Kabaria_Promoter_Editing_DIAL

Data analysis, figure, and modeling code for Kabaria et al "Programmable Promoter Editing for Precise Control of Transgene Expression"

Reference data

Code in this repository requires data available as listed in the main manuscript.

Python setup for data analysis and figures

1. Create a virtual environment in the repository directory using python -m venv env
2. Activate the virtual environment using source env/bin/activate (MacOS, Linux) or .\env\Scripts\activate (Windows)
3. Install the current package versions for this project using pip install -r requirements.txt (Note if package mismatches - for Figure 6 ZF DIAL use requirements_bal.txt, for Figure 4/S19/S21 use requirements_mee.txt, and check the document requirementsold.txt for alternative package versions.) Alternatively, manually install packages as needed using the versions in the requirements.txt document. Most commonly used packages are in functions.py

These notebooks were coded with Python 3.9.1.

5. Download the raw and analyzed data from Zenodo Specifically, download the specified zip files and extract them into a new folder attune_data.
6. Create a file in the root directory of the repo called datadir.txt that contains the absolute path to the data directory attune_data you just downloaded. This should be a single line.

Importing data into ipynb

1. Select your datadir.txt at the beginning of the notebook. For example the datadir_srk.txt is selected in datadir = Path(Path('./datadir_srk.txt').read_text())
2. Ensure each of the file_path and yaml_path in each occurrence is modified to be a local reference to your data directory for the downloaded attune_data folder. For example, if datadir contained the absolute path for the attune_data folder, the code for importing data would change from:

# Original
# Import Data1
# Import Data -BioRep1
folder = '20240121flow_exp20240118_p2'
file_path = datadir/'instruments'/'data'/'attune'/'Sneha'/folder/'export_singlets' #Assign file paths
yaml_path = datadir/'instruments'/'data'/'attune'/'Sneha'/folder/'well_metadata.yaml' #Assign yaml paths 
data1 = rd.flow.load_csv_with_metadata(file_path, yaml_path) #Pull data

and become:

# Your File, with directory datadir for the Data folder
# Import Data1
# Import Data -BioRep1
folder = '20240121flow_exp20240118_p2'
file_path = datadir/'Sneha'/folder/'export_singlets' #Assign file paths
yaml_path = datadir/'Sneha'/folder/'well_metadata.yaml' #Assign yaml paths 
data1 = rd.flow.load_csv_with_metadata(file_path, yaml_path) #Pull data

3. The input columns in the function rd.flow.load_csv_with_metadata may be removed to return all single cell data parameters. The line #rd.plot.plot_well_metadata(yaml_path) can be removed as well (this line shows the plate layout.)

Exporting data

1. Each file concatenates data across bioreplicates into dataframes labeled data, df, etc. These dataframes can be exported with a function like to_excel() to obtained a merged file containing single cell data with labels of conditions and bioreplicates.
2. Alternatively, the .ipynb notebooks specify the folders and .yaml plate maps for all data per experiment, and can be used directly with the single cell data exports available at Zenodo.
3. Summary statistic data exports are available in Zenodo.

Exporting plots

1. In most notebooks you can set the directory of exported plots. Edit lines such as figure_folder = './figs_2024_promoter_editing_paper/f_diff-min-prom/'. Ensure that the directory specified exists. If you do not wish to save exported plots, comment out lines with the savefig() function, such as g.figure.savefig(figure_folder + plottitle + '.svg',dpi=300,bbox_inches='tight').

MATLAB for modeling

MATLAB code is included in a folder MATLAB_modeling_plasmid_titration. The code enclosed was used for modeling. Relevant experimental data is included in datafiles. This data was exported from the experiments in Fig2_S8_S9A_ZF37plasmidtitration.ipynb and FigS8_ZF43plasmidtitration.ipynb. Shorted notebooks for data export are modeling_ZF37plasmidtitration.ipynb and modeling_ZF43plasmidtitration.ipynb. The version used was MATLAB 24.1 (2024a).