Aggregate uploaded results by removing timeseries_id

Drops the timeseries_id column and collapses rows by taking the maximum fdr_corrected_pvalue_logp per group of remaining columns. This dramatically reduces the size of the results data frame for large uploads where many timeseries map to the same parameter/site/feature.

Usage

aggregate_results(results_df)

Arguments

results_df

Data frame read from the uploaded results file.

Value

A data frame without timeseries_id, with one row per unique combination of the remaining columns.