daf2-dauer-comparisons

Comparison analysis of _daf-2 dauer connectome dataset with wildtype nondauer connectome datasets

Required Software

• Python 3.8.2

Usage

Install project dependencies pip3 install -r requirements.txt

Make new folders output, graphs and analysis

1. Edit generate_tables.py with your values for:

   • connection_type: cell-to-cell or neuron_pair
   • zero_filter: 10 (all data) or 'early_development' (filter to allow 1 zero in early development)
   • compare: daf2-dauer or L3 (L3 was added for proof of concept purposes)
   • pvalue_cutoff: Your desired pvalue threshold, 0.05 is default

2. Run generate_tables.py

3. Repeat steps 1-2 until your required conditions are completed.

4. Make new folders in analysis folder named based on a description of your zero_fliter from above i.e. all_connections, 1_zero_in_early_development, and no_zeros

5. Copy and paste the outputs from steps 1-3 into their appropriate folders in analysis

6. Make new folder in graphs folder, name using your value entered for compare in the previous step

7. Within the {your value for compare} folder, make new folder called nonparametric_bootstrapping

8. Within nonparametric_bootstrappingfolder, make new folder based on description of your zero_filter i.e. all_connections, 1_zero_in_early_development, and no_zeros

9. Edit connection_classification.py with your values for:

   • filter = '1_zero_in_early_development' or 'all_connections'
   • cutoff = pvalue threshold, 0.05 is default
   • find_shared_stable_pvalues = True or False (Whether you want to only look at connections with pvalues under the threshold for all 3 normalization methods)
   • compare = daf2-dauer or L3

   If you want to compare with L3:

   • Make test folder in analysis folder,

   • Make new folders in test folder named based on a description of your zero_fliter from above i.e. all_connections, 1_zero_in_early_development, and no_zeros

   • Edit job_dir to:

     job_dir = f'./output/connection_lists/test/{filter}

   • Edit load_data function to:

     df_total = make_connection_key(f'./analysis/test/{filter}/count/total_changes.csv')

     df_input = make_connection_key(f'./analysis/test/{filter}/count/input_changes.csv')

     df_output = make_connection_key(f'./analysis/test/{filter}/count/output_changes.csv')

10. Make new folder in output folder called connection_lists

11. Run connection_classification.py

12. Edit summary_plot.py with your values for:

    • compare = daf2-dauer or L3 (Make sure this is the same value as above)

13. Run summary_plot.py

Optional

• run venn_diagram.py to get a summary of the similarities between analyses of size vs. count