Gut Microbiome Profiles for Early Lung Adenocarcinoma Diagnosis

175 early-stage lung adenocarcinoma patients and 107 healthy controls form the core training cohort for this gut microbiome study. Weidong Zhang published this dataset on figshare in May 2026, aiming to characterize microbial signatures for non-invasive cancer detection. The dataset includes results validated across multiple independent cohorts from Jiangsu and Hainan provinces.

Use Cases

• Train random forest classifiers for lung cancer detection based on gut microbiome operational taxonomic units.
• Analyze microbial diversity and composition differences between cancer patients and healthy controls.
• Validate diagnostic biomarker models across geographically distinct patient cohorts.
• Compare the performance of multiple machine learning models for medical diagnostic tasks.

Strengths

• Training cohort includes 175 early-stage LUAD patients and 107 healthy controls.
• Diagnostic model achieved an AUC of 0.998 in the training cohort and was validated in two external cohorts.
• Data is licensed under CC-BY-4.0, facilitating open reuse and sharing.

Limitations

• Column-level documentation is absent; field semantics must be inferred after download.
• Row count is unknown, which may limit suitability assessment.
• Data may reflect geographic bias inherent to cohorts from Jiangsu and Hainan.

Provenance

Source

Weidong Zhang

Collection Method

Prospective, randomized, multi-center study characterizing the gut microbiome.

Freshness

Last updated 2026-05-14 05:47:15; freshness should be verified.

Geography

Cohorts from Jiangsu and Hainan provinces.