| Title: | A toolkit for analyzing protein quantification results |
|---|---|
| Description: | What the package does (one paragraph). |
| Authors: | Kai Aragaki [aut, cre] |
| Maintainer: | Kai Aragaki <[email protected]> |
| License: | MIT + file LICENSE |
| Version: | 0.1.0 |
| Built: | 2024-11-03 06:08:10 UTC |
| Source: | https://github.com/KaiAragaki/qp |
Takes an absorbance and converts it to a hexidecimal color. For the default qp_pal palette, this should provide a color that approximates real life color at the given absorbance.
abs_to_col(abs, pal)abs_to_col(abs, pal)
abs |
Numeric. Absorbances. |
pal |
Character. A vector of hexidecimal colors. |
The absorbances have typical baseline absorbance (~ 0.07) removed, and then an index is calculated with a logistic curve of maximum 100 and a center of 0.15.
Character. Hexidecimal colors corresponding to absorbances.
Absorbances from a BCA protein quantification in a data.frame
absorbancesabsorbances
absorbancesA data.frame with 96 rows and 5 columns:;
The row of the 96 well plate, where 1 refers to the top row.
The column of the 96 well plate, where 1 refers to the left column.
The absorbance of the contents of the well at 562nm.
Denotes whether the sample is a standard or an unknown (sample).
Denotes individual standards/samples, where each gets its own index.
Calculate dilution from known concentrations
dilute(c1, c2 = min(c1), v2, round_for_pipettes = TRUE)dilute(c1, c2 = min(c1), v2, round_for_pipettes = TRUE)
c1 |
Numeric. Initial concentration of sample. |
c2 |
Numeric. Target concentration of sample. |
v2 |
Numeric. Target final volume of sample. If |
round_for_pipettes |
Logical. If TRUE, rounds values to the accuracy of
standard pipettes using |
a data.frame, with sample_to_add as the volume of sample to add,
and add_to as the volume to dilute the sample into.
dilute(203, 70, 10) dilute(203, 70, 10, round_for_pipettes = FALSE) # Vectorized: dilute(c(8, 10, 12), c(4, 5, 6), c(7, 8, 9))dilute(203, 70, 10) dilute(203, 70, 10, round_for_pipettes = FALSE) # Vectorized: dilute(c(8, 10, 12), c(4, 5, 6), c(7, 8, 9))
Round volume to be pipette-compatible
make_pipette_vol(x)make_pipette_vol(x)
x |
Numeric. Volume to be rounded |
Numeric. Rounded volume.
make_pipette_vol(104.13398) make_pipette_vol(15.3331) make_pipette_vol(9.9211) # Vectorized: make_pipette_vol(c(104.13398, 15.3331, 9.9211, NA, -100.1))make_pipette_vol(104.13398) make_pipette_vol(15.3331) make_pipette_vol(9.9211) # Vectorized: make_pipette_vol(c(104.13398, 15.3331, 9.9211, NA, -100.1))
Quantify protein concentration from a MicroBCA assay
qp( x, replicate_orientation = c("h", "v"), sample_names = NULL, remove_empty = TRUE, ignore_outliers = c("all", "samples", "standards", "none"), standard_scale = c(0, 2^((2:7) - 5)), n_replicates = 3, wavelength = 562 )qp( x, replicate_orientation = c("h", "v"), sample_names = NULL, remove_empty = TRUE, ignore_outliers = c("all", "samples", "standards", "none"), standard_scale = c(0, 2^((2:7) - 5)), n_replicates = 3, wavelength = 562 )
x |
A |
replicate_orientation |
Either 'h' or 'v' - see Details. |
sample_names |
Optional character vector of sample names. |
remove_empty |
Should wells that have less absorbance than the lowest standard be dropped? |
ignore_outliers |
Character. From which group - samples or standards - should outliers be detected and removed? |
standard_scale |
Numeric. Known concentrations of standards, in the order they appear. |
n_replicates |
Numeric. The number of techinical replicates. |
wavelength |
Numeric. The wavelength absorbance was captured. |
If x is a spectramax, the standards must start in the upper left
corner in the order dictated by standard_scale. Whether this is from from
left to right or top to bottom can be specified in replicate_orientation.
Note that replicate_orientation specified the direction that REPLICATES
lie, NOT the direction the samples flow (which will be perpendicular to
the replicates).
Note: replicate_orientation, n_replicates, and wavelength will
be silently ignored if x is not a spectramax or path to a
spectramax
a tibble
data <- system.file("extdata", "absorbances.txt", package = "qp") qp(data, replicate_orientation = "h")data <- system.file("extdata", "absorbances.txt", package = "qp") qp(data, replicate_orientation = "h")
Add sample names
qp_add_names(x, ...) ## S3 method for class 'list' qp_add_names(x, sample_names = NULL, ...) ## S3 method for class 'data.frame' qp_add_names(x, sample_names = NULL, ...)qp_add_names(x, ...) ## S3 method for class 'list' qp_add_names(x, sample_names = NULL, ...) ## S3 method for class 'data.frame' qp_add_names(x, sample_names = NULL, ...)
x |
A |
... |
Unused |
sample_names |
Optional character vector. If NULL, uses sample index. In a standard workflow, the index is the order the sample appears in the plate |
df <- expand.grid( index = c(1, 1, 2, 2, 2, 3), sample_type = c("standard", "unknown") ) df # You don't get to name standards: qp_add_names(df, c("a", "b", "c")) # If there aren't enough names, will use index qp_add_names(df, c("a", "b")) # No names provided will use index by default qp_add_names(df)df <- expand.grid( index = c(1, 1, 2, 2, 2, 3), sample_type = c("standard", "unknown") ) df # You don't get to name standards: qp_add_names(df, c("a", "b", "c")) # If there aren't enough names, will use index qp_add_names(df, c("a", "b")) # No names provided will use index by default qp_add_names(df)
Add known concentrations of protein to standard samples
qp_add_std_conc(x, standard_scale = c(0, 2^((2:7) - 5)), ...) ## S3 method for class 'data.frame' qp_add_std_conc(x, standard_scale = c(0, 2^((2:7) - 5)), ...) ## S3 method for class 'list' qp_add_std_conc(x, standard_scale = c(0, 2^((2:7) - 5)), ...)qp_add_std_conc(x, standard_scale = c(0, 2^((2:7) - 5)), ...) ## S3 method for class 'data.frame' qp_add_std_conc(x, standard_scale = c(0, 2^((2:7) - 5)), ...) ## S3 method for class 'list' qp_add_std_conc(x, standard_scale = c(0, 2^((2:7) - 5)), ...)
x |
A |
standard_scale |
A numeric vector giving the concentrations of the standards. The units are arbitrary, but will determine the units of the output concentrations. |
... |
Unused |
Input is expected to have two columns:
sample_type: A character vector denoting which samples are standards
with "standard". All other values will be considered unknowns.
index: A numeric column denoting the sample number. Index 1 will
correspond to the first item in standard_scale, 2 will be the second,
etc.
Same type as x, with a .conc column
abs <- expand.grid( sample_type = c("standard", "unknown"), index = 1:7 ) abs qp_add_std_conc(abs) # Can add custom scale - doesn't have to be 'in order' or unique: qp_add_std_conc(abs, c(1, 4, 2, 2, 3, 0.125, 7)) # Will warn - more values in `standard_scale` than standard indices # Will drop extra qp_add_std_conc(abs, 1:8) # Will error - fewer values in `standard_scale` than standard indices if (FALSE) { qp_add_std_conc(abs, 1:6) }abs <- expand.grid( sample_type = c("standard", "unknown"), index = 1:7 ) abs qp_add_std_conc(abs) # Can add custom scale - doesn't have to be 'in order' or unique: qp_add_std_conc(abs, c(1, 4, 2, 2, 3, 0.125, 7)) # Will warn - more values in `standard_scale` than standard indices # Will drop extra qp_add_std_conc(abs, 1:8) # Will error - fewer values in `standard_scale` than standard indices if (FALSE) { qp_add_std_conc(abs, 1:6) }
Calculate absorbance means with optional outlier removal
qp_calc_abs_mean(x, ignore_outliers = c("all", "standards", "samples", "none")) ## S3 method for class 'data.frame' qp_calc_abs_mean(x, ignore_outliers = c("all", "standards", "samples", "none")) ## S3 method for class 'list' qp_calc_abs_mean(x, ignore_outliers = c("all", "standards", "samples", "none"))qp_calc_abs_mean(x, ignore_outliers = c("all", "standards", "samples", "none")) ## S3 method for class 'data.frame' qp_calc_abs_mean(x, ignore_outliers = c("all", "standards", "samples", "none")) ## S3 method for class 'list' qp_calc_abs_mean(x, ignore_outliers = c("all", "standards", "samples", "none"))
x |
A |
ignore_outliers |
Which sample types should have outliers ignored from
their mean calculations? If |
Input data.frame must contain the following columns:
sample_type. Character. Must contain values either "standard" or
"unknown"
index. Numeric. Denotes sample number.
.abs. Numeric. Contains absorbance values.
If a boolean .is_outlier is supplied, that will be used instead.
The input tibble with an .is_outlier column and a .mean column
library(dplyr) abs <- expand.grid( sample_type = c("standard", "unknown"), index = 1:7, rep = 1:3 ) |> dplyr::arrange(sample_type, index, rep) abs$.abs <- abs(rnorm(nrow(abs), mean = abs$index)) # Selecting different subsets for outlier removal qp_calc_abs_mean(abs, "none") qp_calc_abs_mean(abs, "standards") qp_calc_abs_mean(abs, "samples") qp_calc_abs_mean(abs, "all") # If an `.is_outlier` column is provided, that will be used instead: abs$.is_outlier <- rep(c(TRUE, FALSE), length.out = nrow(abs)) qp_calc_abs_mean(abs)library(dplyr) abs <- expand.grid( sample_type = c("standard", "unknown"), index = 1:7, rep = 1:3 ) |> dplyr::arrange(sample_type, index, rep) abs$.abs <- abs(rnorm(nrow(abs), mean = abs$index)) # Selecting different subsets for outlier removal qp_calc_abs_mean(abs, "none") qp_calc_abs_mean(abs, "standards") qp_calc_abs_mean(abs, "samples") qp_calc_abs_mean(abs, "all") # If an `.is_outlier` column is provided, that will be used instead: abs$.is_outlier <- rep(c(TRUE, FALSE), length.out = nrow(abs)) qp_calc_abs_mean(abs)
Predict concentrations from standards fit
qp_calc_conc(x, ignore_outliers = TRUE, group_cols = c("sample_type", "index"))qp_calc_conc(x, ignore_outliers = TRUE, group_cols = c("sample_type", "index"))
x |
A list. See details. |
ignore_outliers |
Boolean. Should outliers be considered when calculating the mean? See details. |
group_cols |
Character vector. Columns to group by before taking the mean. |
The supplied list should contain two items - fit, generated by
qp_fit, and qp, a data.frame. qp should contain the following:
Columns used in fit. Usually, this is .log2_abs
Any columns in group_cols
If ignore_outliers = TRUE, .is_outlier will be used if supplied,
or created if not.
Returns a list with the input fit and data.frame, with additional
columns:
.pred: The predicted value from the provided model
.pred_conc: .pred, transformed by conc_transform
.pred_conc_mean: The mean of .pred_conc, sans samples where column
.is_outlier == TRUE
data <- system.file("extdata", "absorbances.txt", package = "qp") calculated <- qp(data, replicate_orientation = "h") # Making a minimal object: calculated$qp <- calculated$qp |> dplyr::select( .log2_abs, sample_type, index, .is_outlier ) calculated qp_calc_conc(calculated)data <- system.file("extdata", "absorbances.txt", package = "qp") calculated <- qp(data, replicate_orientation = "h") # Making a minimal object: calculated$qp <- calculated$qp |> dplyr::select( .log2_abs, sample_type, index, .is_outlier ) calculated qp_calc_conc(calculated)
Calculate dilutions from predicted concentrations
qp_dilute(x, ...) ## S3 method for class 'data.frame' qp_dilute( x, target_conc = NULL, target_vol = 15, remove_standards = FALSE, pipette_vol_compat = TRUE, ... ) ## S3 method for class 'list' qp_dilute( x, target_conc = NULL, target_vol = 15, remove_standards = FALSE, ... )qp_dilute(x, ...) ## S3 method for class 'data.frame' qp_dilute( x, target_conc = NULL, target_vol = 15, remove_standards = FALSE, pipette_vol_compat = TRUE, ... ) ## S3 method for class 'list' qp_dilute( x, target_conc = NULL, target_vol = 15, remove_standards = FALSE, ... )
x |
A |
... |
Unused |
target_conc |
Numeric vector. Target concentration in (mg/mL) protein. If length == 1, recycled. |
target_vol |
Target volume in uL. If length == 1, recycled. |
remove_standards |
Boolean. Should standards be removed from results? |
pipette_vol_compat |
Boolean. Shold returned numbers be rounded to the typically precision of a pipette? |
Same as input, with the volumes of lysate and volumes of diluent to add.
df <- data.frame(.pred_conc = 1) qp_dilute(df, target_conc = 0.5, target_vol = 30) # Many sample and target concentrations df2 <- data.frame(.pred_conc = 1:3) qp_dilute(df2, target_conc = c(0.1, 0.4, 0.8), target_vol = 30) # Takes a list, so long as it has a data.frame named qp as one of the items: ls <- list(qp = data.frame(.pred_conc = 3)) qp_dilute(ls, target_conc = 0.5, target_vol = 30)df <- data.frame(.pred_conc = 1) qp_dilute(df, target_conc = 0.5, target_vol = 30) # Many sample and target concentrations df2 <- data.frame(.pred_conc = 1:3) qp_dilute(df2, target_conc = c(0.1, 0.4, 0.8), target_vol = 30) # Takes a list, so long as it has a data.frame named qp as one of the items: ls <- list(qp = data.frame(.pred_conc = 3)) qp_dilute(ls, target_conc = 0.5, target_vol = 30)
Fit an lm using standards absorbances
qp_fit(x) ## S3 method for class 'data.frame' qp_fit(x) ## S3 method for class 'list' qp_fit(x)qp_fit(x) ## S3 method for class 'data.frame' qp_fit(x) ## S3 method for class 'list' qp_fit(x)
x |
A |
The supplied data.frame must have the following columns:
sample_type. Character. If not 'standard', assumed to be a sample
.is_outlier. Boolean. If TRUE, assumed to be outlier and removed from
fitting. If FALSE or NA, used for fitting. If unsupplied, will create one
with all values set to NA.
.conc. Numeric. Known concentration of standard.
.log2_abs. Numeric. The log2 of the absorbances
A list containing:
fit, an lm object fit with the formula .log2_conc ~ .log2_abs, fit
using non-outlier standards
qp, the input data
absorbances |> qp_add_std_conc() |> qp_fit()absorbances |> qp_add_std_conc() |> qp_fit()
Mark absorbance outliers
qp_mark_outliers(x, ignore_outliers = c("all", "standards", "samples", "none")) ## S3 method for class 'data.frame' qp_mark_outliers(x, ignore_outliers = c("all", "standards", "samples", "none")) ## S3 method for class 'list' qp_mark_outliers(x, ignore_outliers = c("all", "standards", "samples", "none"))qp_mark_outliers(x, ignore_outliers = c("all", "standards", "samples", "none")) ## S3 method for class 'data.frame' qp_mark_outliers(x, ignore_outliers = c("all", "standards", "samples", "none")) ## S3 method for class 'list' qp_mark_outliers(x, ignore_outliers = c("all", "standards", "samples", "none"))
x |
A |
ignore_outliers |
Which sample types should have outliers marked? |
Input data.frame must contain the following columns:
sample_type. Character. Must contain values either "standard" or
"unknown"
index. Numeric. Denotes sample number.
.abs. Numeric. Contains absorbance values.
The input tibble with an .is_outlier column
df <- data.frame( sample_type = rep(c("standard", "unknown"), each = 3), index = c(1, 1, 1, 2, 2, 2), .abs = c(1, 1, 1, 1, 1, 2) ) qp_mark_outliers(df, ignore_outliers = "all") qp_mark_outliers(df, ignore_outliers = "standards") qp_mark_outliers(df, ignore_outliers = "samples") qp_mark_outliers(df, ignore_outliers = "none")df <- data.frame( sample_type = rep(c("standard", "unknown"), each = 3), index = c(1, 1, 1, 2, 2, 2), .abs = c(1, 1, 1, 1, 1, 2) ) qp_mark_outliers(df, ignore_outliers = "all") qp_mark_outliers(df, ignore_outliers = "standards") qp_mark_outliers(df, ignore_outliers = "samples") qp_mark_outliers(df, ignore_outliers = "none")
It attempts to match the real life colors of a protein quantification
experiment, in combination with abs_to_col
qp_palqp_pal
An object of class character of length 100.
qp as they were on the plateView the absorbances of an analyzed qp as they were on the plate
qp_plot_plate(x, size = 15)qp_plot_plate(x, size = 15)
x |
A |
size |
The size of the points used to illustrate the wells. Passed to geom_point. |
a ggplot
qp_plot_plate(absorbances)qp_plot_plate(absorbances)
View an absorbance/concentration plot
qp_plot_standards(x)qp_plot_standards(x)
x |
The output of |
a ggplot
absorbances |> qp() |> qp_plot_standards()absorbances |> qp() |> qp_plot_standards()
Remove empty wells from data
qp_remove_empty(x) ## S3 method for class 'data.frame' qp_remove_empty(x) ## S3 method for class 'list' qp_remove_empty(x)qp_remove_empty(x) ## S3 method for class 'data.frame' qp_remove_empty(x) ## S3 method for class 'list' qp_remove_empty(x)
x |
A |
This function keeps any columns with positive .pred_conc or
sample_type == "standard"
Same as input
df <- expand.grid( .pred_conc = 0:1, sample_type = c("standard", "unknown") ) df qp_remove_empty(df)df <- expand.grid( .pred_conc = 0:1, sample_type = c("standard", "unknown") ) df qp_remove_empty(df)
Create a report for a protein quantificaiton experiment
qp_report(qp, output_file, other = list())qp_report(qp, output_file, other = list())
qp |
Likely the output from |
output_file |
Character. The path of the file to export, including
|
other |
Generally used for Shiny application. Assumes a named list of key-values that will be used to document report parameters. |
## Not run: absorbances |> qp() |> qp_dilute() |> qp_report( "~/my_report.html", other = list(key = "value") # Essentially metadata ) ## End(Not run)## Not run: absorbances |> qp() |> qp_dilute() |> qp_report( "~/my_report.html", other = list(key = "value") # Essentially metadata ) ## End(Not run)
Summarize output from qp pipeline
qp_summarize(x) ## S3 method for class 'data.frame' qp_summarize(x) ## S3 method for class 'list' qp_summarize(x)qp_summarize(x) ## S3 method for class 'data.frame' qp_summarize(x) ## S3 method for class 'list' qp_summarize(x)
x |
A |
A tibble with the sample name, sample_type, and the mean of its
predicted concentration (.pred_conc_mean)
Read in and wrangle protein quantification data
qp_tidy(x, ...) ## S3 method for class 'character' qp_tidy(x, ...) ## S3 method for class 'spectramax' qp_tidy( x, replicate_orientation = c("h", "v"), n_standards = 7, n_replicates = 3, wavelength = 562, ... ) ## S3 method for class 'gp' qp_tidy(x, ...) ## Default S3 method: qp_tidy(x, ...)qp_tidy(x, ...) ## S3 method for class 'character' qp_tidy(x, ...) ## S3 method for class 'spectramax' qp_tidy( x, replicate_orientation = c("h", "v"), n_standards = 7, n_replicates = 3, wavelength = 562, ... ) ## S3 method for class 'gp' qp_tidy(x, ...) ## Default S3 method: qp_tidy(x, ...)
x |
A |
... |
Arguments passed to relevant methods. |
replicate_orientation |
Character. Specified the direction the
replicates lie, not the direction the samples flow (which will be
perpendicular to |
n_standards |
Numeric. The number of different concentrations of standards. Does not include replicates. |
n_replicates |
Numeric. The number of replicates per sample. |
wavelength |
Numeric. For SPECTRAmax files and objects, the wavelength measured. Otherwise, ignored. |
qp assumes that if you read in data not in a spectramax file or
object, you probably have a custom workflow in mind - therefore, tidying
will be minimal and mostly focused on checking for validity.
a data.frame
data <- system.file("extdata", "absorbances.txt", package = "qp") qp_tidy(data)data <- system.file("extdata", "absorbances.txt", package = "qp") qp_tidy(data)