Each module uses a configuration file (INI format), but some of the parameters it contains can also be specified through the command line. To see the parameters available through the command line and their descriptions, run each module with the help option, -h (e.g., python DMcalibrator.py -h).
This module adapts search engine results from Comet-PTM, ReCom, and MSFragger for analysis with SHIFTS. It removes any extra lines before the column headers and adds a column containing the input file name. If the input file type does not have extra lines before the headers and already includes an input file name column, this step can be skipped. Every subsequent module assumes that the first line of the input file contains the column headers.
For MSFragger inputs, the module calculates the left and right positions of modifications.
In all cases, a 'Spectrum_File' column is added.
-
Example:
python SHIFTSadapter.py -i path/to/input/file -o path/to/output/directory
-
Input:
- A tab-separated file containing results from Comet-PTM, ReCom, or MSFragger search engines.
-
Output:
- A tab-separated/Feather file (without extra header lines).
- A log file containing header information.
-
This module will remove scan duplicates, keeping only the scan candidate that meets the following criteria: highest rank, highest score (if there are several with the same rank), highest sp_score (if there are several with the same score).
-
Example:
python DuplicateRemover.py -i path/to/input/file -c path/to/config/file
-
Input:
- A file from SHIFTSadapter output.
- Command-line or Config INI file with the parameters:
-s (--scan) = ScanID column name (case sensitive, column must contain an unique ID for each scan) -n (--num) = Rank column name (case sensitive) -x (--score) = Score column name (case sensitive) -p (--spscore) = SpScore column name (case sensitive) -
Output:
- The same file, with duplicates removed.
- A log file.
-
This module calculates the calibrated values for experimental masses and deltamasses. The dataset is filtered by a user-defined minimum score and maximum ppm, to obtain a subset of high-quality identifications that can be used for calibration. The median relative error (median of the absolute error column divided by the MZ column) is calculated to obtain the ‘alpha’ value: alpha = (df_filtered[abs_error]/df_filtered[mzcolumn]).median() Then, this value is used to calibrate the experimental MZ values of the entire dataset as follows:
cal_exp_mz = exp_mz * (1 - alpha)
-
Example:
python DMcalibrator.py -i path/to/input/file -c path/to/config/file
-
Input:
- A file from SHIFTSadapter or DuplicateRemover output.
- A configuration file (INI). The following parameters are used:
score_min: Minimum score to filter by ppm_max: Maximum PPM error to filter by scorecolumn: Name of column containing score (case-sensitive) zcolumn: Name of column containing charge (case-sensitive) mzcolumn: Name of column containing experimental M/Z (case-sensitive) seqcolumn: Name of column containing sequence (case-sensitive) seqdmcolumn: Name of column containing sequence with deltamass within square brackets (case-sensitive) proteincolumn: Name of column containing protein IDs (case-sensitive) decoyprefix: Prefix used for decoy protein IDs (case-sensitive) calseqcolumn: Name of output column containing sequence with calibrated deltamass (case-sensitive) decimal_places: Number of decimal places to use in sequence+deltamass output columns
Also, the "config" ini file has the three sections:
Aminoacid masses Fixed modifications for each aminoacid (if there are any) Other masses (Hydrogen, Oxygen, proton) -
Output:
- The same file with additional columns for theoretical MH and MZ, calibrated MH and MZ, calibrated deltamass MH and MZ, absolute error, ppm error, sequence with calibrated deltamass.
- A log file containing systematic error before and after calibration, alpha, StdDevMAD_ppm (the mean absolute deviation for the ppm error), number of PSMs before and after filtering (size of the filtered, high quality dataset that was used for calibration)
-
This module concatenates a group of files, generates a histogram grouped by deltamass bins of a user-specified width, and calculates the frequency and slopes (first and second derivatives) for each bin, using linear regression.
-
Example:
python PeakModeller.py -i path/to/input/file -c path/to/config/file
-
Input:
- A text file containing a list of calibrated files to be modelled together
- A configuration file (INI). The following parameters are used:
bins: Width of the bins slope_points: Number of points (bins) to use for slope calculation smooth_points: Number of points (bins) to use for smoothing separate_modelling: Flag to model histograms of target and decoy separately or together.
-
Output:
- A tab-separated file containing all the input files together (DMTable), including a new ‘Filename’ column that tracks the origin of each PSM.
- A tab-separated file containing the histogram (DMHistogram). This table contains columns with the bin, bin midpoint, frequency, slope1 (first derivative) and slope2 (second derivative).
- A log file.
-
This module filters a DMHistogram table (as generated by PeakModeller) according to user-specified thresholds for the columns containing slope and frequency. This filtered dataset is used to calculate a list of apexes. For each bin (row) where the value of the first derivative (slope) changes from positive to negative, the exact DM value where the first derivative is 0 is interpolated using a user-defined number of points surrounding the apex bin. Taking the x (bin midpoint) and y (first derivative) values for those points it will identify the apex as the x value where y equals 0.
-
Example:
python PeakSelector.py -i path/to/input/file -c path/to/config/file
-
Input:
- A DMHistogram table (as generated by PeakModeller) to be filtered.
- A configuration file (INI). The following parameters are used:
frequency = 0 # Threshold for number of PSMs apex_points = 4 # Number of points (bins) to use for apex calculation x2= -0.0000006 # Quadratic term of the function to calculate the zones to search max slope m= 1.0002 # Lineal term of the function to calculate the zones to search max slope intercept= -0.0367 # Intercept of the function to calculate the zones to search max slope dm0= -230 # Minimum mass of the spectra dm1= 500 # Maximum mass of the spectra ci_interval= 84.13 # % of 1-tailed CI for removing outliers (84.13% recommended)
-
Output:
- A text file containing the apex list.
- A log file containing the number of apexes that were calculated.
-
This module will assign every PSM to the closest deltamass peak found in the provided apex list and identify it as either belonging to that peak, or as an orphan. To do so, it will calculate the absolute distance between the deltamass value and the assigned peak and then the error in ppm:
distance = abs(assigned_peak - delta_MH)
distance_ppm = (distance / (theoretical_mass + assigned_peak)) * 1e6
If that value is larger than the user-defined threshold, it will be considered an orphan.
-
Example:
python PeakAssignator.py -i path/to/input/file -c path/to/config/file
-
Input:
- A DMtable.
- An apex list.
- A configuration file (INI). The following parameters are used:
ppm_max: Maximum ppm difference for peak assignation peak_label: Label for peaks orphan_label: Label for orphans caldeltamh_column: Name of column containing calibrated Delta MH theomh_column: Name of column containing theoretical MH for ppm error calculation closestpeak_column: Output column that will contain the closest peak peak_column: Output column that will contain the peak/orphan labels deltamass_column: Output column that will contain the assigned deltamass ppm_column: Output column that will contain the ppm error mod_peptide_column: Name of column containing sequence with deltamass in XXX[DM]XXX or XXXXXX_DM format (case-sensitive) assign_seq_column: Name of output column containing sequence with assigned deltamass (case-sensitive) decimal_places: Number of decimal places to use in sequence+deltamass output columns
-
Output:
- A DMtable with additional columns for the closest peak, assignation as peak or orphan, assigned deltamass, ppm error, sequence with assigned deltamass.
- A log file.
-
This module will calculate global, local, and peak FDR values for a DMtable subdivided by experiment. Global FDR is calculated per experiment, local and peak FDR is calculated for the whole dataset. Experiment groups are created using the information in the “Filename” column (created by the DMcalibrator module) of the DMTable, and a user-provided list that assigns each filename to a group. For the global FDR, rather than taking an entire experiment, this module will separate it in two deltamass regions (defined by the parameter dm_region_limit, default value -56) and calculate a global FDR for each region.
-
Example:
python PeakFDRer.py -i path/to/input/file -c path/to/config/file
-
Input:
- A DMtable.
- A tab-separated file containing, in order:
A column with the batch name A column with the experiment name A column with the file path (must match the file paths in the Filename column of your DMtable).- A configuration file (INI). The following parameters are used:
score_column: Name of column containing score (case-sensitive) dm_region_limit: Deltamass region limit for Global FDR. Two regions will be created, DM equal to or above and DM below this value dm_column: Name of column containing deltamass for region limits (case-sensitive) peak_outlier_value: Peak FDR value to be assigned to orphans (default=1) peak_label: Label for peaks peak_column: Output column that will contain the peak/orphan labels caldeltamh_column: Name of column containing calibrated Delta MH closestpeak_column: Output column that will contain the closest peak
-
Output:
- One DMtable for each batch with additional columns for the global, local and peak FDR ranks and values.
- A log file.
-
DM0Solver is a module that detects if a modified peptide has a Δmass belonging to a list provided by the user (Table 1), for that purpose absolute error is calculated. In such a case, the Δmass is appended at the end of the clean sequence (DM0Sequence output column) and the corresponding label is added in an additional column named DM0Label. If Δmass does not belong to the list, the module passes the modified sequence without any modification to the output columns.
-
Example:
python DM0Solver.py -i path/to/input/file -c path/to/config/file
-
Input:
- A DMtable.
- A tab-separated file containing, in order:
A column with the batch name A column with the experiment name A column with the file path (must match the file paths in the Filename column of your DMtable).- A configuration file (INI). The following parameters are used:
score_column: Name of column containing score (case-sensitive) dm_region_limit: Deltamass region limit for Global FDR. Two regions will be created, DM equal to or above and DM below this value dm_column: Name of column containing deltamass for region limits (case-sensitive) peak_outlier_value: Peak FDR value to be assigned to orphans (default=1) peak_label: Label for peaks peak_column: Output column that will contain the peak/orphan labels caldeltamh_column: Name of column containing calibrated Delta MH closestpeak_column: Output column that will contain the closest peak
-
Output:
- DM0Solver output (default suffix: “DM0S”). New columns:
- DM0Sequence_output_column_name: sequence corrected by DM0Solver.
- DM0Label_output_column_name: selected label of the list provided by the user.
- DM0Label_error_output_column_name: absolute error resulting from the selection of the label that appears in DM0Label_output_colummn_name.
- A log file.
- DM0Solver output (default suffix: “DM0S”). New columns:
-