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

Supplementary file 5 contains metabolomic data from a study comparing intrinsic and idiosyncratic drug-induced liver injury (DILI). The dataset, authored by Xianni Wei and last updated in March 2026, includes results from high-performance chemical isotope labeling liquid chromatography–mass spectrometry (HP-CIL LC-MS) analysis of 44 serum samples. It was used to identify four differential metabolites and develop machine learning diagnostic models with AUC values greater than 0.8.

Use Cases

• Training machine learning classifiers to distinguish intrinsic from idiosyncratic DILI based on serum metabolite profiles.
• Identifying and validating candidate biomarkers like Alanyl-Glycine and N2-Acetyl-L-Cystathionine for liver injury subtypes.
• Analyzing metabolic pathway disruptions in amino acid metabolism, including tryptophan and tyrosine pathways.
• Benchmarking the performance of different machine learning models (e.g., multivariate regression) for clinical metabolomics.

Strengths

• Includes data from 44 clinically classified DILI cases (17 intrinsic, 27 idiosyncratic).
• Machine learning models achieved high diagnostic performance, with a holdout validation AUC of 0.935.
• Metabolomic profiling was performed using high-performance chemical isotope labeling liquid chromatography–mass spectrometry (HP-CIL LC-MS).
• Identified four specific differential metabolites with potential biomarker utility.

Limitations

• Row count and column-level documentation are absent; field semantics must be inferred after download.
• The sample size of 44 cases is relatively small for broad model generalization.
• Data may reflect bias inherent to the specific clinical cohort and collection methodology described.

Provenance

Source

Xianni Wei via figshare.

Collection Method

Serum samples analyzed using HP-CIL LC-MS; cases classified per EASL and NEJM guidelines.

Freshness

Last updated 2026-03-18 07:36:41.