This code accompanies our paper available on BioRxiv, analyzing subjective reports of psychedelic experiences and mapping them to the discovered patterns to the brain. See for example, the strongest canonical component:
Create the python environment:
conda env create -f brain_environment.yml
Activate it:
conda activate brain
Then to run the CCA analysis:
python cca_on_erowid.py
--drug_folder final_drugs_sansdxm
--limit 300
--pca_components 800
--cca_components 8
--id example_run
--seed 19376
Put the components in brain regions using receptor expressions:
python cca_in_brain.py
--receptor_cca_tsv tsvs/tester_max_300_pca_800_None_cca_8_on_final_drugs_sansdxm.tsv
Plot each component as a NIFTI file.
python dump_weights_in_brain.py
--cca_components 8
--cca_by_brain_region_csv tsvs/cca_tester_max_300_pca_800_None_cca_8_on_final_drugs_sansdxm_loadings_by_brain_region_schaeffer.csv