antibiotic that inhibits bacterial growth. This method, fastidious and long (between 24h and 72h), is only applicable to cultivable bacteria, which excludes analysis of the emergence and spread of antimicrobial resistance in diverse and complex microbial communities with large fractions of currently uncultured bacteria.\u003Cbr>Our method allows:\u003Cbr>✓ Fast screening of antibiotics to determine the most effective antibiotic and the right dose against a specific bacterial infection\u003Cbr>✓ Highlight new antibiotic-resistance genes and biomarkers\u003Cbr>Resistance mechanisms are mainly acquired by genome modifications, e.g., mutations and horizontal gene transfers. |\n| --- | --- |\n\n\n| \u003Cbr>Data Generation\u003Cbr>\u003Cbr>➢ 1312 Streptococcus pneumoniae newly sequenced genomes of strains isolated from patients\u003Cbr>❑ Different sequence types (ST) and 64 distinct serotypes\u003Cbr>❑ Many strains are multidrug resistant\u003Cbr>❑ Diversity of geographical origin (isolated from 28 hospitals in 18 provinces of China between 2007 and 2020) |  |  |  |  |  |  |  |\n| --- | --- | --- | --- | --- | --- | --- | --- |\n|  |  |  |  |  |  |  |  |\n| ➢ MIC values (Minimum Inhibitory Concentration) of each strain measured by antibiotic susceptibility testing mainly for 6 Beta-lactam antibiotics :\u003Cbr>❑ Penicillin ❑ Ceftriaxone\u003Cbr>❑ Amoxicillin ❑ Cefepime\u003Cbr>❑ Cefuroxime ❑ Imipenem |  |  |  | ❑ +400 000 SNPs positions identified\u003Cbr>❑ About 4000 gene abundance profiles |  |  |  |\n\nFeature Selection  \nThe Challenge :  \n❑ The identification of resistance-related genes supports the relevance of the model.  \n❑ GWAS-based approach resulted in a highly correlated matrix with insufficient resolution for MIC prediction.  \n❑ Collinear features are a thread for Machine Learning efficiency.  \n❑ The initial number of SNPs does not allow us to compute the correlation matrix.  \n❑ Different population types required different approaches.  \nThe Results :  \n✓ We reduced the input data from 1, 000, 000 to  \n1, 000 features  \n✓ Clusters contain all highly correlated features that allow us to study them afterward  \nLow Variance Filtering  \n• SNPs with a variance lower than 0.05% are removed  \n• ≈ 0.95 > prevalence > 0.05  \nPer Gene SNP Clustering  \n• SNPs from the same orthogroup are clustered together according to their Spearman correlation (>95%)  \n• All features inside each cluster are represented by one single SNP  \nInter Gene Clustering  \n• Representative SNPs that share the same Spearman correlation to a random vector are clustered together (>95%)  \n• All features inside each cluster are represented by one single SNP  \nStatistical Univariate Cluster Ranking  \n• Un","cbCairXZ4mukMB2I","https://ap.wps.com/l/cbCairXZ4mukMB2I","pdf",839860,1,"English","en",105,"# Background\n## Data generation\n## Feature selection\n## Conclusion and perspectives","[{\"question\":\"Why is antibiotic susceptibility testing (AST) insufficient for antimicrobial resistance studies?\",\"answer\":\"AST based on bacterial culture is slow (24–72 h) and applies only to cultivable bacteria, which limits the analysis of resistance emergence and spread in diverse microbial communities with many uncultured organisms.\"},{\"question\":\"What data does ARSENAL use to predict resistance?\",\"answer\":\"It uses 1,312 newly sequenced Streptococcus pneumoniae genomes along with measured MIC values from antibiotic susceptibility testing, mainly for six beta-lactam antibiotics, plus the SNP and gene abundance profiles derived from the genomic data.\"},{\"question\":\"How does ARSENAL reduce the genomic feature space for machine learning?\",\"answer\":\"It applies low-variance filtering, then clusters SNPs using Spearman correlation within orthogroups and across genes to represent each cluster by a single representative SNP, followed by univariate cluster ranking and optimized multivariate feature selection to choose the most predictive feature combinations.\"}]","ARSENAL - Antimicrobial ReSistance prEdictioN by a mAchine Learning method | PDF",1785723919,3,{"code":4,"msg":30,"data":31},"ok",{"site_id":23,"language":22,"slug":32,"title":13,"keywords":33,"description":14,"schema_data":34,"social_meta":84,"head_meta":86,"extra_data":88,"updated_unix":27},"arsenal-antimicrobial-resistance-prediction-by-a-machine-learning-method","",{"@graph":35,"@context":83},[36,52,66],{"@type":37,"itemListElement":38},"BreadcrumbList",[39,43,47,49],{"item":40,"name":41,"@type":42,"position":20},"https://docshare.wps.com","Home","ListItem",{"item":44,"name":45,"@type":42,"position":46},"https://docshare.wps.com/document/","Document",2,{"item":48,"name":12,"@type":42,"position":28},"https://docshare.wps.com/document/research-report/",{"item":50,"name":13,"@type":42,"position":51},"https://docshare.wps.com/document/arsenal-antimicrobial-resistance-prediction-by-a-machine-learning-method/119363/",4,{"url":50,"name":13,"@type":53,"author":54,"headline":13,"publisher":56,"fileFormat":59,"inLanguage":22,"description":14,"dateModified":60,"datePublished":60,"encodingFormat":59,"isAccessibleForFree":61,"interactionStatistic":62},"DigitalDocument",{"name":9,"@type":55},"Person",{"url":40,"name":57,"@type":58},"DocShare","Organization","application/pdf","2026-08-03",true,{"@type":63,"interactionType":64,"userInteractionCount":4},"InteractionCounter",{"@type":65},"ViewAction",{"@type":67,"mainEntity":68},"FAQPage",[69,75,79],{"name":70,"@type":71,"acceptedAnswer":72},"Why is antibiotic susceptibility testing (AST) insufficient for antimicrobial resistance studies?","Question",{"text":73,"@type":74},"AST based on bacterial culture is slow (24–72 h) and applies only to cultivable bacteria, which limits the analysis of resistance emergence and spread in diverse microbial communities with many uncultured organisms.","Answer",{"name":76,"@type":71,"acceptedAnswer":77},"What data does ARSENAL use to predict resistance?",{"text":78,"@type":74},"It uses 1,312 newly sequenced Streptococcus pneumoniae genomes along with measured MIC values from antibiotic susceptibility testing, mainly for six beta-lactam antibiotics, plus the SNP and gene abundance profiles derived from the genomic data.",{"name":80,"@type":71,"acceptedAnswer":81},"How does ARSENAL reduce the genomic feature space for machine learning?",{"text":82,"@type":74},"It applies low-variance filtering, then clusters SNPs using Spearman correlation within orthogroups and across genes to represent each cluster by a single representative SNP, followed by univariate cluster ranking and optimized multivariate feature selection to choose the most predictive feature combinations.","https://schema.org",{"og:url":50,"og:type":85,"og:title":13,"og:site_name":57,"og:description":14},"article",{"robots":87,"canonical":50},"index,follow",{"doc_id":7,"site_id":23},{"code":4,"msg":5,"data":90},[91,95,99,103,108,113,118,121,126,129,133],{"id":20,"doc_module":4,"doc_module_name":45,"category_name":92,"show_sort_weight":93,"slug":94},"Story & Novel",90,"story-novel",{"id":46,"doc_module":4,"doc_module_name":45,"category_name":96,"show_sort_weight":97,"slug":98},"Literature",80,"literature",{"id":51,"doc_module":4,"doc_module_name":45,"category_name":100,"show_sort_weight":101,"slug":102},"Exam",70,"exam",{"id":104,"doc_module":4,"doc_module_name":45,"category_name":105,"show_sort_weight":106,"slug":107},5,"Comic",60,"comic",{"id":109,"doc_module":4,"doc_module_name":45,"category_name":110,"show_sort_weight":111,"slug":112},6,"Technology",50,"technology",{"id":114,"doc_module":4,"doc_module_name":45,"category_name":115,"show_sort_weight":116,"slug":117},7,"Healthcare",40,"healthcare",{"id":11,"doc_module":4,"doc_module_name":45,"category_name":12,"show_sort_weight":119,"slug":120},30,"research-report",{"id":122,"doc_module":4,"doc_module_name":45,"category_name":123,"show_sort_weight":124,"slug":125},9,"Religion & Spirituality",20,"religion-spirituality",{"id":124,"doc_module":4,"doc_module_name":45,"category_name":127,"show_sort_weight":124,"slug":128},"World Cup","world-cup",{"id":130,"doc_module":4,"doc_module_name":45,"category_name":131,"show_sort_weight":130,"slug":132},10,"Lifestyle","lifestyle",{"id":134,"doc_module":4,"doc_module_name":45,"category_name":135,"show_sort_weight":104,"slug":136},19,"General","general"]