PBFGen Usage

PBFGen reads in instrument data files and converts them to PBF, the format used elsewhere in the Informed-Proteomics suite.

Input format support

Supported natively

• *.mzML and *.mzML.gz

Supported with additional software

Note: x86/x64 matter with these additional softwares. PBFGen and other Informed-Proteomics tools that try to read these formats will give you an error if you do not have a correct version installed, telling you which version you need to install.

• *.raw (Thermo *.raw format)
  • Requires an installation of Thermo's MSFileReader, available here (requires registration)
  • Also can be read via ProteoWizard (see the following)
• The following formats (and others) can be read if ProteoWizard is installed (Download here):
  • *.mzXML, *.mzXML.gz
  • *.raw (Thermo *.raw format, also can be read via Thermo MSFileReader)
  • *.mzML, *.mzML.gz (Supported internally, but used for reading gzipped mzML files when available for speed/memory usage purposes)
  • *.mgf (Mascot Generic Format)
  • *.d folders (Agilent and Bruker .d folder formats)
  • *.wiff (AB Sciex *.wiff format)
  • *.u2, FID (older Bruker formats)
  • *.raw folders (Waters *.raw folder format)
  • *.lcd (Shimadzu *.lcd format)

Input format warning

We have only fully tested PBFGen, ProMex and MSPathFinder on *.mzML, *.mzML.gz, *.raw (Thermo), *.mzXML, and *.mzXML.gz formats (with centroided data in the mzML and mzXML formats; with Thermo *.raw we use Thermo's centroiding for profile data). While the full list of files above can be read for conversion into PBF, their valid conversion into PBF or their utility for performing feature finding or database searches is not tested and cannot be guaranteed in any fashion. For best results, the file provided to PBFGen or other tools in the Informed-Proteomics suite should contain scan data with MS and MS/MS data, with centroided peaks (when not a vendor format).

PBFGen Syntax

PbfGen.exe -s [RawFilePath (*.raw or directory)] [-o OutputDir]

-s [RawFilePath (*.raw or directory)]

• Path to a file/folder of one of the supported input formats.
  • Can also perform batch conversion of all *.raw files in a directory
    • Support for directories containing multiple files of other supported formats is not implemented at this time.

[-o OutputDir] (optional)

• If not specified, the converted file(s) will be written to the same directory as the input file.
• If specified, the converted file(s) will be written to the directory at the provided path.

Examples

These command will create MyDataset.pbf in the same directory as MyDataset.raw or MyDataset.mzML:

PbfGen.exe -s MyDataset.raw
PbfGen.exe -s MyDataset.mzML

These command will create MyDataset.pbf in the directory C:\WorkFolder:

PbfGen.exe -s MyDataset.raw -o C:\WorkFolder
PbfGen.exe -s MyDataset.mzML -o C:\WorkFolder

System Requirements and Recommendations

Minimum required:

• .NET 4.5.1 installed
• Will probably try to run on anything where you can install .NET 4.5.1.

Minimum recommended:

• 2.4 GHz, quad-core CPU
• 16 GB RAM
• Windows 7 or newer
• 250 GB hard drive

PBFGen needs a good amount of RAM available when writing the MS1 and MSn Full XICs, due to a need to hold a larger number of peaks in memory and sort them before writing them to disk. It will try to scale the amount it holds it memory according to the amount of free memory when it starts, but it does have a lower limit that it requires based on the number of scans in the input file. The amount of memory available can drastically affect the time it takes to process and write the full XICs to disk; the more memory available, the less time it may take.

The output file size will often be larger than the input file size, due to the data duplication in the PBF format to reduce the processing time for other programs. The primary exception to this is with data that is stored in profile mode, since the data will be centroided (if possible) before it is written in the PBF format.