In [1]:

    

import os
import pandas as pd
import matplotlib.pyplot as plt
import seaborn as sns
import numpy as np


    

In [3]:

    

os.chdir('../')
print(os.getcwd())


    

    




/ifs/projects/adam/pipeline_tests/data/test1/test_chiptools.dir

In [4]:

    

def plotTagCount(TagDir):
    TagFile = 'tagCountDistribution.txt'
    File = "homer/Tag.dir/" + TagDir + "/" + TagFile 
    tagCountDistribution=pd.read_csv(File, sep ="\t", header=1)
    tagCountDistribution.columns = ['Reads per position', 'Fraction of total reads (%)']
    tagCountDistribution['Fraction of total reads (%)'] = tagCountDistribution['Fraction of total reads (%)']*100
    plot_tagCountDistribution = tagCountDistribution[0:10]
    
    plt.figure(figsize=(10,6))
    sns.set_style("whitegrid", {'axes.grid' : False})
    ax = sns.barplot(x="Reads per position", y="Fraction of total reads (%)", data=plot_tagCountDistribution)
    plt.show()


    

In [5]:

    

names = set()

design = pd.read_table('design.tsv')
samples = design.bamReads.values
for sample in samples:
    #print(sample.strip(".bam"))
    names.add(sample.strip(".bam"))

for (dirpath, dirnames, filenames) in os.walk(os.getcwd()):
    for dirname in dirnames:
        if dirname in names:
            plotTagCount(dirname)