gemini_score: Scoring Combination Effect

In sellerslab/gemini: GEMINI: Variational inference approach to infer genetic interactions from pairwise CRISPR screens

Scoring Combination Effect

Description

Score genetic interactions from a gemini.model and produce a gemini.score object, and generate p-values and FDRs if negative control pairs (nc_pairs) are available.

Usage

gemini_score(
  Model,
  pc_gene = NA,
  pc_threshold = NULL,
  pc_weight = 0.5,
  nc_pairs = NA
)

Arguments

Model	an object of class gemini.model
pc_gene	a character vector of any length naming genes to use as positive controls
pc_threshold	a numeric value to indicate the LFC corresponding to a positive control, if no pc_gene is specified.
pc_weight	a weighting applied to the positive control (pc_gene) to filter genes whose individual phenotype is more lethal than pc_weight*y(pc_gene).
nc_pairs	a set of non-interacting gene pairs to define statistical significance.

Value

An object of class gemini.score containing score values for strong interactions and sensitive lethality and recovery, and if nc_pairs is specified, statistical significance for each scoring type.

Examples

data("Model", package = "gemini")
Score <- gemini_score(Model, pc_gene = "EEF2")

sellerslab/gemini documentation built on Dec. 5, 2022, 8:56 a.m.