Although human beings seem very different from the outside, when it comes to our biological info, aka DNA - we are very similar to each other, and mostly we express the same genes respectivally to each cell type. With that being said, external factor like our life-style, the food we eat and such, can have an impact of the expression levels of some genes.

There is a cool noninvasive method called cfRNA, with which we can research the RNA content in the blood. Meaning that after analysing the results of the RNA seq, we might be able to find interesting differences between blood samples of different patients.

The processed in simple terms goes like this:

1. Cells die due to Apoptosis / necrosis and spill their content to the blood.
2. Another option - cells secrete RNA via extra-cellular vesicle.
3. Either way - there are RNA molecules circulating freely in the blood stream.
4. Blood sample can be harvest, processed, and RNA material is sent to sequencing.

More information can be found here

This project is a part of a bigger project that looks for differences in the gene expression of human blood samples, that are processed using the cfRNA method.

Spesifically here, I want to create an analysis based on the GTEx database - that contains information of RNA seq data. The RNA was harvest from biopsies of most tissues in the body.

Here I'm interested in finding the marker genes of the fat tissues - both Adipose Subcutaneous and Adipose Visceral. The markers I find will be used for analysing the RNA seq data that would be received from our patients blood samples that went through the cfRNA processing.

This way, I might be able to find interesting phenomenas in regards to the fat tissues - perhaps the fat marker genes are represented more in people with obisity, or represented less in those who suffer from anorexia.

1. Downloading the GTEx database.
2. Importing relevant python libraries and modules.
3. Uploading the GTEx database and reading it as pandas table.
4. Filtering the genes - deleting mitochondrial genes and ribo genes.
5. Adjusting the table - setting gene names as indexes, dropping the ENSG names.
6. Creating 3 sub tables, each contains 2 out of 3 fat tissues - visceral, subcutaneous and brest.
7. Defining a function for finding marker genes.
8. Applying the function for the 3 sub tables, to receive the marker genes.
9. Combining the marker genes from visceral and subcutaneous adipose tissues.
10. Finding the intersection between the union list, and the marker genes in breast tissue.
11. Deleting from both the sub and vis lists the marker genes that appear in the breast tissue.
12. Extracting the marker genes for vis and sub to a csv file.

Ensure you have the following installed:

• Python 3.10 or later
• pandas
• pytest
• nbimporter

The needed installations can be found in requirements.txt

pip install -r requirements.txt

1. The code:
   Clone this repository (https://github.com/Avitalrosner/Fat_tissue_analysis.git). Or press on download ZIP.

Open GTEx.py file and run the code:

python GTEx.py

2. The input (already in the code) is: GTEx_Analysis_2017-06-05_v8_RNASeQCv1.1.9_gene_median_tpm.gct.

3. To receive the output files locally, copy the path to your directory, and paste it in the GTEx.py as shown here:

NOTE - make sure the "/" in the path are correct, should be like in the picture above.

4. The output: 2 marker genes files - vis.csv & sub.csv (provided in the repository as well).

Run-

pytest

The output should look like this:

The project was originally implemented as part of the Python programming course at the Weizmann Institute of Science taught by Gabor Szabo.