rm_outlier: Identify and Remove Outliers Based on Bonferroni-Holm...

In Qploidy: Estimation of Ploidy and Detection of Aneuploidy Using Genotyping Data

Identify and Remove Outliers Based on Bonferroni-Holm Adjusted P-values

Description

This function detects and removes outlier observations from a vector of 'theta' values using externally studentized residuals and the Bonferroni-Holm adjustment for multiple testing. It is typically used during genotype cluster center estimation to clean noisy values.

Usage

rm_outlier(data, alpha = 0.05)

Arguments

data	A data.frame containing a 'theta' column. This is usually a subset of the full dataset, representing samples within a single genotype class.
alpha	Significance level for identifying outliers (default is '0.05'). Observations with adjusted p-values below this threshold will be removed.

Details

The method fits a constant model ('theta ~ 1') and computes standardized residuals. Observations with significant deviation are flagged using the Bonferroni-Holm procedure and removed if their adjusted p-value is below the defined 'alpha' threshold.

This function was originally developed by **Kaio Olympio** and incorporated into the Qploidy workflow.

Value

A data.frame containing only the non-outlier observations from the input. If fewer than two non-NA 'theta' values are present or if all values are identical, the input is returned unmodified.

Author(s)

Kaio Olympio

Qploidy documentation built on June 8, 2025, 10 a.m.