Doc: Data Cleaning

Missing Value Imputation

fcdc_impute(fc, start_col=fccu_first_feat_col(fc), type="det", method=median)

Impute missing values in an Fc Array data frame on a per-group basis. Thus requires fc$group exists. This function can handle both single time point and longitudinal Fc Array data frames.
@param fc The Fc Array data frame. Can be single time point or longitudinal.
@param start_col Where you want the imputing to start iterating along columns from, presumably fccu_first_feat_col(fc).
@param type String denoting the type of imputation. Currently, only "det" for deterministic is suppoted.
@param method For calculating imputed values. Default is median.
@return The Fc Array data frame, with missing values replaced based on method. By default, this will be the median of a group's values for that feature.

fcdc_normalize(fc, center=TRUE, scale=TRUE)

Perform normalization on an Fc Array data frame. Just a glorified wrapper for scale() that calls fcdc_impute if necessary.
@param fc The Fc Array data frame.
@param center A boolean or vector. See ?scale
@param scale A boolean or vector. See ?scale
@return The Fc Array data frame, with feature vectors scaled/centered as specified.

fcdc_sub_back(fc, base_ids)

Perform background signal subtraction from an Fc Array data frame.
@param fc The Fc Array data frame.
@param base_ids The row indices in fc that correspond to the subjects who represent baseline (say, those in the PLACEBO group).
@return The Fc Array data frame, with fc[base_ids,] having all feature measurements set to 0, and with non-baseline subjects having had their MFIs subtracted by the median of the baseline group's MFI values for each feature, and then scaled by the sd of those baseline MFI values.