Home
- input are mzXML files from Pride which where analyzed as part of the eLife paper. Priority = MHC class I peptides (not class II from Mtb mzXML files).
- Annotation of mzXML files including IEDB metadata
- 2-3 search engines: comet, myrimatch, (msgf+) for identification of MHC peptides from centroided MS/MS data.
- TPP - peptide prophet and iprophet for rescoring and FDR estimation
- NetMHCcons 1.1 for filtering/verification and annotation of identified peptides
- recomputation of the FDR for verified peptides
- spectrast for swath library creation
- Data visualization (heat map (see figure 2 in eLife) and distribution of predicted HLA binding affinities (see Figure 2—figure supplement 4 and Figure 2—figure supplement 5; other visualization tools are optional)
- Create an R/Bioconductor package and an open source graphical user interface
- Validation using a new HLA peptidomics dataset (Buffy samples acquired in Tübingen; data stored in OpenBis)
- Build the database, design website and implementation (www.systemhcatlas.org)
- Publication
The workflow is implemented using the applicake plumbing toolkiet
Open questions:
- how does the output of the workflow looks like:
- meta information about the sample
- annotation of the identified peptides
pip install Uniprot
pip install lxml
pip install ruffus
pip install pandas
sudo apt-get install git
git clone https://github.com/systemhc/systemhc
git clone https://github.com/SysteMHC/SysteMHC_Binaries
git clone https://github.com/SysteMHC/SysteMHC_Data
git clone https://github.com/applicake-tools/applicake2
git clone https://github.com/applicake-tools/searchcake
git clone https://github.com/applicake-tools/searchcake_binaries
To run the application first, change into the installation directory. On the systemhc server, it is ~/prog change into
~prog/systemhc
source setup.sh
This sets two variables:
export SYSTEMHC=$HOME/prog
export LC_ALL=C
To run the workflow go to:
/mnt/Systemhc/Data/version3
python ~/prog/searchcake2/samples/pepidentWFapp.py
