# Dataflow diagram

1. **Rat proteomics**
1. [Rodent stroke proteomics from paper supplement](https://pubs.acs.org/doi/suppl/10.1021/acs.jproteome.9b00220/suppl_file/pr9b00220_si_002.xlsx).

`data/rat.proteomics/ratbr-ais.xlsx`

2. Table for converting UNIPROT accession to gene names is downloaded from UNIPROT interface

`annotations/uniprot-rat-entries-to-gns.txt`

4. The quantities normalized to sham

`data/rat.proteomics/rat-proteomics-norm-by-sham.tsv` is created from 1.2., 1.1 by `notebooks/rat-AIS-proteomics.ipynb`

5. WGCNA `meig` (module eigengene) and `g2m` (gene-to-module) files

`data/rat.proteomics/wgcna-pr-rats-meigs.csv`, `data/rat.proteomics/wgcna-pr-rats-g2m.csv` - these files are generated from 1.4 by `annotations/wgcna-proteomics-rats.ipynb`

6. GOEA analysis produces xlsx reports from gene lists of every WGCNA module (`data/rat.proteomics/wgcna-pr-rats-g2m.csv`)

form 1.5 reports are produced by `notebooks/wgcna-proteomics-rats.ipynb`

7. *.. find better representation ..*

2. **Mouse proteomics**
1. Mouse stroke proteomics from paper [(Gu, 2021)](https://10.1021/acs.jproteome.1c00259) deposit.

`data/mouse.proteomics/protein-?.tsv`
`data/mouse.proteomics/proteins_combined.tsv`

2. The quantities normalized to sham

2.1 are processed into `data/mouse.proteomics/mice-Gu-28days-normalized.tsv` by the notebook `notebooks/mouse-proteomics-preprocess.ipynb`