Figure 5 - Source data 1:

This folder contains the quantification results used as source data for Figure 5E, Figure 5 - Figure supplement 1C, and Figure 5 - Figure supplement 2 C-E.

In brief, we used binary segmentation masks derived from the probabilistic softmax output of the trained models for each image to quantify the number of cFOS-positive ROIs in a selected area (e.g. PVT for Lab-Mue).

The analyses were performed with the code provided in the Jupyter Notebook "fig_5_S9_S10_quantification.ipynb" in our GitHub repository:

https://github.com/matjesg/bioimage_analysis/blob/master/notebooks/fig_5_S9_S10_quantification.ipynb

Method details are provided in the Material & Methods section of our paper.
_______________________________________

Additional file descriptions:

ROIs_*.csv:

Contains the raw data of the ROI quantifications for the respective lab (Mue, Inns1, Inns2, or Wue2): 
 - mean_intensity: 	mean cFOS signal intensity of individual ROI
 - label: 		ID of annotated cFOS-positive ROI
 - area:		area of individual ROI in pixel
 - Nummer:		ID of analyzed image
 - region area:		area of analyzed region in pixel
 - model, unet, ens:	identification of model / ensemble used for analysis
 - subarea:		analyzed region
 - init:		type of initialization


quantifications_*.csv:

Data of "ROIs_*.csv" further processed to ultimately calculate for each image ("Nummer" = image id) the normalized number of cFOS-positive ROIs per analyzed area ("norm_cfos_per_area") and the normalized mean cFOS signal intensity ("norm_mean_intensity"). This file also contains information about treatment condition ("Kondition"; [Lab-Mue - Ext: early retrieval, eRet; Ret: late retrieval, lRet], [Lab-Wue2 - WT: wild type, wt; KO: gad1b knock-down, kd], [Lab-Inns1: Ctrl: control, Ext: Extinction], [Lab-Inns2: Saline, L-DOPA responder, L-DOPA non-responder]), and the model or ensemble that was used for the respective analysis ("ens", "unet", "model", "init"). 



stat_results_quantifications_*.csv:

Results of all individual statistical analyses of "quantifications_*.csv", including calculated effect sizes (eta^2) that are plotted in the respective figure:
 - 1: group 1 (Lab-Mue: eRet, Lab-Wue2: wt, Lab-Inns1: Ctrl, Lab-Inns2: Saline)
 - 2: group 2 (Lab-Mue: lRet, Lab-Wue2: kd, Lab-Inns1: Ext, Lab-Inns2: L-DOPA responder)
 - 3: group 3 (Lab-Inns2: L-DOPA non-responder)
 - anova_ok: Indicates whether the pre-requisites for parametrical testing (normal distribution and euqality of variances) were met (True), or not (False). 
 - H_kwa_all: H-statistic of Kruskal-Wallis-ANOVA between all three groups
 - U_mwu: U-statistic of Mann-Whitney-U test between indicated groups, e.g.: mwu1_vs_2 - Mann-Whitney-U test between group 1 and group 2
 - eta^2_*: eta-squared of indicated test
 - p_*: p-value of indicated test
 - c_*: critical eta-squared of indicated test at p = 0.05
 - c0.01_*: critical eta-squared of indicated test at p = 0.01
 - c0.001_*: critical eta-squared of indicated test at p = 0.001
