Basic Plotting

Getting Ready

Import packages

import numpy as np

from numpy.random import randn

import matplotlib.pyplot as plt

import pandas as pd

from pandas import Series, DataFrame,date_range

from matplotlib import cm

Read in data

dat=pd.read_csv("C:\\Users\duan\Desktop\PythonDataProcessingVisualization\metaDataMean.txt", sep='\s+')

dat

Line Plot

%matplotlib inline

plt.figure()
dat.samplemeans.plot()

plt.figure()
dat.samplemeans.plot(secondary_y=True, style='g')

For list of named colors see here

Change plotting style and add legend

plt.figure();dat.samplemeans.plot(style='k--', label='sample means'); plt.legend()

Bar Plot

plt.figure();
dat.samplemeans.plot(kind='bar')

add a horizontal line

plt.figure();
dat.samplemeans.plot(kind='bar'); plt.axhline(10,linewidth=4, color='r')

plot horizontal bars

plt.figure();
dat.samplemeans.plot(kind='barh',color="pink")

plt.figure();
dat.samplemeans.plot(kind='barh',color = [cm.rainbow(i) for i in np.linspace(0, 1, len(dat.samplemeans))])

Histogram Plot

plt.figure();
dat.samplemeans.hist()

Box Plot

plt.figure();
bp = dat.boxplot()

Define a title

plt.figure();
bp=dat.boxplot()
bp.set_title('My box plot')

Boxplot by celltype

plt.figure();
bp = dat.boxplot(by='genotype')

Better layout by excluding the automatic title

plt.figure();
bp = dat.boxplot(by='genotype')
plt.suptitle('') # that's what you're after
plt.show()

More sophisticated plotting can better reveal the trend

plt.figure();
bp = dat.boxplot(column=['samplemeans'], by=['celltype','genotype'])
title_boxplot = 'expression means'
plt.title( title_boxplot )
plt.suptitle('')
plt.show()