[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"detail-sidebar-cat-0-en-105":3,"doc-seo-340936-105":59,"doc-detail-340936-en":130},{"code":4,"msg":5,"data":6},0,"success",[7,13,18,23,28,33,38,43,48,51,55],{"id":8,"doc_module":4,"doc_module_name":9,"category_name":10,"show_sort_weight":11,"slug":12},1,"Document","Story & Novel",90,"story-novel",{"id":14,"doc_module":4,"doc_module_name":9,"category_name":15,"show_sort_weight":16,"slug":17},2,"Literature",80,"literature",{"id":19,"doc_module":4,"doc_module_name":9,"category_name":20,"show_sort_weight":21,"slug":22},4,"Exam",70,"exam",{"id":24,"doc_module":4,"doc_module_name":9,"category_name":25,"show_sort_weight":26,"slug":27},5,"Comic",60,"comic",{"id":29,"doc_module":4,"doc_module_name":9,"category_name":30,"show_sort_weight":31,"slug":32},6,"Technology",50,"technology",{"id":34,"doc_module":4,"doc_module_name":9,"category_name":35,"show_sort_weight":36,"slug":37},7,"Healthcare",40,"healthcare",{"id":39,"doc_module":4,"doc_module_name":9,"category_name":40,"show_sort_weight":41,"slug":42},8,"Research & Report",30,"research-report",{"id":44,"doc_module":4,"doc_module_name":9,"category_name":45,"show_sort_weight":46,"slug":47},9,"Religion & Spirituality",20,"religion-spirituality",{"id":46,"doc_module":4,"doc_module_name":9,"category_name":49,"show_sort_weight":46,"slug":50},"World Cup","world-cup",{"id":52,"doc_module":4,"doc_module_name":9,"category_name":53,"show_sort_weight":52,"slug":54},10,"Lifestyle","lifestyle",{"id":56,"doc_module":4,"doc_module_name":9,"category_name":57,"show_sort_weight":24,"slug":58},19,"General","general",{"code":4,"msg":60,"data":61},"ok",{"site_id":62,"language":63,"slug":64,"title":65,"keywords":66,"description":67,"schema_data":68,"social_meta":123,"head_meta":125,"extra_data":127,"updated_unix":129},105,"en","language-model-enables-end-to-end-accurate-detection-of-cancer-from-cell-free-dna","Language model enables end-to-end accurate detection of cancer from cell-free DNA","","Affordable Cancer Interception and Diagnostics (ACID) presents an autoregressive language model for cancer diagnosis using only raw cell-free DNA (cfDNA) sequencing reads. ACID is pretrained by concatenating raw sequencing reads with diagnostic labels, then benchmarked against three existing methods. On whole-genome sequencing test data, ACID improves cancer classification (AUROC 0.924 vs 0.853; P \u003C0.001) and hepatocellular carcinoma detection (AUROC 0.981 vs 0.917; P \u003C0.001), and reaches high accuracy with 10,000 reads per sample.",{"@graph":69,"@context":122},[70,84,105],{"@type":71,"itemListElement":72},"BreadcrumbList",[73,77,79,82],{"item":74,"name":75,"@type":76,"position":8},"https://docshare.wps.com","Home","ListItem",{"item":78,"name":9,"@type":76,"position":14},"https://docshare.wps.com/document/",{"item":80,"name":40,"@type":76,"position":81},"https://docshare.wps.com/document/research-report/",3,{"item":83,"name":65,"@type":76,"position":19},"https://docshare.wps.com/document/language-model-enables-end-to-end-accurate-detection-of-cancer-from-cell-free-dna/340936/",{"url":83,"name":65,"@type":85,"image":86,"author":91,"headline":65,"publisher":94,"fileFormat":97,"inLanguage":63,"description":67,"dateModified":98,"datePublished":99,"encodingFormat":97,"isAccessibleForFree":100,"interactionStatistic":101},"DigitalDocument",{"url":87,"@type":88,"width":89,"height":90},"https://docshare.wps.com/thumbnails/language-model-enables-end-to-end-accurate-detection-of-cancer-from-cell-free-dna/340936.png","ImageObject",300,407,{"name":92,"@type":93},"Anda","Person",{"url":74,"name":95,"@type":96},"DocShare","Organization","application/pdf","2026-11-11","2026-09-22",true,{"@type":102,"interactionType":103,"userInteractionCount":52},"InteractionCounter",{"@type":104},"ViewAction",{"@type":106,"mainEntity":107},"FAQPage",[108,114,118],{"name":109,"@type":110,"acceptedAnswer":111},"What is ACID and how does it perform cancer diagnosis?","Question",{"text":112,"@type":113},"ACID is an autoregressive language model formulated for cancer diagnosis using only raw cfDNA sequencing reads. It converts raw sequencing reads and diagnostic labels into a language-model pretraining setup and then performs end-to-end classification.","Answer",{"name":115,"@type":110,"acceptedAnswer":116},"How does ACID compare with benchmark methods on whole-genome sequencing data?",{"text":117,"@type":113},"ACID significantly outperforms the best benchmarked method for cancer diagnosis, achieving AUROC 0.924 vs 0.853 (P \u003C0.001). It also improves hepatocellular carcinoma detection, with AUROC 0.981 vs 0.917 (P \u003C0.001).",{"name":119,"@type":110,"acceptedAnswer":120},"How many sequencing reads are needed for ACID to achieve high accuracy?",{"text":121,"@type":113},"ACID can achieve high accuracy with just 10,000 reads per sample, according to the study summary.","https://schema.org",{"og:url":83,"og:type":124,"og:title":65,"og:site_name":95,"og:description":67},"article",{"robots":126,"canonical":83},"index,follow",{"doc_id":128,"site_id":62},340936,1790181814,{"code":4,"msg":5,"data":131},{"doc_id":128,"user_id":132,"nickname":92,"user_avatar":133,"doc_module":4,"category_id":39,"category_name":40,"doc_title":65,"doc_description":67,"doc_content":134,"file_id":135,"file_url":136,"file_type":137,"file_size":138,"view_count":52,"is_deleted":4,"is_public":8,"is_downloadable":8,"audit_status":8,"page_count":52,"language":139,"language_code":63,"site_id":62,"html_lang":63,"table_of_contents":140,"faqs":141,"seo_title":142,"seo_description":67,"update_tm":143,"read_time":144,"preview_page_count":19,"preview_urls":145},962075006959,"https://ap-avatar.wpscdn.com/avatar/e0002397efbe92a78e?_k=1776741047341049297","Briefings in Bioinformatics, 2024, 25(2), 1–10  \n[https://doi.org/10.1093/bib/bbae053](https://doi.org/10.1093/bib/bbae053)  \nProblem Solving Protocol  \nLanguage model enables end-to-end accurate detection of cancer from cell-free DNA  \nHongru Shen†, Jilei Liu†, Kexin Chen and Xiangchun Li   \nCorresponding authors: Kexin Chen, Department of Epidemiology and Biostatistics, Tianjin Medical University Cancer Institute and Hospital, Huanhu Xi Road, Tiyuan Bei, Hexi District, Tianjin 300060, China. Tel.:/Fax: (86)022-23372231; E-mail: [chenkexin@tmu.edu.cn](chenkexin@tmu.edu.cn); Xiangchun Li, Tianjin Cancer Institute, Tianjin Medical University Cancer Institute and Hospital, Huanhu Xi Road, Tiyuan Bei, Hexi District, Tianjin 300060, China. Tel.:/Fax: (86)022-23372231;  \nE-mail: [lixiangchun@tmu.edu.cn](lixiangchun@tmu.edu.cn)  \n†Hongru Shen and Jilei Liu contributed equally to this work.  \nWe present a language model Affordable Cancer Interception and Diagnostics (ACID) that can achieve high classification performance in the diagnosis of cancer exclusively from using raw cfDNA sequencing reads. We formulate ACID as an autoregressive language model. ACID is pretrained with language sentences that are obtained from concatenation of raw sequencing reads and diagnostic labels. We benchmark ACID against three methods. On testing set subjected to whole-genome sequencing, ACID significantly outperforms the best benchmarked method in diagnosis of cancer [Area Under the Receiver Operating Curve (AUROC), 0.924 versus 0.853; P \u003C0.001] and detection of hepatocellular carcinoma (AUROC, 0.981 versus 0.917; P \u003C0.001) . ACID can achieve high accuracy with just 10 000 reads per sample. Meanwhile, ACID achieves the best performance on testing sets that were subjected to bisulfite sequencing compared with benchmarked methods. In summary, we present an affordable, simple yet efficient end-to-end paradigm for cancer detection using raw cfDNA sequencing reads.  \nKeywords: cell-free DNA; cancer detection; generative language model  \nINTRODUCTION  \nCell-free DNA (cfDNA) is fragmented DNA circulating in the blood [1 , 2] through cell necrosis and apoptosis [3 , 4] . cfDNA fragments shed into plasma by tumor cells bear cancer-specific genomic and epigenetic alterations that can be leveraged for early cancer detection [5 , 6] . Tumor-derived cfDNA fragments are shorter in size than hematopoietic-originated background cfDNA fragments [7] . cfDNA fragments exhibit preferential patterns in patients with and without cancer as well as among different cancer types [8 , 9] . The fragmentation patterns of cfDNA have been widely studied in early cancer detection [10–15] . The Circulating Cell-free Genome Atlas (CCGA) is among one of the most ambitious and representative endeavors in investigating the use of cfDNA-based multi-cancer early cancer detection [16–18] .  \nThe size profiles of cfDNA are different in patients with and without cancer in which shorter cfDNA fragments are more prevalent in patients with cancer. The median size of background cfDNA is ∼166 bp [19–21] . Mouliere and colleagues reported that length selection of short cfDNA fragments increases enrichment of circulating tumor DNA in advanced cancer patients [22] . There are several characteristics manifesting unique fragmentation patterns of cfDNA in patient with cancer, including alterations in nucleosome positions [20 , 23] and preference of specific sequence  \nat the end of cfDNA fragments [24] . Whole-genome sequencing (WGS) offers a high-resolution view of the genome. However, it remains challenging to unveil accurate chromosomal aberrations from cfDNA [25–27] . Cristiano and colleagues developed a machine learning approach called DNA evaluation of fragments for early interception (DELFI) [28] by integrating numerous genomic alterations of cfDNA for cancer detection. DELFI incorporates characteristics of cancer-specific profiles with respect to mutation and chromatin organization obtained from ","cbCaitbjWRguDMf6","https://ap.wps.com/l/cbCaitbjWRguDMf6","pdf",1188656,"English","# Introduction\n## cfDNA fragmentation patterns and cancer relevance\n## Prior machine learning and integrative approaches\n## cfDNA end-motif and nuclease activity background\n## Study aim and contributions","[{\"question\":\"What is ACID and how does it perform cancer diagnosis?\",\"answer\":\"ACID is an autoregressive language model formulated for cancer diagnosis using only raw cfDNA sequencing reads. It converts raw sequencing reads and diagnostic labels into a language-model pretraining setup and then performs end-to-end classification.\"},{\"question\":\"How does ACID compare with benchmark methods on whole-genome sequencing data?\",\"answer\":\"ACID significantly outperforms the best benchmarked method for cancer diagnosis, achieving AUROC 0.924 vs 0.853 (P \\u003c0.001). It also improves hepatocellular carcinoma detection, with AUROC 0.981 vs 0.917 (P \\u003c0.001).\"},{\"question\":\"How many sequencing reads are needed for ACID to achieve high accuracy?\",\"answer\":\"ACID can achieve high accuracy with just 10,000 reads per sample, according to the study summary.\"}]","Language model enables end-to-end accurate detection of cancer from cell-free DNA | PDF",1790039609,25,[146,150,154,158],{"page_no":8,"url":147,"width":148,"height":149},"https://res-academy.cache.wpscdn.com/upload/kdocshare/2026-11-16/f573d9490414695d8921b91808c794e3.webp",983,1291,{"page_no":14,"url":151,"width":152,"height":153},"https://res-academy.cache.wpscdn.com/upload/kdocshare/2026-11-16/3438b619b5d101246f6c3da0049088fe.webp",982,1290,{"page_no":81,"url":155,"width":156,"height":157},"https://res-academy.cache.wpscdn.com/upload/kdocshare/2026-11-16/4d42a5f2dc15e53c65eead1375c84167.webp",981,1289,{"page_no":19,"url":159,"width":160,"height":161},"https://res-academy.cache.wpscdn.com/upload/kdocshare/2026-11-16/80bcc942cbd61893df667816b7862d29.webp",980,1287]