Respiratory Microbiome Features for Chicken Age Prediction via Machine Learning

A list of microbial taxa identified as predictive features for chicken age using a machine learning model. The dataset was created by Alejandro Moreno-León and published on figshare under a CC-BY-4.0 license in April 2026. It contains features selected via Recursive Feature Elimination (RFE) and retained only if they persisted in at least 70% of independent Leave-One-Study-Out (LOSO) iterations.

Use Cases

• Training machine learning models for age prediction in poultry based on respiratory microbiome signatures.
• Analyzing the stability and importance of specific microbial taxa across different study cohorts.
• Benchmarking feature selection methods in microbiome studies.
• Investigating the relationship between respiratory microbiota maturation and host development.

Strengths

• Features were selected using a rigorous Recursive Feature Elimination (RFE) process.
• A consensus signature was formed requiring features to persist in at least 70% of independent LOSO iterations, suggesting cross-study stability.
• Data is openly available under a permissive CC-BY-4.0 license.

Limitations

• The dataset is very small (35.0 KB), indicating a limited scope, likely containing only feature names and selection statistics.
• Column-level documentation is absent; field semantics must be inferred after download.
• Row count is unknown, which may limit suitability assessment for model training.

Provenance

Source

figshare, authored by Alejandro Moreno-León.

Collection Method

Features were identified via Recursive Feature Elimination (RFE) across Leave-One Study-Out (LOSO) iterations from a study on respiratory microbiota.

Time Range

The dataset itself does not specify a temporal coverage for the underlying samples.

Freshness

Last updated 2026-04-27 03:51:40; freshness should be verified.

Geography

The dataset does not specify the geographic origin of the chicken samples.