daa-interpret

To interpret differential abundance analysis results, the Skill parses a results TSV containing standard columns like feature_id, estimate, p_value, and q_value to extract top significant features and generate a concise structured report. It automatically determines the analysis method to present effect sizes appropriately.

A DAA results TSV requires standard columns including feature_id, coefficient, estimate, std_error, statistic, p_value, q_value, prevalence, mean_abundance, prevalence_tier, and confidence. The Skill parses these defined columns to compute effect sizes and surface top features with direction.

Effect size calculation works by identifying the analysis method based on the result structure, then computing and presenting effect sizes in the appropriate scale for that specific method. This ensures the generated interpretation accurately reflects the magnitude and direction of feature abundance changes.

Yes, you can check for potential compositional artifacts. The generated interpretation report includes quality checks for compositional artifacts alongside the top significant features, their direction, and confidence levels, enabling downstream validation of the differential abundance analysis outputs.

The best way to translate DAA outputs into actionable insights is applying a structured workflow that computes effect sizes, determines the method used, and generates a concise report listing top significant features with direction and confidence. This highlights quality checks to enable downstream validation and communication.

You do not need to specify a file path strictly. You can provide a results TSV path, but the Skill can also apply to the most recent results file to extract standard columns and generate the interpretation report immediately without explicit path input.