In benjbuch/summerrband: Analyse Gel-shift Assays

knitr::opts_chunk$set(
  collapse = TRUE,
  comment = "#>",
  fig.path = "man/figures/README-",
  out.width = "100%"
)

summerrband

summerrband plays for you gel-shift data from "tapes" recorded on Cytiva's ImageQuant TL. Wherever the data comes from, once it's in a table (data.frame), this package gives you also a handle on fitting dissociation constants in various ways.

Installation

Since this package is not part of CRAN, you must install and update the development version from GitHub with:

# install.packages("devtools")
devtools::install_github("benjbuch/summerrband")

Showcases

library(summerrband)

A Song on Import-Me

my_file <- system.file("extdata", "gel_01.txt", package = "summerrband")

This is how a raw IQTL file looks like (truncated):

data.table::fread(my_file)[, 1:14]

This is how the same file will look like after tidy importing. As an extra benefit of calling iqtl_view instead of iqtl_read, we'll play a preview alongside:

iqtl_view(my_file)

Let's assign some metadata along with the tunes:

my_data <- iqtl_meta(my_file, meta_data = list(conc = c(2^seq(10, 0), 1),
                                               protein = "protA",
                                               ligand  = "DNA1"))

head(my_data)

Of course, there is a batch function (iqtl_import_all) too.

Fitting the Tune and Tuning the Fit

Four lines to sing:

library(magrittr)  # for the pipe operator

my_data %>% 
  dplyr::filter(band_id == "band_1") %>% 
  dplyr::group_by(protein, ligand) %>% 
  model_cleanly_groupwise(fit_Kd, formula = vol_frac ~ conc, 
                          newdata = data.frame(conc = 10^seq(0, 3, length.out = 100))) %>% 
  model_display(color = ligand) + ggplot2::facet_wrap(ggplot2::vars(protein))

my_data %>% 
  dplyr::filter(band_id == "band_1") %>% 
  dplyr::group_by(protein, ligand) %>% 
  model_cleanly_groupwise(fit_Kd, formula = vol_frac ~ conc) %>% 
  dplyr::select(tidy) %>% tidyr::unnest(tidy)

We can also employ a model that doesn't assume excess of the non-titrated species R. So, when we provide its intial (total) concentration R0 before the equilibrium is reached:

my_data %>% 
  dplyr::filter(band_id == "band_1") %>% 
  dplyr::group_by(protein, ligand) %>% 
  model_cleanly_groupwise(fit_Kd, formula = vol_frac ~ conc, R0 = 2.0) %>% 
  dplyr::select(tidy) %>% tidyr::unnest(tidy)