Multi-Omics Biomarkers for Early Tuberculosis Detection in Type 2 Diabetes Patients

A prospective cohort study of 198 participants (71 healthy controls, 67 with type 2 diabetes, 60 with T2DM-TB comorbidity) conducted transcriptomic and proteomic profiling to identify immune-related biomarkers. The research identified 13 cross-platform biomarkers and developed 11 logistic regression models with reported AUC values between 0.93 and 0.99. The dataset, authored by Zhaoyang Ye and last updated on 2026-03-18, is shared under a CC-BY-4.0 license on figshare.

Use Cases

• Training diagnostic classifiers for tuberculosis in diabetic patients based on the 13 identified immune-related biomarkers.
• Analyzing immune-metabolic regulatory networks based on the described Toll-like receptor, NF-κB, and JAK-STAT pathways.
• Validating competing endogenous RNA (ceRNA) network interactions, such as novel-miR-109 targeting FPR1, LILRB3, and MSTRG.4908.1.
• Studying immune cell infiltration patterns in T2DM-TB comorbidity based on the described 'pro-inflammatory-suppressive-reconstructive' imbalance.

Strengths

• Dataset is derived from a prospective cohort study with 198 participants across three distinct clinical groups.
• Research validated 11 diagnostic models with reported high performance (AUC 0.93-0.99) and used 5-fold cross-validation.
• Identified a specific set of 13 cross-platform biomarkers, including mRNAs, miRNAs, lncRNAs, and proteins.

Limitations

• Column-level documentation is absent; field semantics must be inferred after download.
• Row count is unknown, which may limit suitability assessment.
• The primary data file is a 21.9 KB DOCX document, suggesting the dataset is likely a summary or results table rather than the raw underlying data.

Provenance

Source

Zhaoyang Ye via figshare

Collection Method

Prospective cohort study involving whole-transcriptome and plasma-proteome profiling, with external validation using GEO datasets and RT-qPCR.

Freshness

Last updated 2026-03-18 07:43:06