knitr::opts_chunk$set( collapse = TRUE, comment = "#>", fig.path = "man/figures/README-", out.width = "100%", dev.args = list(png = list(type = "cairo")), echo = TRUE, dpi = 150, fig.align = "center")
The (future, long term, eventual) goal of aqua is to provide tools that help with different assay qualification and validation project analysis and reporting.
You can install aqua. I hope. You're on your own.
devtools::install_github("cgtc/aqua") # fingers crossed
library(aqua) sd(c(2,3,2)) # boring sd.u(c(2,3,2)) # amazing
sd.u
produces a very unbiased standard deviation estimate. Makes every small sample size data look horrible. The sad but true reality. Uses a chi-distributed correction factor.
library(lme4) funky_model <- lmer(total.fruits ~ 1 + (1|amd) + (1|status), data = Arabidopsis) get_intermed_levels(funky_model) |> knitr::kable(escape = F) |> kable_style()
Boom. Clearly the contribution of amd
and status
to variations in fruit.growth
(in a vacuum) in this model I just invented that makes no sense is what it says there.
You can get repeatability and overall intermediate precision / reproducibility of an assay if you turned it into an lme4 random model with the only fixed term being an intercept. Just describe your model like 1 + (1|Lab/Operator/Day) + (1|SampleType)
or whatever your assay design structure is and fire it up.
You can also do the whole TCID50 assay calculations. There's a lot of functions for that that tend to even work! Does anyone use TCID50 anymore? Hello?
Currently 0.2 ALPHA
Add the following code to your website.
For more information on customizing the embed code, read Embedding Snippets.