glaDIAtor

Description

glaDIAtor is a software package for analyzing mass spectrometry data acquired using data independent acquisition (DIA) mode. It enables untargeted analysis of DIA proteomics and metaproteomics data without the need for any DDA data. The glaDIAtor software runs on x86/64 platforms supported by Docker. Having at least 128 GB RAM is recommended for running the example dataset presented in this Documentation.

Installation

glaDIAtor is designed to run under container technologies such as Docker and Podman. Container technologies enable running glaDIAtor under multiple platforms (x64). These instructions are written for Linux operating system, which we recommend for running glaDIAtor.

The following folders are required:

1. A folder for input files such as raw data and sequence database files (/data here).
2. A folder for storing analysis results and intermediate files (/run-files here)

First, download glaDIAtor container from Docker Hub.

$ docker pull elolab/gladiator

Start glaDIAtor

Then, deploy glaDIAtor in server mode and expose input files and result storage location to the container:

“docker run --name gladiator --rm -d --network="host" -v /data:/data -v /run-files:/run-files gladiator pserve /opt/gladiator/UI/server.ini”

Note: Single workstation deployment is done by switching from server.ini to workstation.ini. Furthermore,--network="host" parameter is omitted. When running container as normal user (supported by Podman container technology), the listening port should be exported to high range with -p 8080:80.

On Windows 10 platform, following command can be used to start glaDIAtor after which glaDIAtor can be accessed from web browser: http://localhost . Following command makes C:\run-files and C:\data folders available to glaDIAtor.

docker run --name gladiator -d --rm -v //c/run-files:/run-files/ -v //c/data:/data -p 80:80 elolab/gladiator pserve /opt/gladiator/UI/server.ini

Access glaDIAtor

After the container is deployed, the glaDIAtor can be accessed by typing the server name to web browser. On workstation deployments, glaDIAtor can be accessed by typing http://localhost (or http://localhost:8080) to web browser address bar. The difference between server and workstation deployments is that with the workstation deployment the GUI can only be accessed from the machine running glaDIAtor.

Stop glaDIAtor

The glaDIAtor can be stopped with “docker stop gladiator” command.

Analysis workflow

1. Choose DIA spectrum files to be analyzed.
   • Open format (mzXML, mzML) are supported, as well as Thermo Fisher Scientific raw files.
2. Choose sequence database files (FASTA-files).
   • The files should contain the proteins of interest, the indexed retention time (iRT) peptides for normalization, peptides related to lysis, the digestion enzyme (typically trypsin), and if desired, possible contaminants.
3. Optional: Choose DDA files to build a spectral library.
   • By default, glaDIAtor builds a pseudospectral library directly from DIA samples.
4. Customize search parameters such as Precursor and Fragment tolerances.
5. Optional: Annotate the identified peptides.
   • Peptides can be annotated by enabling the annotation option from the settings panel. In this case, an annotation table file in TSV-format is required that contains an ID column matching to FASTA IDs and multiple annotation columns such as KEGG functional group.
6. Choose a name for the analysis and start the analysis run.

Analysis Results

Once the analysis has completed successfully, the analysis results are available on the results tab. The result files contain the identified peptides and proteins and their intensities in each sample. Also, if annotation was enabled, corresponding annotation file is created and a quick overview of the results can be seen from the barplots.

If glaDIAtor detects an interrupted analysis run, an option is provided to continue the run. Furthermore, there is an option to load settings from an existing run. This sets the analysis settings according to those of an existing run.

APPENDIX

Example dataset

A test run of glaDIAtor can be performed by downloading a DIA metaproteomic human fecal dataset from ProteomeXchange Consortium PRIDE repository (https://www.ebi.ac.uk/pride/) identifier PXD008738.

• 170825_HF_1ug_DIA_1column_1.raw
• 170825_HF_1ug_DIA_1column_2.raw
• 170825_HF_1ug_DIA_1column_3.raw
• 170825_HF_1ug_DIA_1column_4.raw
• 170825_HF_1ug_DIA_1column_5.raw
• 170825_HF_1ug_DIA_1column_6.raw

Microbiota protein database

https://db.cngb.org/microbiome/genecatalog/genecatalog_human/

Convert the raw mass spectrometry files to open format

Convert the DIA raw files to mzML format using the MSConvert program from the ProteoWizard software with the following options. The conversion needs to be done on a Windows platform using the ProteoWizard software.

• Output format: mzML
• Extension: empty
• Binary encoding precision: 64bit
• Write index: checked
• TPP compatibility: checked
• Use zlib compression: unchecked
• Package in gzip: unchecked
• Use numpress linear compression: unchecked
• Use numpress short logged float compression: unchecked
• Use numpress short positive integer compression: unchecked
• Only titleMaker filter

Note, if the data are from Thermo Fisher Scientific instrument, glaDIAtor is able to read its raw files and this step is not needed.

How to customize peptide search parameters of the spectral/pseudospectral library

The default parameters are for a nanoflow HPLC system (Easy-nLC1200, Thermo Fisher Scientific) coupled to a Q Exactive HF mass spectrometer (Thermo Fisher Scientific) equipped with a nano-electrospray ionization source. Below is the summary of the default settings:

• Precursor mass tolerance: 10 ppm
• Fragment ion tolerance: 0.02 Da
• Cleavage site: Trypsin_P
• Fixed modification: Carbamidomethyl (C)
• Variable modification: Oxidation (M)

glaDIAtor provides a small set of search parameter settings that can be adjusted. If more settings needs to be adjusted, the default settings (comet_settings_template.xml and xtandem_settings_template.xml) can be accessed at the container image location /opt/gladiator/.

Run the analysis from the command line

Get help of all command line parameters:

$ docker run --rm gladiator /opt/gladiator/gladiator.py --help

Example:

$ docker run -it --rm \
-u $(id -u):$(id -g) \
-v /files/data/:/data \
-v /files/out:/run-files \
gladiator /opt/gladiator/gladiator.py \
--project-name my-first-analysis \
--sample-data /data/fecal/DIA/mzML/170825_HF_1ug_DIA_1column_1.raw /data/fecal/DIA/mzML/170825_HF_1ug_DIA_1column_2.raw /data/fecal/DIA/mzML/170825_HF_1ug_DIA_1column_3.raw /data/fecal/DIA/mzML/170825_HF_1ug_DIA_1column_4.raw /data/fecal/DIA/mzML/170825_HF_1ug_DIA_1column_5.raw /data/fecal/DIA/mzML/170825_HF_1ug_DIA_1column_6.raw \
--databases /data/ref/IGC.pep.fasta /data/ref/irtfusion.fasta /data/ref/uniprot_human.fasta /data/ref/trypsin.fasta /data/ref/Q7M135.fasta

After the processing has completed, the "out" folder contains dia-peptide-matrix.tsv and dia-protein-matrix.tsv files. The files are TSV formatted and can be loaded to spreadsheet programs like MS Excel or to a statistical analysis program like R. The dia-peptide-matrix.tsv contains the detected peptides and their intensity values for each of the samples. The first column contains the peptide sequence and a list of possible source proteins. The rest of the columns indicate the samples and contain the peptide intensity values in each sample. Similarly, dia-protein-matrix.tsv contains the intensity values at the protein level.

To perform an optional differential expression analysis between the sample groups, the groups must be provided using an additional parameter in the command: --design-file . The design file must be defined as a tab-separated file (see example-design-file.tsv), where the column Filename refers to the filename of a sample, the column Condition is the group to which the sample belongs, the column BioReplicate refers to the biological replicate, and the column Run to the MS run.

Annotate the peptide intensity matrix from command line

$ docker run -it --rm \
-u $(id -u):$(id -g) \
-v /files/data/:/data \
-v /files/results:/run-files \
gladiator /opt/gladiator/annotate-matrix.py \
--project-name my-first-analysis \
--annotation-files /data/ref/IGC.annotation.summary.v2-with-header.tsv \
--id-column 'Gene Name'

There are two optional parameters

• --threshold (The number of different annotations after which a peptide is labeled ambiguous. Default is 2.)
• --merge-unimods (Merge modifications by summing the intensities of peptides with the same amino acid sequence. Peptide sequences become unique.)

Note, the IGC annotation file IGC.annotation.summary.v2.tsv file, available at IGC homepage, is missing the column headers and the headers need to be added to the annotation file (IGC.annotation.summary.v2-with-header.tsv) before using the file in glaDIAtor. The list of column headers are listed below:

• Gene ID
• Gene Name
• Gene Length
• Gene Completeness Status
• Cohort Origin
• Taxonomic Annotation(Phylum Level)
• Taxonomic Annotation(Genus Level)
• KEGG Annotation
• eggNOG Annotation
• Sample Occurence Frequency
• Individual Occurence Frequency
• KEGG Functional Categories
• eggNOG Functional Categories
• Cohort Assembled

Use DDA data for the library

Download the following human fecal DDA data files for library construction from ProteomeXchange Consortium PRIDE repository (https://www.ebi.ac.uk/pride/) identifier PXD008738:

• 170825_HF_1_6_pool_2ug_DDA_library_1column_1.raw
• 170825_HF_1_6_pool_2ug_DDA_library_1column_2.raw
• 170825_HF_1_6_pool_2ug_DDA_library_1column_3.raw
• 170825_HF_1_6_pool_2ug_DDA_library_1column_4.raw
• 170825_HF_1_6_pool_2ug_DDA_library_1column_5.raw
• 170825_HF_1_6_pool_2ug_DDA_library_1column_6.raw

Convert the DDA raw files to mzXML format on a Windows platform using the ProteoWizard software.

FOR %i IN (*.raw) DO \
"\Program Files\ProteoWizard\ProteoWizard 3.0.11252\qtofpeakpicker.exe" \
--resolution=2000 \
--area=1 \
--threshold=1 \
--smoothwidth=1.1 \
--in %i \
--out %~ni.mzXML

DDA-files can be specified to the gladiator.py with --library-data.

Example:

--library-data /data/fecal/DDA/mzXML-peak-picked/170825_HF_1_6_pool_2ug_DDA_library_1column_1.raw \
/data/fecal/DDA/mzXML-peak-picked/170825_HF_1_6_pool_2ug_DDA_library_1column_2.raw \
/data/fecal/DDA/mzXML-peak-picked/170825_HF_1_6_pool_2ug_DDA_library_1column_3.raw \
/data/fecal/DDA/mzXML-peak-picked/170825_HF_1_6_pool_2ug_DDA_library_1column_4.raw \
/data/fecal/DDA/mzXML-peak-picked/170825_HF_1_6_pool_2ug_DDA_library_1column_5.raw \
/data/fecal/DDA/mzXML-peak-picked/170825_HF_1_6_pool_2ug_DDA_library_1column_6.raw 

Build the glaDIAtor Docker image using the Dockerfile in this repository

$ docker build -t glaDIAtor . -f Dockerfile