Curation and ISA representation of a SARS-Cov2/Covid-19 Proteomics Dataset - PXD107710 - ISA representation

Curation and ISA representation of a SARS-Cov2/Covid-19 Proteomics Dataset deposited in PRIDE database with accession number: PXD107710

ISA-Tab annotation for the  "SARS-CoV-2 infected host cell proteomics reveal potential therapy targets" publication. 

Github repository: https://github.com/ISA-tools/PXD017710

This is part of an effort to (re-)annotate: https://dx.doi.org/10.21203/rs.3.rs-17218/v1

Additional work done as part of:

 https://github.com/virtual-biohackathons/covid-19-bh20

 https://github.com/virtual-biohackathons/covid-19-bh20/wiki/FairData

Proteomics data

Available from PRIDE at https://www.ebi.ac.uk/pride/archive/projects/PXD017710
and [MassIVE/CCMS Maestro+MSstats reanalysis of MSV000085096 / PXD017710]

ISA-Tab representation:

Rationale: Demonstrate suitability of the ISA format for representing MS based protein profiling experiment with more granularity and details, thus providing a better representation of the experiment design.
The formatting and re-annotation are based on information extracted from:
- the original publication
- the supplementary tables available from the publishers site
- the 'filtered-results.csv' helper file as supplied to @sneumann during the HUPO-PSI meeting March 2020

Viewing the ISA-tab formatted and re-annotated PXD017710 with ISATab-Viewer

Viewing the ISA-tab formatted and re-annotated PXD017710 locally, do the following:

```bash
python -m http.server 8000
```

Then point your browser to `http://0.0.0.0:8000/isaviewer-demo.html`

Curation tasks performed:

* initial structure of the study design in ISA format:

* linkage of Proteome and Translatome data (supplementary material) to ISA assay tables (via Derived Data File)

* processing the Proteome and Translatome data (supplementary material) with python pandas library to generate the following csv files:

    - proteome_intensities_long_table_ggplot2.txt
    - proteome_diffanal_ratio_pvalue_long_table_ggplot2.txt
    - translatome_intensities_long_table_ggplot2.txt    
    - translatome_diffanal_ratio_pvalue_long_table_ggplot2

    The files are `long table` corresponding to a `melt` on the Excel file originally generated by the users and can be readily loaded in R ggplot2 library for graphical representation.
    The statistical relevant elements have been annotated with the STATO ontology and the tables comply with a Frictionless.io Data Package.
    The jupyter notebook for the transformation is available.

* conversion of raw data to mzML format: detailed in https://github.com/ISA-tools/PXD017710

install docker: 
```bash
        >brew update
        >brew install docker
```

sign in to docker
```bash
        >docker start
        >docker login
```

pull docker container for ProteoWizard:
```bash
>docker pull chambm/pwiz-i-agree-to-the-vendor-licenses
```

:warning: be sure to sign-up and login to https://hub.docker.com/

in order to be able to reach

https://hub.docker.com/r/chambm/pwiz-skyline-i-agree-to-the-vendor-licenses

run the pwiz tool from the container over the raw data:
```bash
 docker run -it --rm -e WINEDEBUG=-all -v /Users/Downloads/PXD017710/raw/:/data chambm/pwiz-skyline-i-agree-to-the-vendor-licenses wine msconvert /data/*.raw --mzML
```

* ontology markup for:
    * declaration of independent variables as ISA Study Factors:{biological agent, dose, time point, replicate} ->OBI
    * Taxonomic information (host cells and virus) -> NCBITaxonomy
    * Cell line: CaCo-2 cells -> Cell Line Ontology
    * Disease: Colon Cancer -> Human Phenotype Ontology
    * MS specific aspect (TMT reagent, instrument ... ) -> PSI-MS
    * Statistical Tests -> STATO

Unresolved curatorial issues:

 1. ambiguities related to Tandem Mass Tag labelling protocol
    - the publication mentions TMT11 (see Figure 2 in https://www.researchsquare.com/article/rs-17218/v1)
    - the information available from PRIDE mentions TMT6 (https://www.ebi.ac.uk/pride/archive/projects/PXD017710)
    This may require another round of annotation on the TMT agents and fractions in the ISA a_assay representation

 2. SARS-Cov2 isolate: no clear NCBI Taxonomic anchoring and unclear origin: -> the markup is made to the parent class (as of 06.04.2020)

Release and packaging as a BDBAG:

The tgz file associated with this upload has been producing using https://github.com/fair-research/bdbag. It contains several manifest files detailing metadata and data files, providing md5 and sha256 checksums.

Github repository: https://github.com/ISA-tools/PXD017710