process example: process_pipe_3d

This example shows how nmrglue can be used to process NMR data. In this script a 3D time domain NMRPipe file is processing into a 3D NMRPipe frequency domain file. For 3D processing the XXX iter3D object is used to loop over XY and ZX planes.After processing the resulting file is compared to one processed with NMRPipe.

[source code]

#! /usr/bin/env python

import nmrglue as ng

# create the iterator (xiter) which will loop over YX planes
xiter = ng.pipe.iter3D("../common_data/3d_pipe/data/test%03d.fid",'x','x')

print "Processing XY planes..."
for i,(dic,plane) in enumerate(xiter):
    
    # process the direct dimention (x)
    dic,plane = ng.pipe_proc.zf(dic,plane,auto=True)
    dic,plane = ng.pipe_proc.ft(dic,plane,auto=True)
    dic,plane = ng.pipe_proc.ps(dic,plane,p0=0.0,p1=0.0)
    dic,plane = ng.pipe_proc.di(dic,plane)

    # process the first indirect (y) dimention
    dic,plane = ng.pipe_proc.tp(dic,plane)
    dic,plane = ng.pipe_proc.zf(dic,plane,auto=True)
    dic,plane = ng.pipe_proc.ft(dic,plane,auto=True)
    dic,plane = ng.pipe_proc.ps(dic,plane,p0=0.0,p1=0.0)
    dic,plane = ng.pipe_proc.di(dic,plane)
    dic,plane = ng.pipe_proc.tp(dic,plane)
   
    # write out the plane
    #print "Writing out Z slice:",i,"/",xiter.i_max
    xiter.write("./ft/test%03d.ft2",plane,dic)


print "Processing ZX planes..."

# create the iterator (ziter) which will loop over XZ planes
ziter = ng.pipe.iter3D("./ft/test%03d.ft2",'z','z')

for i,(dic,plane) in enumerate(ziter):

    # process the z-dim
    dic,plane = ng.pipe_proc.zf(dic,plane,auto=True)
    dic,plane = ng.pipe_proc.ft(dic,plane,auto=True)
    dic,plane = ng.pipe_proc.ps(dic,plane,p0=-92.0,p1=65.0)
    dic,plane = ng.pipe_proc.di(dic,plane)

    # write out the plane
    #print "Writing out Y slice:",i,"/",ziter.i_max
    ziter.write("./ft/test%03d.ft3",plane,dic)

print "Done"
# check against a file processed with NMRPipe
dic1,data1 = ng.pipe.read_lowmem("../common_data/3d_pipe/ft/test%03d.ft3")
dic2,data2 = ng.pipe.read_lowmem("./ft/test%03d.ft3")
# we'll check two 2D slices as the whole 3D takes a long time
print ng.misc.pair_similar(dic1,data1[5],dic2,data2[5],verb=True)
print ng.misc.pair_similar(dic1,data1[52],dic2,data2[52],verb=True)