{"id":32596350,"url":"https://github.com/schlosslab/severe-cdi","last_synced_at":"2025-10-30T04:59:19.376Z","repository":{"id":62614153,"uuid":"352146007","full_name":"SchlossLab/severe-CDI","owner":"SchlossLab","description":"Predicting CDI severity from OTUs","archived":false,"fork":false,"pushed_at":"2023-07-12T14:15:42.000Z","size":288021,"stargazers_count":2,"open_issues_count":0,"forks_count":1,"subscribers_count":4,"default_branch":"main","last_synced_at":"2024-05-02T02:29:40.879Z","etag":null,"topics":[],"latest_commit_sha":null,"homepage":"http://www.schlosslab.org/severe-CDI/","language":"R","has_issues":true,"has_wiki":null,"has_pages":null,"mirror_url":null,"source_name":null,"license":"other","status":null,"scm":"git","pull_requests_enabled":true,"icon_url":"https://github.com/SchlossLab.png","metadata":{"files":{"readme":"README.md","changelog":null,"contributing":null,"funding":null,"license":"LICENSE.md","code_of_conduct":null,"threat_model":null,"audit":null,"citation":null,"codeowners":null,"security":null,"support":null}},"created_at":"2021-03-27T18:21:34.000Z","updated_at":"2023-06-01T14:56:15.000Z","dependencies_parsed_at":"2022-11-03T23:22:01.301Z","dependency_job_id":null,"html_url":"https://github.com/SchlossLab/severe-CDI","commit_stats":null,"previous_names":[],"tags_count":0,"template":false,"template_full_name":null,"purl":"pkg:github/SchlossLab/severe-CDI","repository_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/SchlossLab%2Fsevere-CDI","tags_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/SchlossLab%2Fsevere-CDI/tags","releases_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/SchlossLab%2Fsevere-CDI/releases","manifests_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/SchlossLab%2Fsevere-CDI/manifests","owner_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/owners/SchlossLab","download_url":"https://codeload.github.com/SchlossLab/severe-CDI/tar.gz/refs/heads/main","sbom_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/SchlossLab%2Fsevere-CDI/sbom","scorecard":null,"host":{"name":"GitHub","url":"https://github.com","kind":"github","repositories_count":281748723,"owners_count":26554822,"icon_url":"https://github.com/github.png","version":null,"created_at":"2022-05-30T11:31:42.601Z","updated_at":"2022-07-04T15:15:14.044Z","status":"online","status_checked_at":"2025-10-30T02:00:06.501Z","response_time":61,"last_error":null,"robots_txt_status":"success","robots_txt_updated_at":"2025-07-24T06:49:26.215Z","robots_txt_url":"https://github.com/robots.txt","online":true,"can_crawl_api":true,"host_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub","repositories_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories","repository_names_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repository_names","owners_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/owners"}},"keywords":[],"created_at":"2025-10-30T04:58:33.112Z","updated_at":"2025-10-30T04:59:19.360Z","avatar_url":"https://github.com/SchlossLab.png","language":"R","funding_links":[],"categories":[],"sub_categories":[],"readme":"\n# severe-CDI\n\n#### Predicting Severity of *C. difficile* Infections from the Taxonomic Composition of the Gut Microbiome\n\nKelly L. Sovacool, Sarah E. Tomkovich, Megan L. Coden, Jenna Wiens,\nVincent B. Young, Krishna Rao, Patrick D. Schloss\n\n## Abstract\n\n*Clostridioides difficile* infection (CDI) can lead to adverse outcomes\nincluding ICU admission, colectomy, and death. The composition of the\ngut microbiome plays an important role in determining colonization\nresistance and clearance upon exposure to *C. difficile*. We\ninvestigated whether machine learning (ML) models trained on 16S rRNA\ngene amplicon sequences from gut microbiota extracted from 1,277 patient\nstool samples on the day of CDI diagnosis could predict which CDI cases\nled to severe outcomes. We then trained ML models to predict CDI\nseverity on OTU relative abundances according to four different severity\ndefinitions: the IDSA severity score on the day of diagnosis, all-cause\nadverse outcomes within 30 days, adverse outcomes confirmed as\nattributable to CDI via chart review, and a pragmatic definition that\nuses the attributable definition when available and otherwise uses the\nall-cause definition. The models predicting pragmatic severity performed\nbest, suggesting that while chart review is valuable to verify the cause\nof complications, including as many samples as possible is indispensable\nfor training performant models on imbalanced datasets. Permutation\nimportance identified *Enterococcus* as the most important OTU for model\nperformance, and increased relative abundance of *Enterococcus* was\nassociated with severe outcomes. Finally, we evaluated the potential\nclinical value of the OTU-based models and found similar performance\ncompared to prior models based on Electronic Health Records. The modest\nperformance of the OTU-based models represents a step toward the goal of\ndeploying models to inform clinical decisions and ultimately improve CDI\noutcomes.\n\n## Manuscript\n\n- [Quarto](paper/paper.qmd)\n- [PDF](paper/paper.pdf)\n- [Markdown](paper/paper-gfm.md)\n\n### Word count\n\n- abstract: 242\n- body: 4885\n","project_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Fschlosslab%2Fsevere-cdi","html_url":"https://awesome.ecosyste.ms/projects/github.com%2Fschlosslab%2Fsevere-cdi","lists_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Fschlosslab%2Fsevere-cdi/lists"}