Skip to contents

Perform Fisher's exact test on a single transcript in ninetails output

Source: R/ninetails_statistic.R
nonA_fisher.Rd

Runs Fisher's exact test for testing the null hypothesis of independence of rows and columns in a 2x2 contingency table representing a given transcript in ninetails output data. This is a wrapper for fisher.test with additional data wrangling features.

Usage

nonA_fisher(
  ninetails_data,
  grouping_factor,
  base,
  min_reads = 0,
  transcript_id_column = NA
)

Arguments

ninetails_data

Data frame. The output of merge_nonA_tables (merged tabular output containing read classification and non-A position data).

grouping_factor

Character string. The name of the factor variable defining groups/conditions (must have exactly 2 levels).

base

Character string. Letter representing the non-A nucleotide for which statistics are computed. Accepted values: "C", "G", "U", or "all" (for all non-A residues combined). Default: "C".

min_reads

Numeric. Minimum number of reads representing a given transcript to include it in the analysis. Default: 0.

transcript_id_column

Character string. Name of the column in which transcript identifiers are stored. Default: NA.

Value

A tibble with results for the given transcript, containing:

p.value

Numeric. Raw p-value from Fisher's exact test, or NA if conditions were not met.

stats_code

Character. Status code describing whether the test conditions were met (see stat_codes_list for definitions).

Details

The function is suitable only for pairwise comparisons (2x2 contingency tables), where two conditions (e.g. WT vs KO) are compared at once. The user can set a cutoff number of reads required for the analysis.

This function is intended to be called internally by calculate_fisher and is not typically used standalone.

Acknowledgements

Inspired by the Nanotail package written and maintained by Pawel Krawczyk (smaegol): https://github.com/LRB-IIMCB/nanotail/blob/dev/R/polya_stats.R. Many thanks to the developer of the original source code.

See also

calculate_fisher for the per-transcript wrapper with multiple-testing correction, merge_nonA_tables for preparing the input, summarize_nonA for contingency table computation

Examples

if (FALSE) { # \dontrun{

test <- ninetails::nonA_fisher(
  ninetails_data = merged_nonA_tables,
  grouping_factor = "sample_name",
  base = "C",
  min_reads = 100
)

} # }