Urinary Protein Biomarkers for Prediabetes Diagnosis and Risk Stratification

Wen-Ling Liao's dataset contains label-free proteomics data from urine samples used to discover biomarkers for prediabetes. The discovery phase included 43 control and 58 prediabetes participants, with validation performed on an expanded set of 91 control and 68 prediabetes subjects. The data identifies alpha-1-acid glycoprotein 1 (AGP1) and zinc-α2-glycoprotein (ZAG) as candidate biomarkers.

Use Cases

• Validate candidate urinary biomarkers (AGP1 and ZAG) for prediabetes based on the proteomic analysis described.
• Compare the discriminative performance of protein biomarkers against fasting blood glucose based on AUC values reported.
• Develop machine learning models for early risk stratification based on combined protein levels, age, and sex.
• Investigate early metabolic alterations associated with prediabetes progression using the identified protein signatures.

Strengths

• Discovery cohort includes 101 participants (43 control, 58 prediabetes).
• Validation cohort includes 159 participants (91 control, 68 prediabetes).
• Protein quantification performed using two independent software platforms for analytical robustness.
• Candidate biomarkers AGP1 and ZAG showed high discriminative performance (AUCs of 0.936 and 0.926).

Limitations

• Row count is unknown, which may limit suitability assessment.
• Column-level documentation is absent; field semantics must be inferred after download.
• Further validation in independent cohorts is warranted, as noted in the description.

Provenance

Source

figshare

Collection Method

Label-free nanoLC-MS/MS-based proteomics applied to urine samples.

Freshness

Last updated 2026-05-05 15:45:52; freshness should be verified.