Identification of two robust subclasses of sepsis with both prognostic and therapeutic values based on machine learning analysis
Source check: Source checked. Not independent scientific confirmation.
Finding: Two transcriptomic subtypes were identified in 1,692 sepsis samples; cluster B had higher observed mortality in the cohort with survival data (60.7% vs 39.3%; p=.044). In a separate dataset, risk scores were associated with response to early supportive therapy (p=.027; AUC .663, 95% CI .516-.789). The two-cluster structure and expression/pathway patterns were assessed in five independent datasets (617 samples). The seven-gene prognostic score used separate training (980) and validation (760) sets. Therapy-response association was assessed in GSE110487 only and is modest/retrospective; it does not establish a treatment effect. The therapy-response association is retrospective and does not establish treatment efficacy or causality.
Technology: K-means clustering of transcriptomic profiles defined the two sepsis subtypes; LASSO and logistic regression produced the seven-gene risk score linked to clinical outcomes.
Publication: Peer-reviewed paper · Test context: retrospective or benchmark. Catalog categories, not an evidence grade.
- AI contribution
- K-means clustering of transcriptomic profiles defined the two sepsis subtypes; LASSO and logistic regression produced the seven-gene risk score linked to clinical outcomes.
- Test and result
- Two transcriptomic subtypes were identified in 1,692 sepsis samples; cluster B had higher observed mortality in the cohort with survival data (60.7% vs 39.3%; p=.044). In a separate dataset, risk scores were associated with response to early supportive therapy (p=.027; AUC .663, 95% CI .516-.789).
- Limitation / still unknown
- The two-cluster structure and expression/pathway patterns were assessed in five independent datasets (617 samples). The seven-gene prognostic score used separate training (980) and validation (760) sets. Therapy-response association was assessed in GSE110487 only and is modest/retrospective; it does not establish a treatment effect.
Evidence sources and annotation context
Evidence context (descriptions follow the cited sources; test type is a catalog classification):
Evidence sources: https://www.frontiersin.org/journals/immunology/articles/10.3389/fimmu.2022.1040286/full
Source checked · editorial details
The source page was checked against this entry. Independent confirmation is not assessed by this label.
Source type: paper. Paper: not linked in this catalog; this does not mean no paper exists.
Review date: 22 Nov 2026. Review scope: Primary source full text Methods and Results; version/date checked against bibliographic record..
No follow-up recorded; this is not a claim that no follow-up exists. Follow-up review scope and date are not recorded. No source-supported relationships recorded.
Editorial assessment: Impact 3/5 · Validation 2/5 · Novelty 3/5. Impact and novelty are editorial judgments; validation follows the annotated test type where one exists. Not an evidence grade. Scoring criteria.