data_selection: Extracts samples information from a lum_export object or...
In drLumi: Multiplex Immunoassays Data Analysis

Description Usage Arguments Details Value Examples

Extracts from a lum_export or data.frame object controls, dilutions points, background and samples to be calibrated. These files can be merged to an expected concentration and flag dataset.

data_selection(x, ecfile = NULL, flagsfile = NULL, backname = "Back",
    stanname = "Stan", posname = "Con", unkname = NULL,
    byvar.ecfile = c("analyte", "sample"), byvar.flagsfile = c("well",
    "analyte"), fsample = "sample", fanalyte = "analyte", fbatch = "plate",
    ...)

`x`	a `lum_export` object or `data.frame` with samples information.
`ecfile`	a `data.frame` or CSV path file with the expected concentration in order to merge to `x`.
`flagsfile`	a `data.frame` or CSV path with the flags information in order to merge to `x`.
`backname`	character vector or list of the background samples identification of `x`.
`stanname`	character vector or list of the standard points identification of `x`.
`posname`	character vector or list of the positive controls identification of `x`.
`unkname`	character vector or list of the samples identification of `x`. Default `NULL`.
`byvar.ecfile`	character vector of the merging variable(s) of `x` and `ecfile`.
`byvar.flagsfile`	character vector of the merging variable(s) of `x` and `flagsfile`.
`fsample`	character vector of the name of the sample variable.
`fanalyte`	character vector of the name of the analyte variable.
`fbatch`	character vector of the name of the Batch variable.
`...`	other options. Ignored.

Default method for identifying background, standard and positives samples is to define a character vector of length one and apply agrep functions in order to extract databases. Samples to be calibrated (unknowns) are identified as the remaining ones (default NULL).

If the arguments are defined as a list the function will subset exactly that information from x object.

The expected concentration file is merged based on byvar.ecfile. Only applies when ecfile is not NULL. Same applies for flags file. A variable named 'flag' must be in flags data in order to perform the merge.

The name of the batch in a list format with the following components: background, standard, positive and unknowns datasets.

# Load data
data(ecdata)
data(mfidata)

dat <- subset(mfidata,plate=="plate_1" & analyte=="FGF")

# Example 1
sdf <- data_selection(dat)

lapply(sdf$plate_1, function(x) head(x))

# Example 2 (merge ecdata)
sdf <- data_selection(dat, ecfile = ecdata,
             byvar.ecfile=c("analyte","sample"))

lapply(sdf$plate_1, function(x) head(x))

# Example 3 (extract specific samples names with list)
sdf <- data_selection(dat,
             stanname=list("Standard10"),
             backname = list("Background0"),
             posname = list("Control1","Control2"),
             unkname = list("B_sid_13_CSP"))

lapply(sdf$plate_1, function(x) head(x))

# Example 4 (extract aproximate names samples)
sdf <- data_selection(dat,
             stanname="Standard1",
             backname = "Background0",
             posname = "Control1",
             unkname = "B_sid_13_CSP")

lapply(sdf$plate_1, function(x) (x))