cumulative_sum_beta_distribution(df, cutoff=0.7, verbose=False, save=False, silent=False)¶
Attempts to filter outlier samples based on the cumulative area under the curve exceeding a reasonable value (cutoff). This method only works on poor quality samples that are better identified using ControlReporter summary (CLI: ‘methylcheck controls’)
- DataFrame – wide format (probes in columns, samples in rows) cutoff (default 0.7) silent – suppresses figure, so justs returns transformed data if False. if save==True: saves figure to disk.
- dataframe with subjects removed that exceed cutoff value.