A Detailed Description of ZNova Parameters
The Build ZNova Parameters File screen shown with default options:
The image below is a clickable image map. Click on any region of the image below to find out about that particular parameter.
It may seem like there are a lot of parameters. However, most you will never need to change. Many (but not all) of the parameters can be accessed through the use of command line options if you are processing a spectrum or data file from a command prompt. The command line options can also be specified in the Xcalibur sample list in the ZNova Params field. Use the ZNova-ize Sequence utility to add parameter strings to your Xcalibur sample list. Please note that the command line options are case sensitive. See the description and examples shown below.
The Toolbar:
The buttons on the toolbar have the following functions:
- Open a parameter file
- Save a parameter file
- Save a parameter file as a different name
- Clear all current parameter settings on the form
- Restore default parameters
- Access the help pages on the ProMass website
- Process the current spectrum on the clipboard using the currently displayed params
- Done
If you forget the toolbar or the other parameter functions, just move your mouse over the toolbar and help text will guide you. Parameter help text appears in the status area at the bottom of the Build ZNova Params box.
Deconvolution Parmeters:
- Adduct Ion = mass
- default setting: 1.0079
- command line option: -a mass
- example: xznova ñd ìc:\mydata\oligo.txtî ña ñ1.0079
The Adduct Ion option sets the mass of the adduct ion that is providing the charge in the ESI mass spectrum. For positive ions, you would normally set this to 1.0079 (for a proton), for negative ions you would normally set this to ñ1.0079. If left blank, the adduct mass will default to 1.0079.
- Comprehensive Deconvolution = checked | unchecked
- default setting: unchecked
- command line option: ñx
- example: xznova ñd ìc:\mydata\myoglobin.txtî ñx
(Previously named the Exhaustive deconvolution option) Normally, when ZNova deconvolutes a spectrum it will remove peaks of found components as the deconvolution proceeds. If the data contain overlapping signals from a very complex mixture, you may want to use the Comprehensive deconvolution mode. In the Comprehensive mode, each m/z peak is evaluated individually and no known component peaks are removed during the deconvolution process. As a result, Comprehensive mode can be somewhat slower than the normal mode of operation. You may want to limit your deconvolution range (e.g., with Minimum Mass and Maximum Mass) to the region of interest to speed the processing. To use Comprehensive mode click the Comprehensive deconvolution checkbox or use the ñx command line option. The Maximum Components and Maximum Iterations options have no effect under Comprehensive mode.
- PeakWidth = p
- default setting: 3
- command line option: ñp pw
- example: xznova ñd ìc:\mydata\bighairyprotein.txtî ñp 5
The Peak Width parameter is used to set the width (at peak base) in the raw m/z data. Typically this is set to a value between 1 ñ 10. As components are identified, the Peak Width setting is used to determine which neighboring datapoints are transformed to the zero charge domain. This setting is also used to remove data points during processing as components are found in the normal (non-Comprehensive) deconvolution mode.
- Merge Width = m
- default setting: 0.5
- command line option: ñm merge
- example: xznova ñd ìc:\mydata\bighairyprotein.txtî ñp 5 ñm 0.3
The Merge Width option determines whether intensity data in a window around the detected peaks are merged (summed) during the scoring process. If set to 0, merging is turned off, and only the signals within the m/z tolerance window are considered during scoring. For profile data, the m/z tolerance window is 0.1 m/z units, and 0.5 m/z units for centroided data. The Merge Width option allows one to obtain a centroid of the deconvoluted peaks. Typically, the Merge Width is set to 10 - 50% of the peak width at the base (e.g 0.2-0.5 for 2 m/z unit peaks. The Merge Width option has no effect on the intensities of transformed peaks if the output is profile deconvolution. If the output is centroided, the intensities are summed in the Merge Width window.
- Normalize Scores = charge
- default setting: 1
- command line option: -n charge
- example: xznova ñd ìc:\mydata\peptide.txtî ñp 2 ñn 3
With the Normalize Scores option set to a charge state value, ZNova will begin normalizing the scores of the candidate charge states based on the ratio of charge states that were observed to those that were possible over the input m/z range. This prevents false scoring of a narrow distribution of peaks that occur due to random noise events in the input data. The normalization process has the effect of providing higher weight to those components that show a full series of predicted peaks. The normalization process has no effect on the intensities of the deconvoluted peaks, it only effects the score that is calculated to determine the charge state of each peak. If you have a protein with a very narrow charge state distribution, you may want to set this option to 0 or change the input First m/z and/or Last m/z values to bracket your data range to get optimum deconvolution. When score normalization is activated, the normalization is applied starting at the chosen charge state. For most applications, you will want to leave normalization on and set to 1. For tryptic peptide applications you will probably want to set this to 3 or 4, which will effectively enhance the scoring of lower charge states relative to higher charge states.
The First m/z option allows you to specify the starting m/z in the raw input data. This is useful if you have chemical background at the low m/z end of the spectrum that you want to exclude from the deconvolution, or the charge state distribution does not cover the full acquired range of the spectrum. If left blank, ZNova will use the lowest m/z in the acquired data.
- Last m/z = mz | blank
- default setting: blank
- command line option: ñl mz
- example: xznova ñd ìc:\mydata\peptide.txtî ñf 700 ñl 1500
The Last m/z option allows you to specify the ending m/z in the raw input data. This is useful if you have chemical background at the high m/z end of the spectrum, or the charge state distribution does not cover the full acquired range of the spectrum. If left blank, ZNova will use the highest m/z in the acquired data.
- Minimum Mass = mass
- default setting: 5000
- command line option: ñF mass
- example: xznova ñd ìc:\mydata\protein.txtî ñf 700 ñl 1500 ñF 5000 ñL 30000
The Minimum Mass option allows you to specify the low mass limit of the output deconvolution. Note that computation time increases with deconvolution mass range.
- Maximum Mass = mass
- command line option: ñL mass
- default setting: 20000
- example: xznova ñd ìc:\mydata\protein.txtî ñf 700 ñl 1500 ñF 5000 ñL 30000
The Maximum Mass option allows you to specify the high mass limit of the output deconvolution. Note that computation time increases with deconvolution mass range.
- Smoothing Width = sw | blank
- command line option: ñw sw
- default setting: 3
- example: xznova ñd ìc:\mydata\bighairyprotein.txtî ñw 5
The Smoothing Width option allows you to specify the width of the moving average that is performed before data are deconvoluted. Smoothing improves the signal-to-noise ratio of the input data. If left blank or set to 0, smoothing will not be performed. Smoothing will not be performed on centroid input data if the centroid plot option is checked. The smoothing algorithm is a simple moving average (e.g., not a Savitzky-Golay algorithm), therefore, excessive smooth width can broaden the peaks in the input data. A smooth width of 3 is adequate for most applications.
- Smoothing Cycles = n | blank
- default setting: 2
- no command line option available
The Smoothing Cycles option allows you to specify the number of successive smooths of the input data that are performed. If left blank to or set to 0, smoothing will not be performed. Smoothing will not be performed on centroid input data if the centroid plot option is checked.
- Intensity threshold % = thresh | blank
- default setting: 0
- command line option: ñt thresh
- example: xznova ñd ìc:\mydata\bighairyprotein.txtî ñt 10
The threshold option allows you to specify the % relative intensity noise threshold for the input data. If threshold is set to 0 or left blank, the noise threshold is determined automatically by taking the average signal level of the input data and mutiplying this number by the Noise Multiplier parameter. Generally, automatic noise thresholding is preferred. However, in some instances, the automatic thresholding may over-estimate the noise level if the spectrum is particularly noisy or has a sloping or uneven baseline that is not effectively removed. The threshold parameter may be helpful in these cases. Generally, a value of 1-5% relative intensity threshold is a good starting point.
- Minimum Score = score
- default setting: 2
- command line option: ñs score
- example: xznova ñd ìc:\mydata\bighairyprotein.txtî ñs 3
The Minimum Score option allows you to specify the Minimum Score threshold. Scores below this value will not be transformed to the zero charge spectrum. This prevents the algorithm from transforming random noise. If Minimum Score is left blank, the Minimum Score will be set to the log(Noise Multiplier). Setting Minimum Score to 0 is allowed, however, you should be aware that without a Minimum Score threshold, the output data could contain many more signals that are due to transformation of random noise signals. This will be particularly evident in low S/N data.
- Maximum Components = comp
- default setting: 0
- command line option: ñc comp
- example: xznova ñd ìc:\mydata\bighairyprotein.txtî ñc 5
The Maximum Components option allows you to specify the maximum number of components to find in the deconvoluted data before stopping execution of the program. If you are only interested in the most abundant component you can set Maximum Components to 1 for example. If set to 0 or left blank, ZNova will continue to process data until all data points are processed or one of the other options for stopping the program (e.g., Maximum Iterations, Maximum Time) has been satisfied. This option has no effect if the Comprehensive deconvolution mode is used.
- Maximum Iterations = iter
- default setting: 0
- command line option: ñi iter
- example: xznova ñD ìc:\mydata\bighairyprotein.txtî ñi 100
The Maximum Iterations option allows you to specify the maximum number of iterations before stopping execution of the program. One iteration is defined as the testing of one data point (+/- peak width) to find its true charge state. If set to 0 or left blank, ZNova will continue to process data until all data points are processed or one of the other options for stopping the program (e.g., Maximum Components, Maximum Time) has been satisfied. This option has no effect if the Comprehensive deconvolution mode is used.
- Maximum Time = time
- default setting: 0
- command line option: ñe time
- example: xznova ñD ìc:\mydata\bighairyprotein.txtî ñe 60
The Maximum Time option allows you to specify the maximum amount of ìdeadî iteration time (in seconds) before stopping execution of the program. ìDeadî iteration time is computation time that has elapsed since ZNova found the last real component. Since ZNova sorts the mass/intensity pairs by intensity, you will generally find that if ZNova is iterating and not finding anything for several seconds that it is cyphering on noise. Therefore, you could use this option when deconvoluting a wide mass range (e.g., 5000 ñ 90000) to exit after 10 ñ 30 seconds of ìdead timeî. If this field is left blank or set to zero, ZNova will continue to process data until all data points are processed or one of the other options for stopping the program (e.g., Maximum Components, Maximum Iterations) has been satisfied. The Maximum Time option may be used in Comprehensive or normal deconvolution mode.
- Weight outputs = checked | unchecked
- default setting: unchecked
- no command line option available
With weighting turned on, the program will weight the intensities of the output deconvoluted peaks with score. This enhances the intensities of higher scoring peaks and may help to pick out ìrealî signals in noisy data. This feature is only active in centroid mode or when the Centroid Plot option is checked. If accurate relative abundances of peaks are important in your application, this feature should be turned off.
- Exclusion List = (m1, m2, m3, m4-m5)
- default setting: blank
- no command line option available
M/z values entered in the Exclusion List box will be excluded from the deconvolution. This is often useful if you have background peak(s) that you want to exclude from processing. The m/z value ± 0.5 times the Peak Width will be exluded from processing. You enter multiple values separated by commas. You can enter a range by entering two numbers separated by a dash (e.g, 400-700). You can combine ranges and comma separated values.
Display Parameters:
- Centroid Plot = checked | unchecked
- default setting: unchecked
- no command line option available
With Centroid Plot turned on (checked), the program will produce ìstickî deconvolution data instead of profile (continuum) data. The intensities of the output centroided spectra are obtained from a centroiding process and are summed in the Merge Width window. The masses are the intensity-weighted average mass obtained in the merge window. You can have ZNova produce centroided output from either profile or centroid input spectra. If you acquired your data in centroid mode, and Centroid Plot is unchecked, the data will be interpolated to 0.1 u and will be treated as profile data. If the data type is profile and Centroid Plot is unchecked, the peaks will still be labeled with their centroid-averaged (Merge Width > 0) or average masses (Merge Width = 0), however, the intensities reported will be those obtained from the profile deconvolution process.
- Image Format = png | jpeg
- no command line option available
- default setting: png
The Image Format option specifies whether the output graphics files will be in png or jpeg format. In general, png output is best for web browser viewing. JPEG format may be better if you are planning on pasting the resulting graphics into Word or PowerPoint.
- Graph Size = X Y
- no command line option available
- default setting: 800 x 600
The Graph Size option specifies the image size in X and Y pixels for the output spectra and chromatogram images. An 800 x 600 size is generally optimum for viewing images in a web browser if your screen resolution is 1024 x 768 or better. When printing, 800 x 600 will fit nicely on a page when printed in landscape mode.
- Zoom Range = z | z1 - z2
- no command line option available
- default setting: 500
The Zoom Range option specifies a window where ZNova will plot a zoomed deconvoluted spectrum. If a single value, e.g., 500, is entered in the Zoom Range box, ZNova will plot a ± 500 Da window around the most intense ion in the deconvoluted spectrum. You can also enter an explicit range, by entering 2 numbers separated by a dash. For example, if you enter 15000 - 18000, ZNova will plot a zoom deconvoluted spectrum over this range, regardless of the value of the base peak deconvoluted mass.
- Logo File Name = logo.gif
- no command line option available
- default setting: promass.gif
The Logo File Name option allows you to customize the HTML report with your own corporate logo. The logo will appear at the upper right hand corner of the report. Either GIF or JPEG file formats may be used.
- Plot Data = checked | unchecked
- no command line option available
- default setting: checked
The Plot Data option specifies whether ZNova will produce graphic output. If Plot Data is unchecked only text-based output will be produced, i.e., without spectra and chromatogram images.
Results Location Parameters:
- Directory Format = raw | file | date | explicit
- no command line option available
- default setting: raw
The Directory Format option specifies where ZNova will store output files. Output files consist of graphic files, text-based results, and the chromatogram and summary HTML reports for each data file processed. If the raw option is selected, the output files will be stored in the raw data path in a subdirectory named promass_results. If the file option is selected, ZNova will create a new subdirectory under Results Directory for each data file processed based on the input data file name. If the date option is used, ZNova will create a new subdirectory under Results Directory based on the current date and will direct all output files to this directory. If the explicit option is used, ZNova will direct all output to a directory that you explicitly select in the Results Directory box.
- Results Directory = C:\Program Files\ProMassXcali
- no command line option available
- default setting: ProMass install directory
The Results Directory option specifies the base directory where ZNova will store its output files. If either the file or date options are selected in the Directory Format box, sub-directories are automatically created under the Results Directory. When the explicit Directory Format option is selected, all results are stored in the Results Directory indicated. For the raw Directory Format option, the Results Directory is automatically set to the data file path if it can be determined at run time.
- Overview of how output files are stored
- raw Directory Format (default):
- drive:\data path\promass_results
- All results files contained in a single directory
- file Directory Format:
- drive:\Results Directory\datafile basename
- Results from each data file are in a separate subdirectory named after the data file.
- date Directory Format:
- drive:\Results Directory\month_day_year
- Results from each processing day are in a separate subdirectory named after the processing date.
- explicit Directory Format:
- drive:\Results Directory
- All results files contained in a single directory
Noise and Data Reduction Parameters:
- Filter Data = checked | unchecked
- default setting: checked
- no command line option available
With data filtering turned on, the program will remove all data points that are below the minimum threshold signal level before processing. The minimum threshold signal is determined by the taking the total average signal level for the spectrum and multiplying by the Noise Multiplier parameter. Alternatively, the user may specify a % relative intensity noise Threshold. Filtering data speeds processing since there are fewer data points in the input spectrum. It is recommended that this option always be turned on.
- Baseline Factor = base
- default setting: 0.5
- command line option: ñb baselineFactor
- example: xznova ñd ìc:\mydata\bighairyprotein.txtî ñb 1.2
With baseline removal activated, the program will automatically determine the noise baseline in the input data and remove this baseline noise from the input spectrum. The range of typical values for Baseline Factor is 0.5 to 1. Use a value of 0 to turn off baseline removal. It is recommended that both baseline removal and data filtering always be turned on, since a flat baseline will allow for the most accurate determination of the data noise level.
- Noise Multiplier = noise
- default setting: 2
- no command line option available
The Noise Multiplier option allows you to specify the noise threshold multiple. A datapoint must have a signal-to-noise ratio of Noise Multiplier to be considered a valid real signal. Typically this parameter is set to a value between 2 - 5. The minimum value for Noise Multiplier is 1.1. If you set Noise Multiplier to less than 1.1, a value of 2 will be used. When data filtering is turned on with the Filter Data option, data points with a S/N below the Noise Multiplier value are discarded from the input data before deconvolution.
- Top N Points = topn
- default setting: 0
- no command line option available
The Top N Points option allows you to instruct ZNova to only consider the top N data points in the input spectrum for deconvolution. In most cases this option is not used. However, this option may be helpful if you are only interested in major components and you are interested in maximum algorithm speed. Generally, this option is best used on centroided data. This option may also be useful for peptides, since most of the signal is generally concentrated in a few peaks. A value of 0 turns this feature off.
Options only available on the command line:
- -d
- example: xznova ñd ìc:\mydata\myoglobin.txtî
The -d option is used to specify a raw mass/intensity text-based datafile as input to ZNova. A ñd, -X, or ñC option is required in order for ZNova to run. If the file contains a path make sure to enclose the filename in double quotes, as shown above. If you want to reprocess a previously generated ProMass text-based results file, do not use ñd, use the ñX option instead. The text file should be in white space-delimited mass/intensity format as shown below: 500.058098 78
500.158082 14
500.258066 0
500.358050 48
500.458033 6
500.558017 0
500.658001 35
500.757985 109
- -X
- example: xznova ñX ìc:\xcalibur\data\myolcms.txtî
The ñX option is used to specify the ProMass text-based datafile to process. If the file contains a path make sure to enclose the filename in double quotes, as shown above. If you want to process a plain mass-intensity text file, do not use ñX, use the ñd option instead. ProMass text files are produced automatically from the ProMass program processed from the Xcalibur Sequence Setup program. You could use the -X command line option to reprocess an Xcalibur text file that was created previously. The ProMass text files are saved in the same directory as the acquired data.
The ñC option is used to specify processing of a spectrum from data on the Windows clipboard. The program will wait for data to appear on the clipboard before processing will begin or will complain if the data is not in the correct format. For example, to process a spectrum directly from Xcalibur/QualBrowser, activate the spectrum view, right mouse click the spectrum to get the Export menu and select Clipboard (Exact Mass). Then enter xznova ñC (and any other optional parameter strings) in a console window to process the current clipboard spectrum.
- -P
- example: xznova ñX ìc:\xcalibur\data\myolcms.txtî -P ìc:\promassxcali\znova\znova.paramsî
The ñP option is used to specify the parameter file to use when running ZNova in lieu of the default znova.params parameter file. ZNova uses the following hierarchy when applying input parameters. If no parameter file is specified, ZNova first looks in the current local directory for a znova.params file. If no parameter file is found, ZNova looks in the ProMassXcali\ZNova install directory and reads in the default znova.params file. If ZNova cannot find a valid parameter file, a fatal error is generated. When a parameter file is specified with the ñP option, ZNova will use this file. If this file is not found, the install directory znova.params file will be applied. If you specify a parameter file along with other command line options, the options specified will override those same options specified in the parameter file.
|
