-
Notifications
You must be signed in to change notification settings - Fork 5
WET List
Updated: December 2018
Farseer-NMR version: +1.3.0
Before or during the calculation run, Farseer-NMR tries to identify possible mistakes introduced by the user that could give rise to wrong analysis results or program crashes. Mistakes can be of several nature, for example, wrong input data format or wrong feature selection.
When an error is spot, a short message rises explaining the issue identified, the run is consequently paused or aborted. Each issue is identified by a WET number to each the user should refer to (in the list bellow) for detailed explanation on how to solve it.
The WET list does not follow an order or any kind.
The PRE Analysis flag is set to True but other flags upon which it fully or partially depends are set to False.
- PRE Analysis is only performed over the analysis of condition 3 (z axis), if this flag (do_along_z) is set to False, PRE Analysis won't be performed.
- PRE Analysis is performed on Height and/or Volume ratio calculations, at least one of those flags should be set True.
- DELTA_PRE Oscilation maps are only performed on Comparisons, Perform Comparisons should True for this plot to be printed.
- Z axis data points folder names must be
diaandparaor00_diaand01_para. Other names are not allowed.
No analysis is activated. Farseer performs analysis on three different dimensions (conditions) of the input dataset, nevertheless, those analysis routines need to be activated in the farseer_user_variables file using variables do_cond1, do_cond2 and do_cond3. If all those variables are set to False, parsed peaklists with lost and unassigned residues identified will be exported but no calculation or plot representation will be performed
No plots activated. Farseer allows the user to configure which parameters are calculated and from those, which kind of plots are drawn. When no plots are turned ON Farseer will export only the calculated parameters.
Input x values for restraints fitting do not match data points for condition 1. Farseer allows fitting of the calculated restraints to specific equations, nevertheless, fitting is only allowed along the condition 1 data points. For fitting to be performed Farseer requires the user to input a list of values that correspond to the experimental values to be used in as x values in the fitting equation. For example, fitting a CSPs evolution to an increasing ligand concentration requires fitting_x_values variable to be a list of the experimentally investigated ligand concentrations. The number of input values has to match the number of peaklist (.csv files) along condition 1 and if condition 1 .csv file names can be converted to number values, those have to match the fitting_x_values variable.
Number of input x values (coordinate axis) do not match the number of input peaklists. Farseer allows fitting of the calculated restraints to specific equations, nevertheless, fitting is only allowed along the condition 1 data points. For fitting to be performed Farseer requires the user to input a list of values that correspond to the experimental values to be used in as x values in the fitting equation. For example, fitting a CSPs evolution to an increasing ligand concentration requires fitting_x_values variable to be a list of the experimentally investigated ligand concentrations. The number of input values has to match the number of peaklist (.csv files) along condition 1, where 0 relates to the reference experiment (peaklist). Even if fitting is disabled, fitting_x_values can be used for Residue Evolution Plot if res_evo_set_x_values flag, fitting_x_values has to match number of peaklists input.
There are negative values in titration_x_values variable. Farseer has implemented fitting only to the Hill Equation (http://www.physiologyweb.com/calculators/hill_equation_interactive_graph.html) which does not consider the use of negative values in the coordinate axis. Please correct your input data in titration_x_values variable.
x values given by the user for fitting look good. This is not an error nor a warning, it's just an alert to the user about the values used in as x coordinates to perform the fitting.
Files are missing in the input folder tree. Farseer assumes that when analysing different titrations together, those are related. Therefore the same number of files (data points) should be given for every condition. For example, there should be the same number of peaklists (.csv files) in each subfolder, as well as the same subfolder tree should be maintained under spectra/.
There are no files of the type file type. You have asked Farseer-NMR to read a certain file type that does not exist in the spectra/ folder.
File names do not match. File names corresponding to the X data points, for example, peaklists .csv files, MUST have the same name in all the y data point folders. Even if files reffer to different concents, names should be normallized to be the same. For example:
WRONG!
298/
M1/
M1_L1_1.csv
M1_L1_2.csv
298/
M2/
M2_L1_1.csv
M2_L1_2.csv
RIGHT!
298/
M1/
L1_1.csv
L1_2.csv
298/
M2/
L1_1.csv
L1_2.csv
Folder names do not match. Folder names corresponding to the Y data points (for example peaklists .csv files) MUST have the same name in all the Z data point folders. For example:
WRONG!
298/
298_M1/
L1_1.csv
L1_2.csv
278/
278_M1/
L1_1.csv
L1_2.csv
RIGHT!
298/
M1/
L1_1.csv
L1_2.csv
278/
M1/
L1_1.csv
L1_2.csv
.fasta files are missing in the input folder tree. If Apply FASTA variable is set to True, a .fasta file must exist in each Y axis folder. Farseer-NMR found that some .fasta files are missing while others are present. Please look for those missing.
Nothing to do with these files. File type you tried to read is not compatible with Farseer-NMR. There's nothing to be done with those files. Please select a compatible file type: .csv or .fasta.
Empty Data points. Farseer-NMR as detected an empty data point in the peaklist files. If you want to add an empty data point from a missing value, please add the header followed by a 'new line' character to that same file. Use the following header: Assign F1,Assign F2,Merit,Details,Fit Method,Vol. Method,Number,#,Position F1,Position F2,Height,Volume,Line Width F1 (Hz),Line Width F2 (Hz)
Missing tag position. The tag position should be writen as header of the *.pre file containing the information on theoretical PRE data. Header format should be, e.g. residue 40: #40.
Unexistent reference residue. The residue you have selected to use as reference for chemical shift correction does not exist or is unassigned or lost. Select a valid residue in the Settings Menu or confirm the usage flag is not activated by mistake.
Unexistent tag residue. The residue selected for the tag position in the *.pre file, for theoretical PRE, is not part of the protein sequence.
The .fasta file is shorted than the reference experiment. Is you have input a FASTA file containing the whole protein sequence, with the intent of identifying the unassigned residues, it is expected that the .fasta file has more residue entries than the reference experiment of that series. This is not the case. Confirm that the .fasta file is correct.
Comparing references that do not exist. The flags expand_lost_yy or expand_lost_zz are activated in the Run Settings. Though, there are is only one datapoint along these axis and, therefore, nothing to compare with. Leaving the flag active won't affect the calculation, though you might want to deactivate it.
Calculation along axis is activated but there are no data points. You have activated a calculation axis flag (x, y or z) but there are no data points along that given axis. The flag may be activated by mistake.
Different .fasta sizes incompatibility. .fasta files must have the same size, that is, number of residues when analysing Series along the Y axis (which is the axis that receives the .fasta files). Read further on axes restrictions. In the log file there is a section that highlights the .fasta files where you can revisit the sequences loaded to identify the wrong file.
FASTA file contains residue numbers (first or last) in between the protein's primary sequence. Which would lead to protein truncation because FASTA numeration overwrites that of original peaklists.
Select one of the available fitting options. See the Documentation file.
Farseer-NMR could not reindex this peaklist. This may be due to different situations. Reported situations are:
- Sorting peaklist rows by residue number fails due to duplicated residue. In other words, there are more than one peaks (rows) that are assigned to the same residue. For example:
14,13,9.50727,123.51900, 81SerH, 81SerN,1.53193e+06,5.61533e+07,62.14512,49.65150,1.00000,None,parabolic,peak fit
15,13,9.50737,123.52000, 81SerH, 81SerN,1.53193e+06,5.61533e+07,62.14512,49.65150,1.00000,None,parabolic,peak fit
Only one entry is allowed per residue, that is, residue numeration (81 in this case) cannot be repeated. Duplicated entries may be originated from multiple peaks sharing the same assignment in the analysis project. You can remove the extra line in the .csv directly or remove the peak itself in the analysis project.
Chemical shift values do not match dimension. This occurs when proton and nitrogen dimensions are swapped, resulting in "Position F1" and/or "Assign F1" columns referent to Nitrogen nuclei instead of Proton nuclei and vice-versa. Farseer-NMR requires that Proton is set to the F1 name and Nitrogen to F2. You can simple rename the column names, "Position F1"<->"Position F2" and "Assign F1"<->"Assign F2", regardless of their position in the table.
Loading NmrView/NmrDraw peaklists require additional information. These two peaklist formats lack information on the residue types and only have residue number information. Therefore, Farseer-NMR requires the selection of a FASTA file to correctly parse these peaklists. Even if you don't want to expand the peaklists to a whole FASTA sequence (to identify the unassigned residues), a file should be selected to allow peaklist parsing. Do this by selecting a FASTA file the same way as for any other situation, under the FASTA submenu. Additionally, you should introduce the correct FASTA starting number, otherwise correct linking of the residue number with the residue nature will fail. If, additionally, you want then to use the FASTA file during your run, simply check the Apply FASTA box.
Required HEX colour. Parameters that configure for generating colour gradients require that initial and end colours are input in the JSON config file as HEX code. The Farseer-NMR GUI already saves colours for these parameters in HEX code, if you are encountering this error most likely is because the colours in these parameters were manually configured and do not match the requirements. Please review them, these are:
- ["dpre_osci_settings"]["color_init"]
- ["dpre_osci_settings"]["color_end"]
- ["cs_scatter_settings"]["mk_start_color"]
- ["cs_scatter_settings"]["mk_end_color"]
- ["cs_scatter_flower_settings"]["mk_start_color"]
- ["cs_scatter_flower_settings"]["mk_end_color"]
Peaklists proposed for series along an axis have different lengths. Farseer-NMR can only generate series with peaklists of the same length, i.e. number for rows or number of residue entries. The most common occurrence of this error is when analysing along Y or Z and NO FASTA file was given or the flags applyFASTA, expand_lost_yy and expand_lost_zz were not activated, leading to hanging residues (rows) in certain peaklists that were not added across the series or dataset. expand_lost_yy and expand_lost_zz flags can be activated under the GUI menu Search lost residues across axes. Consider that if you are analysing paramagnetic NMR data, expand_lost_zz flag should be set to true in order to identify those residues that were lost beyond detection just by the addition of the paramagnetic tag. The same principle applies when comparing different constructs (of the same length) along the Y axis, although in this case, a missing .fasta file is the most probable reason to raise WET#28.
Misleading characters in peaklist or absent information. In order to avoid parsing errors, Farseer-NMR searches for the presence of misleading characters in the peaklists. Those may come from the most varied reasons: typos or rows with no assignment information are the most common. Specifications:
- Assign F1 and Assign F2 columns allow only digit chars. WET is risen whenever regex
\Wis found. - Other float containing columns are scanned for
\Wexcept+and.. Farseer-NMR identifies the presence of those characters but those should be removed manually by the user. - negative values on Height and Volume. Farseer-NMR cannot yet analyse intensity information from peaks with negative phase. These values should be turned positive.
Could not read a peaklist file. The indicated peaklist has a valid extension though Farseer-NMR could not read it. This is mostly due to a wrong peaklist formatting style. Please review the valid peaklist files under our Documentation folder. This issues may be tricky to solve, please do not hesitate in contacting us through our mailing list for further help.
Some users in Linux have experienced the following error:
This application failed to start because it could not find or load the Qt platform plugin "xcb"
in "".
Available platform plugins are: minimal, offscreen, xcb.
Reinstalling the application may fix this problem.
Solution: add the following to the beginning of your rinning bash script or terminal env: "LD_LIBRARY_PATH=/usr/lib"