This repository was submitted for The BrainHack School 2020 and QLSC612 Fundamentals of Neuro Data Science (McGill University and Dr. Jean-Baptiste Poline)

Running myanalysis.ipynb requires the following dependencies:

• pandas (including pandas.plotting)
• numpy
• random
• csv
• scipy.stats
• matplotlib.pyplot

1. To install the required dependencies, use pip install -r requirements.txt.
2. To run myanalysis.ipynb and print the expected outputs (statistics and figures), open the jupyter notebook using the command jupyter notebook and run all kernels.

myanalysis.ipynb creates two continuous random variables: partY and partY2. Related t-test correlations (using the command scipy.stats.ttest_rel(data[{var1}], data[{var2}], nan_policy="omit")) are run on both random variables with each of the existing variables in the brainsize.csv file: FSIQ, VIQ, PIQ, Weight, Height, and MRI_Count.

The expected statistical test outputs are as follows:

It is clear from these two tables that each correlation with partY or partY2 is significant because all of the p-values with these two random variables are < 0.05. The negative t values mean that the sample mean was less than the hypothesized mean, while the positive t value for MRI_Count and partY or partY2 means that the sample mean was greater than the hypothesized mean. Plots of the most significant correlation for both partY and partY2 are below.

The expected figure outputs should look as follows:

1. When plotting the statistical relationship between the most significant correlation for partY, the figure should look like:

2. When plotting the statistical relationship between the most significant correlation for partY2, the figure should look like: