Gut Microbiome Pathways Predicting Immunotherapy Response in Lung Cancer

77 pretreatment fecal samples from Japanese patients with non-small cell lung cancer receiving immune checkpoint inhibitors, profiled via 16S rRNA sequencing. The dataset, authored by Feifei Wei and shared under CC-BY-4.0, compares taxonomic versus functional microbiome features for predicting clinical outcomes. It includes a four-pathway signature identified as the strongest predictor of treatment response.

Use Cases

• Train machine learning models to predict immune checkpoint inhibitor response based on MetaCyc pathway features.
• Compare the predictive power of taxonomic profiles versus functional pathway features for clinical outcomes.
• Interpret model predictions using SHAP analysis to identify key microbial metabolic drivers like nitrogen metabolism pathways.
• Validate the performance of a specific four-pathway signature comprising urea cycle and short-chain fatty acid biosynthesis pathways.

Strengths

• Includes data from 77 patients, providing a specific cohort size for analysis.
• Features a validated four-pathway predictive signature identified through machine learning.
• Uses nested internal and external validation and SHAP analysis for model robustness and interpretability.

Limitations

• Row count is unknown, which may limit suitability assessment.
• Column-level documentation is absent; field semantics must be inferred after download.
• Data is from a specific cohort of 77 Japanese patients, which may limit generalizability.

Provenance

Source

figshare

Collection Method

16S rRNA sequencing of pretreatment fecal samples from a clinical cohort.

Freshness

Last updated 2026-05-15 04:21:48; freshness should be verified.

Geography

Japan