read_nmrpipe: Read NMRPipe spectrum
In Smith-Group/fitnmr: FitNMR multidimensional spectrum analysis

read_nmrpipe

R Documentation

Read NMRPipe spectrum

Description

This function reads 1D-4D spectra stored in the NMRPipe format.

Usage

read_nmrpipe(inFormat, dim_order = NULL, complex_data = FALSE)

Arguments

`inFormat`	character with file name or format for multiple files
`dim_order`	integer vector used to reorder dimensions or character specifying
`complex_data`	logical value indicating whether complex data should be read

Details

For three and four dimensional datasets, the spectral data is often spread across multiple files. To read those, inFormat should be a sprintf-style string that describes how the files are named. For instance, if the files are named 001.ft3, 002.ft3, etc., then inFormat should be "%03i.ft3". If there is no zero-padding, as in this case, 0 should be omitted from the format. If there are fewer digits, then the first 3 should be changed accordingly.

The default 2D NMRPipe scripts only have a single transpose ("TP") command, leaving the the indirect dimension as the first dimension in the resulting array. The 2D plotting functions in fitnmr usually plot this first dimension along the x-axis, which will make for generally non-standard contour plots. Furthermore, when peak fitting is employed, this will also be the first dimension. To fix this, you can change the spectral order with the dim_order parameter. The order of the dimensions should be specified in the same way that would be done for aperm. Alternatively, you can specify a character argument to have fitnmr attempt to automatically detect and correct the array order. The only currently supported type is "hx", which will put the dimension with the greatest observe frequency first.

This function is partly based on the pipe2rnmr function from rNMR.

Value

a named list with four elements:

int: multidimensional array with spectrum intensities (ppm values are given in the dimnames)
ppm: list of numeric vectors giving the ppm values associated with each int array dimension
fheader: matrix with dimension-specific header information
header: numeric vector with the complete header contents

The fheader row definitions are as follows (taken from NMRPipe fdatp.h):

Row Name	Description
SIZE	Number of points in dimension
APOD	Current valid time-domain size
SW	Sweep Width, Hz
ORIG	Axis Origin (Last Point), Hz
OBS	Obs Freq, MHz
FTFLAG	1=Freq Domain 0=Time Domain
QUADFLAG	Data Type Code (See Below)
UNITS	Axis Units Code (See Below)
P0	Zero Order Phase, Degrees
P1	First Order Phase, Degrees
CAR	Carrier Position, PPM
CENTER	Point Location of Zero Freq
AQSIGN	Sign adjustment needed for FT
APODCODE	Window function used
APODQ1	Window parameter 1
APODQ2	Window parameter 2
APODQ3	Window parameter 3
C1	Add 1.0 to get First Point Scale
ZF	Negative of Zero Fill Size
X1	Extract region origin, if any, pts
XN	Extract region endpoint, if any, pts
OFFPPM	Additional PPM offset (for alignment)
FTSIZE	Size of data when FT performed
TDSIZE	Original valid time-domain size
LB	Extra Exponential Broadening, Hz
GB	Extra Gaussian Broadening, Hz
GOFF	Offset for Gaussian Broadening, 0 to 1
OBSMID	Original Obs Freq before 0.0ppm adjust

In addition several rows contain information inferred from the header data:

Row Name	Description
aq_s	Acquisition time, seconds
sw_ppm	Original Sweep Width, PPM
direct	Direct (1) or indirect (0) dimension
alias	Aliasing (0/1) with inversion (-1)