has_signif_negative_cluster: Does a well have a statistically significant number of... In ddpcr: Analysis and Visualization of Droplet Digital PCR in R and on the Web

Description

Classify a well as having a significant negative cluster (eg. a mutant well) or not using a binomial test.

We can call a well as mutant if it is statistically significantly more than 1 mutant drops, then the mutant frequency is 1.4 significantly more than 1 P(x >= 7) = 1 - P(x <= 7) + P(x = 7) = 1 - pbinom(7, 500, .01) + dbinom(7, 500, .01) = 0.237 > 0.01 So not statistically significantly enough, so we say it's a wildtype well. But if there are 5000 drops and 70 mutant drops (same 1.4 with higher absolute numbers), then P(x >= 70) = 1 - pbinom(70, 5000, .01) + dbinom(70, 5000, .01) = 0.004 So this is indeed significant, and this well would be deemed mutant.

Usage

 `1` ```has_signif_negative_cluster(plate, neg, pos) ```

Arguments

 `plate` A ddPCR plate `neg` Number of negative (or mutant) drops `pos` Number of positive (or wildtype) drops

Value

`TRUE` if the number of negative drops is statistically significant, `FALSE` otherwise.

ddpcr documentation built on July 8, 2020, 6:49 p.m.