milk study plots

plots.py

from __future__ import division
import numpy as np
import pandas as pd
import matplotlib.pyplot as plt
from matplotlib import rc, rcParams

np.random.seed(0)
rc('font',**{'family':'sans-serif','sans-serif':['Arial'],'size':16})

def myboxplot(boxdata,pointdata):
    def getboxbounds(df):
        df = pd.DataFrame(df)
        return [(np.min(d), np.min(d), np.median(d), np.max(d), np.max(d))
                    for _, d in df.iteritems()]
    ### plot boxes
    bp = plt.boxplot(getboxbounds(boxdata),sym='') # sym='' means no outliers
    plt.setp(bp['boxes'], color='black')
    plt.setp(bp['medians'], color='black')
    plt.setp(bp['whiskers'], visible=False)

    ### plot points
    for idx, (name, col) in enumerate(pointdata.iteritems()):
        N = len(col)
        xvals = idx + 1 + np.linspace(-0.1,0.1,N,endpoint=True)[np.random.permutation(N)]
        plt.plot(xvals,col,'ko')

def makefig(df,names):
    plt.figure(figsize=(8,6)) # figure size in inches

    milk = df[df['Type'] == 1] # "milk" rows are where Type is 1
    replacer = df[df['Type'] == 2] # "replacer" rows are where Type is 2

    myboxplot(milk[names],replacer[names])

    plt.xticks(np.arange(1,len(names)+1),names)
    plt.ylabel('Concentration (g/1,000 kcal)',labelpad=15)

    plt.grid('off')
    plt.grid(axis='y')

    ymin, ymax = plt.ylim()
    plt.ylim(ymin-0.05*(ymax-ymin),ymax)

    ax = plt.gca()
    ax.set_axisbelow(True)
    ax.yaxis.tick_left()
    ax.xaxis.tick_bottom()

if __name__ == '__main__':
    # load the data into a Pandas dataframe named df
    df = pd.read_excel('milk study data for matt.xlsx','Sheet1')

    # rename the columns to match the figures
    df.columns = [
            'Sample ID','Type','Protein','Carbohydrate','Fat','Calcium',
            'Phosphorus','Ca:P ratio','ALA','ARA','EPA','DHA',
            ]

    makefig(df,['Protein','Fat','Carbohydrate'])
    plt.yticks(10*np.arange(2,12))
    plt.savefig('figure1.pdf')

    makefig(df,['Calcium','Phosphorus','Ca:P ratio'])
    plt.savefig('figure2.pdf')

    makefig(df,['ALA','ARA','EPA','DHA'])
    plt.savefig('figure3.pdf')

    plt.show()