Resources - MSCypher

Resources - MSCypher

Researcher: 

MSCypher

MSCypher is a command-line workflow currently running on Microsoft Windows OS and Linux platforms. The input to MSCypher are MGF files which can be generated by Maxquant and the APLToMGFConverter program or MSConvert (part of Proteowizard).

MSCypher’s novel contribution to the field is better support for identifying chimeric (co-fragmented) peptide features which are common in complex datasets, integrated machine learning, end-to-end label-free quantitative analysis and extensive support for third-party tools.

The workflow currently supports the following steps:

  1. Customised sequence database creation using the MSPnr100 sequence database as input. Any FASTA sequence database can be created by specifying multiple taxonomic identifiers by species. These sequence databases are written to disk for future use.
  2. MSConvert support for converting RAW instrument data to MGF input files.
  3. DeNovo peptide sequencing using the PepNovo program. All denovo sequences are written into the combined MSCypher table for comparison with database searching output.
  4. Database searching using the Digger search algorithm which is fast and sensitive.
  5. Retention time prediction using the program Elude.
  6. Machine learning to classify peptide identifications. Current tools are RandomForest, Percolator and Newbee (a newly developed tool in our laboratory). The RandomForest classifier is the default tool in this category.
  7. Protein Inference at 1%FDR.
  8. Quantitative proteomics at the peptide and protein level supporting spectral counting via NSAF and intensity based quant and statistics using MSstats and Limma and an LFQ R package for automated generation of volcano plots, QC metrics and others.
  9. Visualisation and support for other tools such as Protter, Skyline, Peptigram and others.
  10. Support for FlashLFQ under development.

Download Windows MSCypher or Linux MSCypher and example input data. The example input data (mgf files) should be copied and unzipped to the data directory. Please e-mail Eugene Kapp for assistance.