dominance_strain_comparison

This repository holds all the scripts (in python and R) that were used for data analysis and figure creation for the dominance strain comparison manuscript. Full disclosure: this repository was reorganized with no double checking about dependencies in terms of calling data or other code; so be ware that that could be an issue while running things. Best MC

Directory Structure

• data/: Contains raw data (reward comp has been updated)
• mpc_scripts/: Contains med pc scripts used during the reward competition and training (written in med pc code) author: mixed
• results/: contains most of the code
  • 2023_05_12_pilot_consolidation: author RI
  • data: should be empty, i put the excel sheets i need for figures 3-5 in there
  • figure 3: contains an old boris analysis script in python, it calculated the percentage of trials across behaviors, does not include stats or figures,
    author: MIC
  • figure 4: contains 3 scripts for
    • contested_vs_uncont.ipynb: fig4. B constested vs uncontested figure creation and some data carpentry (no stats) author: AH
    • rc_diff.ipynb: fig4 C percent trials won across winner vs losers, data carpentry only author: AH
    • feature_extraction_sleap.ipynb: fig4 D-I all the sleap analysis (feature extraction, umap, gif creation, enrichment plots etc) all figure creation is in here and stats, should also exist inpose_tracking_repo; author: MC
  • figure 5: has some old elo score calculations and an old correlations script, author RI for elo, correation author MC
    • david_score_calculation.py: calculates david score, author CY
    • david_score_plotting.py: plots correlation matrices (fig 5 E + F, fig S3), author CY
    • linearity_matrix_generation.ipynb: takes in the template dat in the outer most data folder and produces an output excel sheet of matrices of wins to be used in the R markdown linearity; author: MC
    • tube_test_matrix.ipynb: create s matrix for tube tes tdata, author KP
  • old_figures: author RI
  • R_GLMs_DCI: has all the R scripts for the DCI calculations and GLMs including the project and R markdowns; author: MC
    • GLM_stats.Rproj: the project file
    • linearity.Rmd: calculates DCI based on output from the py script linearity_matrix_generation.ipynb
    • lmer_tubetest.RMD: stats for fig 3, does the mixed model on the boris data for tube test
    • rewardcomp.Rmd: does stats on the reward comp data (number of trials won per subject per strain per winer vs loser)
    • urine_marking.Rmd: runs a GLM mixed effects model on urine spots, winner/loser, subject, stats for fig 2
  • reward_training: calculates training progession, latencies and port probabilities; author RI
  • rewardcomp_sleap: calculates distance to port and other things; author RI
• src/: src code for elo scores and med pc extracting (this might throw an error as i reorganized without checking to see if the elo scores and med pc scripts would run); author: RI