Calculate features based on the cell-based or full scan registration procedures

The feature_registration_all function calculates the seven registration-based features in one call. The feature_registration_summary function is an exported helper.

Usage

feature_registration_all(comparisonData, id_cols = NULL)

feature_registration_summary(
  cellIndex,
  direction,
  fft_ccf,
  summaryVar,
  summaryFun = mean,
  imputeVal = NA
)

Arguments

comparisonData

tibble such as one returned by the comparison_cellBased() or comparison_fullScan() functions that contains results from the cell-based or full scan comparison procedure

id_cols

variable(s) to group by prior to calculating the summary statistics

cellIndex

tibble column containing cell indices

direction

tibble column indicating whether the associated row came from the "reference_vs_target" or "target_vs_reference" comparison

fft_ccf

tibble column containing cross-correlation function values

summaryVar

tibble column that is to be summarized

summaryFun

function that will be used to summarize the values in the summaryVar column

imputeVal

value to return if the feature calculation results in a non-numeric (i.e., NA, NULL) value

Note

The feature_registration_all function can be used on comparison data from a full-scan or cell-based comparison. For a full-scan comparison, we recommend using only the average CCF and average pairwise-complete correlation.

To calculate the visual diagnostic features, you must set the returnX3Ps argument in the comparison_cellBased() or comparison_fullScan() functions to TRUE.

Examples

data("K013sA1","K013sA2")

compData_cellBased <- comparison_cellBased(reference = K013sA1,
                                           target = K013sA2,
                                           thetas = c(-3,0,3))
#> Error in mutate(., direction = "reference_vs_target"): could not find function "mutate"

feature_registration_all(compData_cellBased)
#> Error in dplyr::summarise(., ccfMean = feature_registration_summary(cellIndex = cellIndex,     direction = direction, fft_ccf = fft_ccf, summaryVar = fft_ccf,     summaryFun = mean, imputeVal = NA), ccfSD = feature_registration_summary(cellIndex = cellIndex,     direction = direction, fft_ccf = fft_ccf, summaryVar = fft_ccf,     summaryFun = sd, imputeVal = NA), pairwiseCompCorAve = feature_registration_summary(cellIndex = cellIndex,     direction = direction, fft_ccf = fft_ccf, summaryVar = pairwiseCompCor,     summaryFun = mean, imputeVal = NA), pairwiseCompCorSD = feature_registration_summary(cellIndex = cellIndex,     direction = direction, fft_ccf = fft_ccf, summaryVar = pairwiseCompCor,     summaryFun = sd, imputeVal = NA), xTransSD = feature_registration_summary(cellIndex = cellIndex,     direction = direction, fft_ccf = fft_ccf, summaryVar = x,     summaryFun = sd, imputeVal = NA), yTransSD = feature_registration_summary(cellIndex = cellIndex,     direction = direction, fft_ccf = fft_ccf, summaryVar = y,     summaryFun = sd, imputeVal = NA), thetaRotSD = feature_registration_summary(cellIndex = cellIndex,     direction = direction, fft_ccf = fft_ccf, summaryVar = theta,     summaryFun = sd, imputeVal = NA)): object 'compData_cellBased' not found

compData_fullScan <- comparison_fullScan(reference = K013sA1,
                                         target = K013sA2,
                                         thetas = c(-3,0,3))

feature_registration_all(compData_fullScan) %>%
  dplyr::select(ccfMean,pairwiseCompCorMean)
#> Error in dplyr::select(., ccfMean, pairwiseCompCorMean): Can't subset columns that don't exist.
#> ✖ Column `pairwiseCompCorMean` doesn't exist.