Difference between revisions of "Tutorial for R1/R2 Relaxation curve-fitting analysis on varian recorded as fid interleaved"

From relax wiki
Jump to navigation Jump to search
(Forced creation of a TOC - this will improve the formatting on the main page 'Did you know...' section.)
(→‎Export peak list from analysis: Use of button and key press templates and the code tag for better formatting.)
Line 138: Line 138:
 
# Peaks export menu
 
# Peaks export menu
 
# Select spectrum which contain your peaks
 
# Select spectrum which contain your peaks
# Select export file. Save to somewhere. For example: '''test.sparky'''.
+
# Select export file. Save to somewhere. For example: <code>test.sparky</code>.
 
# Export
 
# Export
# In the last menu which pop-up, slect OK for moelcular system
+
# In the last menu which pop-up, select {{button|OK}} for molecular system
# Then be SURE to have '''Proton first in Dim 0'''.
+
# Then be SURE to have <code>Proton first in Dim 0</code>.
  
 
Remove potential un-assigned peaks '?'.
 
Remove potential un-assigned peaks '?'.
Line 183: Line 183:
 
</source>
 
</source>
  
In NMRDraw, '''Shift + k'''. Push '''read'''. Now zoom into a peak. Be "sure", that you have not moved the spectrum in analysis or so.
+
In NMRDraw, {{key press|shift|k}}. Push {{button|read}}. Now zoom into a peak. Be "sure", that you have not moved the spectrum in analysis or so.
  
 
== Do intensity measurements for all spectra ==
 
== Do intensity measurements for all spectra ==

Revision as of 11:47, 18 November 2015

Method 1

Intro

Made from Tutorial for Relaxation dispersion analysis cpmg fixed time recorded on varian as fid interleaved

Get the process helper scripts

  1. stPeakList.pl

Check the peak list matches

Check the peak list matches

sparky 0/test.ucsf

Prepare files

Make standard peak list

stPeakList.pl 0/test.ft2 peak_list_SPARKY.list > peak_list_SPARKY.tab
cat peak_list_SPARKY.tab

Find *.ft2 files

ls -v -d -1 */*.ft2 > ft2_files.ls
cat ft2_files.ls

Measure heights

seriesTab -in peak_list_SPARKY.tab -out peaks_list_max_standard.ser -list ft2_files.ls -max
cat peaks_list_max_standard.ser

Extract the spectra settings from Varian procpar file

set RELAXT=`awk '/^relaxT /{f=1;next}f{print $0;exit}' procpar`; echo $NCYCLIST
set SFRQ=`awk '/^sfrq /{f=1;next}f{print $2;exit}' procpar`; echo $SFRQ

foreach I (`seq 2 ${#RELAXT}`)
  set T=${RELAXT[$I]}; echo $T $SFRQ >> relaxt.txt
end

cat relaxt.txt

Measure RMSD in spectra

Measure the backgorund noise RMSD in each of the .ft2 files

Very fast way - RMSD via nmrpipe showApod

We can also use the showApod rmsd.

set FIDS=`cat ft2_files.ls`
set OUT=${PWD}/apod_rmsd.txt
set CWD=$PWD
rm $OUT

foreach I (`seq 1 ${#FIDS}`)
set FID=${FIDS[$I]}; set DIRN=`dirname $FID`
cd $DIRN
set apodrmsd=`showApod *.ft2 | grep "REMARK Automated Noise Std Dev in Processed Data:" | awk '{print $9}'`
echo $apodrmsd $DIRN >> $OUT
cd $CWD
end
cat $OUT

paste relaxt.txt $OUT > settings.txt

Relax script

File: relax_1_ini.py

import os
from auto_analyses.relax_fit import Relax_fit

# Taken from the relax disp manual, section 10.6.1 Dispersion script mode - the sample script
# Create the data pipe.
pipe_name = 'base pipe'
pipe_bundle = 'relax_fit'
pipe.create(pipe_name=pipe_name, bundle=pipe_bundle, pipe_type='relax_fit')
 
# Create the spins
spectrum.read_spins(file='peaks_list_max_standard.ser', dir=None)
 
# Read the spectrum from NMRSeriesTab file. The "auto" will generate spectrum name of form: Z_A{i}
spectrum.read_intensities(file="peaks_list_max_standard.ser", dir=None, spectrum_id='auto', int_method='height')
 
# Loop over the spectra settings.
timesfile=open('settings.txt','r')
 
i = 0
for line in timesfile:
    timeT = line.split()[0]
    set_sfrq = float(line.split()[1])
    rmsd_err = float(line.split()[2])
 
    print timeT, set_sfrq, rmsd_err
 
    # Set the current spectrum id
    current_id = "Z_A%s"%(i)

    # Set the relax time
    relax_fit.relax_time(time=float(timeT), spectrum_id=current_id)

    # Set the peak intensity errors, as defined as the baseplane RMSD.
    spectrum.baseplane_rmsd(error=rmsd_err, spectrum_id=current_id)
 
    i += 1
 
# Save the program state before run.
# This state file will also be used for loading, before a later cluster/global fit analysis.
state.save('ini_setup', force=True)

# Set settings for run.
results_directory = os.path.join(os.getcwd(),"model_sel_analyt")
pipe_name = 'base pipe'; pipe_bundle = 'relax_fit'
#GRID_INC = 11; MC_NUM = 500
GRID_INC = 21; MC_NUM = 100

Relax_fit(pipe_name=pipe_name, pipe_bundle=pipe_bundle, file_root='rx', results_dir=results_directory, grid_inc=GRID_INC, mc_sim_num=MC_NUM, view_plots=True)

Analyse

relax_trunk relax_1_ini.py -t log_relax_1_ini.log

Method 2

Intro

You have measured a range of HSQC with different settings of T1 and T2.

You have imported one spectrum in analysis, made peak assignment, and now wan't to analyse for all the spectrum.

Export peak list from analysis

We want to use the NMRPipe program, seriesTab.
First export to Sparky format.

Go to

  1. Other
  2. Format converter
  3. Export
  4. Sparky
  5. Peaks export menu
  6. Select spectrum which contain your peaks
  7. Select export file. Save to somewhere. For example: test.sparky.
  8. Export
  9. In the last menu which pop-up, select OK for molecular system
  10. Then be SURE to have Proton first in Dim 0.

Remove potential un-assigned peaks '?'.

grep -v "?" test.sparky > test2.sparky

Then sort the file, according to residue number. Be aware of small tricks.

# Extraxt assignments, removing first letter
awk '{print $1}' test2.sparky | cut -d'-' -f1 | cut -c2- > sort.txt

# Merge columns to prepare for sort
paste sort.txt test2.sparky > sort2.tx

# Sort after natural sort of (version) numbers within text
sort -V sort2.tx | grep -v "Assignment" > sort3.txt

# Now extract columns
awk '{print $2, $3, $4, $5, $6, $7}' sort3.txt > test3.sparky
cat test3.sparky

You should now have a sorted peak list. Congratulations.

Now we convert the peak list from Sparky to NMRPipe format. First find directory with files.

 
# Go into directory, and copy the peak file
cd "/sbinlab2/rma/Desktop/Documents/.../XYZ.fid"
cp /sbinlab2/rma/Desktop/test/test3.sparky .

set SPEAKS=test3.sparky
set FTFILE=0/test.ft2

stPeakList.pl $FTFILE $SPEAKS > test.tab
cat test.tab

# Now check file
nmrDraw $FTFILE

In NMRDraw, Shift+k. Push read. Now zoom into a peak. Be "sure", that you have not moved the spectrum in analysis or so.

Do intensity measurements for all spectra

# '-1': single column, 'v'  natural sort of (version) numbers within text, 
ls -v -d -1 */*.ft2 > ft2_files.ls
cat ft2_files.ls

# So. -in 'test.tab' contains information about x,y positions in frequency units.
# -out tells where to store file.
# -list tells which ft2 files to compare.
# -max takes the height.
# -max -dx 1 -dy 1, takes the height within a freqency box of +/- 1 freqeuncy unit.
seriesTab -in test.tab -out peaks_list_max_standard.ser -list ft2_files.ls -max
seriesTab -in test.tab -out peaks_list_max_dx1_dy1.ser -list ft2_files.ls -max -dx 1 -dy 1

Extract time delays from procpar

# '^' look for line starting with. 
# Then take next line.
# print from column 2 to end of line.
set RELAXT=`awk '/^relaxT /{f=1;next}f{print $0;exit}' procpar | cut -d" " -f 2-`; echo $RELAXT

set SFRQ=`awk '/^sfrq /{f=1;next}f{print $2;exit}' procpar`; echo $SFRQ

Now extract noise from spectrums

set FIDS=`cat ft2_files.ls`
set OUT=../apod_rmsd.txt
 
foreach I (`seq 1 ${#FIDS}`)
  set FID=${FIDS[$I]}; set DIRN=`dirname $FID`
  cd $DIRN
  set apodrmsd=`showApod *.ft2 | grep "REMARK Automated Noise Std Dev in Processed Data:" | awk '{print $9}'`
  set RELAXTI=${RELAXT[$I]}
  echo $DIRN $RELAXTI $apodrmsd $SFRQ >> $OUT
  cd ..
end

cat apod_rmsd.txt

You should now a file with settings, which tell the delay time and spectrum noise for the measurements.
You also have a file, which have measured intensities for spins at each spectrum.

Analyse in relax

In GUI

gedit analyse_relax_gui.py
# Create the spins
spectrum.read_spins(file="peaks_list_max_standard.ser", dir=None)

# Read the spectrum from NMRSeriesTab file. The "auto" will generate spectrum name of form: Z_A{i}
spectrum.read_intensities(file="peaks_list_max_standard.ser", dir=None, spectrum_id='auto', int_method='height')

# Open the settings file.
set_file = open('apod_rmsd.txt', 'r')

for i, line in enumerate(set_file):
    # Set the current spectrum id
    current_id = "Z_A%s"%(i)

    # Get the relax time.
    relaxT_i = float(line.split()[1])
    rmsd_i = float(line.split()[2])
    sfrq_i = float(line.split()[3])

    print relaxT_i, rmsd_i, sfrq_i

    # Set time
    relax_fit.relax_time(time=relaxT_i, spectrum_id=current_id)

    # Set noise
    spectrum.baseplane_rmsd(error=rmsd_i, spectrum_id=current_id)

    # Set sfrq
    spectrometer.frequency(id=current_id, frq=sfrq_i, units='MHz')

Then do

  1. relax -g
  2. New analysis
  3. R2 relaxation
  4. Create pipe
  5. User functions (n-z) -> script -> analyse_relax_gui.py
  6. Set in top NMR freqeuncy label to 750.
  7. Go

In GUI

gedit analyse_relax.py
from auto_analyses.relax_fit import Relax_fit

# Taken from the relax disp manual, section 10.6.1 Dispersion script mode - the sample script
# Create the data pipe.
pipe_name = 'base pipe'
pipe_bundle = 'RMA'
pipe.create(pipe_name=pipe_name, bundle=pipe_bundle, pipe_type='relax_fit')
 
# Create the spins
spectrum.read_spins(file="peaks_list_max_standard.ser", dir=None)

# Read the spectrum from NMRSeriesTab file. The "auto" will generate spectrum name of form: Z_A{i}
spectrum.read_intensities(file="peaks_list_max_standard.ser", dir=None, spectrum_id='auto', int_method='height')

# Open the settings file.
set_file = open('apod_rmsd.txt', 'r')

for i, line in enumerate(set_file):
    # Set the current spectrum id
    current_id = "Z_A%s"%(i)

    # Get the relax time.
    relaxT_i = float(line.split()[1])
    rmsd_i = float(line.split()[2])
    sfrq_i = float(line.split()[3])

    print relaxT_i, rmsd_i, sfrq_i

    # Set time
    relax_fit.relax_time(time=relaxT_i, spectrum_id=current_id)

    # Set noise
    spectrum.baseplane_rmsd(error=rmsd_i, spectrum_id=current_id)

    # Set sfrq
    spectrometer.frequency(id=current_id, frq=sfrq_i, units='MHz')

Relax_fit(pipe_name=pipe_name, pipe_bundle=pipe_bundle, file_root='R2', results_dir=None, grid_inc=11, mc_sim_num=500, view_plots=False)

Run relax

relax analyse_relax.py

See also