Running MSFragger and Sage through Scaffold

Avatar
Jacob Lippincott
Follow

Scaffold 5 introduced the ability to search raw directly by connecting Scaffold to MSFragger, which can be used on Windows, macOS or Linux.

Scaffold DDA can also integrate with MSFragger and Sage in a similar fashion.

The following document provides additional information on running MSFragger in Scaffold Software. See our documentation on FragPipe for more information on configuring for Scaffold.

Notes

At this stage our direct support for MSFragger is still new. Thus, there are some known issues to be aware of. We are working to get these corrected in a future version of Scaffold. While we recommend using the most current supported version, here is the MSFragger change log for reference.

  1. Make sure you have updated to Scaffold 5.2 before using MSFragger 3.5. If using Scaffold 5.0 - 5.1.2, use MSFragger 3.4.
  2. Some digestion enzymes in the dropdown menu in the MSFragger search parameters are not currently supported. These include: pepsin 1.3, pepsin 2, Glu-C (bicarb), Gluc-C (phosphate), Asp N, Asp N (ambic), V8, Proteinase K, Trypsin/CnBr, Thermolysin, Lys N, Lys C, pepsina, endopeptidase, Trypsin K, Trypsin p DP and Lys C P DP. If interested in using an alternative digestion enzyme, consider running FragPipe and loading the data into Scaffold. Tryptic searches are working correctly
  3. There was a bug in previous MSFragger versions that caused the MGF file to not be written in some cases. This was fixed in version 3.3. If you are running a version of MSFragger prior to 3.3, please update. 
  4. The SemiTrypic digestion enzyme option reverts to Trypsin. Again, consider using FragPipe.
  5. If using MSFragger version 3.1 or 3.1.1 do not use LFDR as the scoring model. In most cases it is generally recommended to use Percolator. 

Quantitation

When run by itself or within Scaffold, MSFragger does not support precursor intensity quantitation. To get precursor values when using MSFragger, run it through FragPipe and then load the resulting files into Scaffold. Please see our Support Data Formats for more information on supported quantitation methods. 

Connecting to MSFragger

Once MSFragger has been downloaded, point Scaffold to the MSFragger JAR file using the Edit > Preferences > Paths menu. Make sure to keep the MSFragger JAR file in the folder it came in as other files in it are required for data processing. 

Note, MSFragger requires a separate license from Scaffold. Please see https://github.com/Nesvilab/MSFragger/wiki/Preparing-MSFragger for information on obtaining a license.

Allocating RAM

Larger data sets require more RAM to be allocated to Scaffold and MSFragger. While RAM can be allocated to Scaffold using the Edit > Preferences > Memory tab, MSFragger RAM needs to be allocated separately.

Navigate to the msfragger.params.template file found in your Scaffold installation directory (parameters > msfragger.params.template), and open it with a text editor. In this file you see -Xmx:8g. This is the default and it means that there is currently 8 GB of RAM allocated to MSFragger. To change the allocation simply replace this number with how much RAM you would like to allocate. For example change this to -Xmx:24g to allocate 24 GB of RAM to MSFragger. Save the file and restart Scaffold if it is running. You should now be using the new setting.

Decoys

MSFragger has specific requirements for how decoys are annotated in the FASTA file. Please see their article on FASTA title lines: https://github.com/Nesvilab/philosopher/wiki/How-to-prepare-a-protein-database#header-formatting. FASTA parsing can easily be changed in Scaffold to accommodate the correct MSFragger annotations.

MSFragger Errors

Occasionally when running MSFragger through Scaffold, you may see an Error executing MSFragger: process exited with non-zero status! error. This generic error indicates that MSFragger has crashed. Often this error indicates that MSFragger has run out of memory. Check to make sure you have allocated more RAM to MSFragger using the method above. If the error text does not indicate that a lack of memory is the issue, please send us the error message and we will be of assistance in troubleshooting. Key words like java heap size or memory in the error message indicate that the error is RAM related.

Additional Search Parameters

All of the search parameters required to complete an MSFragger search are available through the Scaffold Load Data Wizard. However, some additional parameters can be modified as needed.

Parameters can be added to the msfragger.params.template file that was edited in the previous section. Please use the closed_fragger.params file found in your MSFragger installation as a reference but do not edit this file. Copy parameters from it into the template file.

For more information on each parameter, please consult the MSFragger documentation.

The following is a list of parameters that should not be changed.

  • num_threads
  • database_name
  • precursor_mass_lower
  • precursor_mass_upper
  • precursor_mass_units
  • fragment_mass_tolerance
  • fragment_mass_units
  • calibrate_mass
  • write_calibrated_mgf
  • decoy_prefix
  • fragment_ion_series
  • search_enzyme_name
  • search_enzyme_cutafter
  • search_enzyme_butnotafter
  • allow_multiple_variable_mods_on_residue
  • output_file_extension
  • output_format

Modifications should either be set in the Scaffold Load Data Wizard or in the template file but not in both. Additionally, Scaffold does not allow for more that one variable modification on a single amino acid.  

The following parameters can be changed by adding the parameter to the msfragger.params.template file.

  • precursor_true_tolerance
  • precursor_true_units
  • isotope_error
  • mass_offsets
  • precursor_mass_mode
  • localize_delta_mass
  • delta_mass_exclude_ranges
  • precursor_charge
  • override_charge
  • max_fragment_charge
  • num_enzyme_termini
  • allow_missed_cleavage
  • digest_min_length
  • digest_max_length
  • max_variable_mods_per_peptide
  • max_variable_mods_combinations
  • minimum_peaks
  • use_topN_peaks
  • minimum_ratio
  • clear_mz_range
  • remove_precursor_peak
  • remove_precursor_range
  • intensity_transform
  • Deisotope
  • min_fragments_modelling
  • min_matched_fragments
  • output_report_topN
  • output_max_expect
  • report_alternative_proteins

 

 

Have more questions? Submit a request