Hi, For such questions, it may be better to ask on the relax-users mailing list as the answer might be useful for other users. You can see what is happening with Sparky peak list files in the relax module lib.software.sparky. If you specifically look at the read_list_intensity() function, you will see the lines: # Skip non-assigned peaks. if line[0] == '?-?': continue This needs to be done in relax as it is not possible to hold data from such peaks in the relax data store. The way I handle such a situation in Sparky is to give all such peaks a temporary or pseudo-assignment, for example SS1000, SS1001, etc. Essentially having a special residue name and numbers starting much higher than the end of your system's sequence numbering. These can then be given 'N' and 'HN' labels. You would then need to create these spin system within relax prior to reading the peak lists, and the data will then be handled as normal. This could either be by creating and then reading in a text file containing all of these spins with residue numbers and names in different columns, using individual calls to the spin.create user function, or within the spin editor window of the GUI (which just provides access to the graphical version of the spin.create user function). For the question about multiple reference spectra, have a look at the specific_analyses/relax_disp/__init__.py file (in the relax_disp branch). Specifically within the calculate() method used for the fixed time period experiments, you will see the lines: # Average the reference intensity data and errors. ref_intensity = average_intensity(spin=spin, frq=frq, point=None, time=time) ref_intensity_err = average_intensity(spin=spin, frq=frq, point=None, time=time, error=True) # Average the intensity data and errors. intensity = average_intensity(spin=spin, frq=frq, point=point, time=time) intensity_err = average_intensity(spin=spin, frq=frq, point=point, time=time, error=True) # Calculate the R2eff value. spin.r2eff[param_key] = calc_two_point_r2eff(relax_time=time, I_ref=ref_intensity, I=intensity) # Calculate the R2eff error. spin.r2eff_err[param_key] = calc_two_point_r2eff_err(relax_time=time, I_ref=ref_intensity, I=intensity, I_ref_err=ref_intensity_err, I_err=intensity_err) So you can see that the data from multiple spectra are averaged. This relies on the my assumption that this is the logical thing to do. For the peak intensity error analysis prior to this, how multiple spectra are handled depends on what you have measured and whether you have measured all spectra replicated or only a subset. This is described in full detail within the documentation for the current spectrum.error_analysis user function (see either the help system or the user manual at http://www.nmr-relax.com/manual/spectrum_error_analysis.html). Regards, Edward On 3 June 2013 18:55, Troels Emtekær Linnet <tlinnet@xxxxxxxxx> wrote:
Hi ! I have been playing around in relax_disp, and wonder. Can relax handle un-assigned peaks? Assignment w1 w2 Height Volume A101N-A101HN 124.571 8.003 0.00E+00 0.00E+00 -- ?-? 124.316 7.962 0.00E+00 0.00E+00 -- We often have 2-3 un-assigned peaks, which is interesting to carry on in our analysis. Can you have several reference spectra? We have some datasets with 2-3 replicates for the reference spectrum. Best Troels Emtekær Linnet _______________________________________________ relax (http://www.nmr-relax.com) This is the relax-devel mailing list relax-devel@xxxxxxx To unsubscribe from this list, get a password reminder, or change your subscription options, visit the list information page at https://mail.gna.org/listinfo/relax-devel