how to calculate correlation coefficient for a scatter-plot in scipy-CodePudding

i am trying to calculate the correlation coefficient for a scatterplot with scipy, the thing is, i have a kind of complex dataset in an ndarray, and the basic syntax does not work for me...

here is my full code:

import pandas as pd
import matplotlib.pyplot as plt
import numpy as np
import matplotlib.patches as mpatches
from matplotlib.pyplot import figure

figure(figsize=(12, 10), dpi=200)


import scipy.stats
from scipy.stats import t

plt.close('all')


data = np.array([
    [22.8, 14.4],
    [19.6, 3.6],
    [0.3, 16.6],
    [8.9, 7],
    [13.7, 13.4],
    [14.7, 1.5],
    [1.9, 0.4],
    [-1.8, 0.3],
    [-3, -15.3],
    [-5.9, -6.3],
    [-13.4, -15],
    [-5.7, -34.8],
    [-6.8, -12.9],

]) 

custom_annotations = ["K464E", "K472E", "R470E", "K464A", "M155E", "K472A", "M155A", "Q539A", "M155R", "D244A", "E247A", "E247R", "D244K"]
class_colours = ["r", "r", "r", "r", "r", "r", "g", "g", "b", "b", "b", "b", "b"]

for i, point in enumerate(data): 
    plt.scatter(point[0], point[1], marker='o', label=custom_annotations[i], c=class_colours[i], edgecolors='black', linewidths=1, alpha=0.75)
    plt.annotate(custom_annotations[i], (data[i,0], data[i,1]))

plt.xlabel(r'$\Delta  V_{0.5}$  Apo wild-type mHCN2 (mV)', fontsize=10)
plt.ylabel(r'$\Delta \psi$  cAMP-bound wild-type mHCN2 (mV)', fontsize=10)
plt.title('$\Delta \psi$  cAMP-bound wild-type mHCN2 (HHU) vs Change in relative current (Jena)', fontsize=10)

plt.axvline(0, c=(.5, .5, .5), ls= '--')
plt.axhline(0, c=(.5, .5, .5), ls= '--')

scipy.stats.pearsonr(data[i,0], data[i,1])

plt.legend(ncol=3, loc=(1.04,0))
plt.show()

CodePudding user response：

pearsonr works fine on your data

scipy.stats.pearsonr(data[:,0], data[:,1]) #change i to : to get the whole col.
# this returns (r_coeff, p_value)

You were passing two floats (namely values at the row i) as the error says, however corr takes two arrays, in your case the two columns.