| prep.norm {mdatools} | R Documentation |
Normalization
Description
Normalizes signals (rows of data matrix).
Usage
prep.norm(data, type = "area", col.ind = NULL, ref.spectrum = NULL)
Arguments
data |
a matrix with data values |
type |
type of normalization |
col.ind |
indices of columns (can be either integer or logical valuws) for normalization to internal standard peak. |
ref.spectrum |
reference spectrum for PQN normalization, if not provided a mean spectrum for data is used |
Details
The "area", "length", "sum" types do preprocessing to unit area (sum of
absolute values), length or sum of all values in every row of data matrix. Type "snv"
does the Standard Normal Variate normalization, similar to prep.snv. Type
"is" does the normalization to internal standard peak, whose position is defined by
parameter 'col.ind'. If the position is a single value, the rows are normalized to the height
of this peak. If 'col.ind' points on several adjucent vales, the rows are normalized to the area
under the peak - sum of the intensities.
The "pqn" is Probabilistic Quotient Normalization as described in [1]. In this case you also
need to provide a reference spectrum (e.g. mean or median of spectra for some reference samples). If
reference spectrum is not provided it will be computed as mean of the spectra to be
preprocessed (parameter data).
Value
data matrix with normalized values
References
1. F. Dieterle, A. Ross, H. Senn. Probabilistic Quotient Normalization as Robust Method to Account for Dilution of Complex Biological Mixtures. Application in 1 H NMR Metabonomics. Anal. Chem. 2006, 78, 4281–4290.