finalplot_combined_LDsvSD.py

# -*- coding: utf-8 -*-
"""
Created on Mon Jul 25 16:04:54 2022

@author: fm02
"""
### Author: federica.magnabosco@mrc-cbu.cam.ac.uk
### Plot results from LDvsSD individual ROIs accuracy


import numpy as np
import pandas as pd
import pickle
import seaborn as sns
import matplotlib.pyplot as plt
from scipy.stats import sem

sns.set(rc={"figure.dpi":300, 'savefig.dpi':300})
sns.set_theme(context="notebook",
              style="white",
              font="sans-serif")

sns.set_style("ticks")

kkROI = ["lATL", "rATL", "AG", "PTC", "IFG", "PVA"]

def rms(example):
    """Compute root mean square of each ROI.
    Input is a dataframe of length=n_vertices."""
    # first transform Series in np array of dimension n_vertics*timepoints, when the input is unstacked.
    example = np.vstack(np.array(example))
    # create np.array where to store info
    rms_example = np.zeros(example.shape[1])
    # loop over timepoints
    for i in np.arange(0,example.shape[1]):
        rms_example[i] = np.sqrt(np.mean(example[:,i]**2))
    
    return rms_example 

# 
kkROI = ['lATL', 'rATL', 'AG', 'PTC', 'IFG', 'PVA']

with open("/imaging/hauk/users/fm02/final_dTtT/combined_ROIs/LDvsSD/scores.P" , 'rb') as f:
    scores = pickle.load(f)

with open("/imaging/hauk/users/fm02/final_dTtT/combined_ROIs/LDvsSD/patterns.P" , 'rb') as f:
    patterns = pickle.load(f)

# # create times array
times = np.arange(-300,900,4)

colors = sns.color_palette(['#FFBE0B',
                            '#FB5607',
                            '#FF006E',
                            '#8338EC',
                            '#3A86FF',
                            '#1D437F',
                            '#1D437F'
                            ])

# IBM colorblind palette
# '#648fff'
# '#785ef0'
# '#dc267f'
# '#fe6100'
# '#ffb000'
# '#000000'
# '#ffffff'

for task in scores.keys():
    scores[task] = np.array(scores[task])
        
# initialise average(scores) key
scores['avg'] = [ [] for _ in range(len(scores)) ]

# calcualte average performance for each participant, across tasks
for i in range(0, 18):
    scores['avg'].loc[i] = np.array([scores['mlk'][i],
                              scores['frt'][i],
                              scores['odr'][i]]).mean(axis=0)
        
# for roi in scores['avg'].keys():
#    scores['avg'][roi] = np.array(scores['avg'][roi])


i = 0
for task in scores.keys():
    # iter to select colours
    # average plot all ROIs in one plot
        # plot the average score across task (= scores['avg'], and across participants)    
    sns.lineplot(x=times, y=np.stack(scores[task]).mean(axis=0), color=colors[i], label=task)
    # plot the standard error of the mean
    plt.fill_between(x=times, \
                      y1=(np.stack(scores[task]).mean(axis=0)) - sem(np.stack(scores[task]),0), \
                      y2=(np.stack(scores[task]).mean(axis=0)) + sem(np.stack(scores[task]),0), \
                      color=colors[i], alpha=.1)
    i += 2
    # plot some line that are useful for inspection
plt.axvline(0, color='k');
plt.title('LD vs average(SD) Decoding ROC AUC')
plt.axhline(.5, color='k', linestyle='--', label='chance');
plt.legend();
#plt.savefig('//cbsu/data/Imaging/hauk/users/fm02/final_dTtT/combined_ROIs/LDvsSD/Figures/average_LDvsSD_accuracy.png', format='png')
plt.show();

patterns_roi = dict.fromkeys(kkROI)

for roi in patterns_roi.keys():
    patterns_roi[roi] = dict.fromkeys(['frt', 'mlk', 'odr'])
    for task in patterns_roi[roi].keys():
        patterns_roi[roi][task] = []

# calculate the ROOT-MEAN-SQUARE for each pattern in each task
# loop over participants    
for i in range(18):
    # loop over each roi
    for roi in patterns_roi.keys():
        # loop over each task
        for task in patterns_roi[roi].keys():
            patterns_roi[roi][task].append(rms(np.array(patterns[task][i].loc[roi])))

for roi in patterns_roi.keys():
    patterns_roi[roi]['avg'] = []

# calculate the average of the RMS(pattern) across each task

# loop over participants    
for i in range(18):
    # loop over each roi
    for roi in patterns_roi.keys():
        patterns_roi[roi]['avg'].append(np.array([patterns_roi[roi]['mlk'][i],
                                 patterns_roi[roi]['frt'][i],
                                 patterns_roi[roi]['odr'][i]]).mean(axis=0))    

i = 0
for roi in patterns_roi.keys():
    sns.lineplot(x=times, y=np.array(patterns_roi[roi]['avg']).mean(axis=0), color=colors[i]) # this takes mean over participants
    i += 1
plt.axvline(0, color='k');
plt.title('LD vs average(SD) RMS patterns')
plt.legend(patterns_roi.keys());
#plt.savefig('/imaging/hauk/users/fm02/final_dTtT/combined_ROIs/LDvsSD/Figures/average_LDvsSD_patterns.png', format='png')
plt.show();

sns.lineplot(x=times, y=np.stack(scores['avg'].mean(axis=0)), color='black')
plt.fill_between(x=times, \
                      y1=(np.mean(np.stack(scores['avg']),0)-sem(np.stack(scores['avg']),0)), \
                      y2=(np.mean(np.stack(scores['avg']),0)+sem(np.stack(scores['avg']),0)), \
                      color='k', alpha=.1)
plt.title(f'LD vs average(SD) Decoding ROC AUC')
plt.axvline(0, color='k')
plt.axhline(.5, color='k', linestyle='--');