Code for my MSc Dissertation on Patterns of gene expression in high grade serous ovarian cancer (HGSOC)

Tromics: Transcriptomics!

$ conda create -n tromics python=3.6 numpy scikit-learn pandas matplotlib pillow jupyterlab seaborn
$ source activate tromics
$ pip install mygene qgrid nimfa goatools biopython nose

The above has been captuured in create_tromics_conda_invironment.sh so you can just:

$ cd HgsocTromics
$ ./create_tromics_conda_invironment

First setup a Python 3.6 environment withthe above packages.

Then:

$ cd ~/Documents/gitrepos
$ git clone git@github.com:ipoole/HgsocTromics.git
$ ls HgsocTromics
Cache Notebooks README.md  RSrc  Src

$ cd HgsocTromics/Src
$ nosetests
..................................................
----------------------------------------------------------------------
Ran 50 tests in 12.040s

OK

Unit tests are based on trivial ('Mini') expression datasets in ...Data/Mini_AOCA and .../Data/Mini_Canon, which are committed.

Note that on the first run of unit tests there will be warnings about factorization algorithms failing to converge (due to 'Mini' datasets). The second run of nosetests will be clean as above since the factorizations are cached.

I have attempted to commit the PyCharm recommended .idea files, so this should work out of the box (except for the python interpreter). If not, a few things to check:

• Specify the correct Python 3.6 interpreter with the above packages installed.
• Unit test run templates to use nosetests and working directory .../Src

You should be able to run init tests from within PyCharm

The substantive data is not committed to git and must be obtained seperately. There are currently three datasets (in addition to the two 'Mini' datasets): AOCS, TCGA and Canon_N200. These are added into the .../Data folder, which will then look like:

$ ls -r *
TCGA_OV_VST:
TCGA_OV_VST_Metadata.tsv  TCGA_OV_VST_Expression.tsv

Mini_Canon:
Mini_Canon_Expression.tsv

Mini_AOCS:
Mini_AOCS_Expression.tsv

Canon_N200:
Canon_N200_Expression.tsv

AOCS_Protein:
AOCS_TPM_VST.csv        AOCS_Protein_Scraped_Metadata.csv  AOCS_Protein_Expression.tsv
aocs_raw_figure_e6.txt  AOCS_Protein_Metadata.tsv

.... TODO