Serum Metabolomics for Intrinsic and Idiosyncratic Drug-Induced Liver Injury Biomarkers

A metabolomic study by Xianni Wei, published on figshare in March 2026, compares serum samples from 44 patients with drug-induced liver injury. The data, generated via high-performance chemical isotope labeling liquid chromatography–mass spectrometry, identifies four differential metabolites. Machine learning models were used to evaluate these metabolites as potential biomarkers for distinguishing intrinsic (n=17) from idiosyncratic (n=27) DILI types.

Use Cases

• Training machine learning classifiers to distinguish intrinsic from idiosyncratic DILI based on serum metabolite profiles.
• Identifying candidate biomarkers like Alanyl-Glycine and N2-Acetyl-L-Cystathionine for diagnostic assay development.
• Analyzing metabolic pathway disruptions in amino acid metabolism related to drug-induced liver injury.
• Validating diagnostic model performance using reported AUC metrics from ROC analysis.

Strengths

• Includes data from 44 clinically classified DILI cases (17 intrinsic, 27 idiosyncratic).
• Reports high diagnostic performance with machine learning model AUC values exceeding 0.8, including a cross-validated AUC of 0.983.
• Identifies four specific differential metabolites with potential biomarker significance.

Limitations

• Row count and column-level documentation are absent; field semantics must be inferred after download.
• The dataset is small in scale at 299.9 KB, which may limit the complexity of models that can be trained.
• Description metadata is limited; actual data quality and structure require manual inspection after download.

Provenance

Source

figshare, author Xianni Wei.

Collection Method

Serum samples analyzed using high-performance chemical isotope labeling liquid chromatography–mass spectrometry (HP-CIL LC-MS).

Freshness

Last updated 2026-03-18 07:36:43; freshness should be verified.