In [1]:

    

# import everything

import os
import random
import sys
import gzip

import matplotlib.lines
import matplotlib.colors
import matplotlib.pyplot as plt
from mpl_toolkits.mplot3d import Axes3D


    

In [2]:

    

# read the gziped files

explib = {}
with gzip.open('run_accession-experiment_lib.tsv.gz', 'r') as f:
    for l in f:
        p = l.strip().split("\t")
        if len(p) == 1:
            p.append('OTHER')
        explib[p[0]] = p[1]


    

In [3]:

    

# Read SRA.partie, Kate's file
# data is in these columns: 
#    0: SRA id (needs to have .sra removed)
#    1: title (percent unique kmer)
#    2: percent unique kmer
#    3: title (percent 16S)
#    4: percent 16S
#    5: title (percent PHAGE)
#    6: percent PHAGE
#    7: title (percent PROKARYOTE)
#    8: percent PROKARYOTE

data = {}
experimentlibraries = {}
with gzip.open('SRA.partie.tsv.gz', 'r') as f:
    for l in f:
        p = l.strip().split("\t")
        # data is unique kmers, percent 16S, percent phage, percent prok, percent prok + phage
        p[0] = p[0].replace('.sra', '')
        data[p[0]] = [float(p[2]), float(p[4]), float(p[6]), float(p[8]), float(p[6]) + float(p[8])]
        if p[0] in explib:
            if explib[p[0]] in experimentlibraries:
                experimentlibraries[explib[p[0]]].append(p[0])
            else:
                experimentlibraries[explib[p[0]]] = [p[0]]
        else:
            sys.stderr.write("No " + p[0] + " in exp\n")


    

In [4]:

    

# initiate the figure
allcolors = ['indigo', 'gold', 'hotpink', 'firebrick', 'indianred', 'yellow',
             'mistyrose', 'olive', 'pink', 'tomato', 'orangered', 'navajowhite', 'lime', 'palegreen', 'greenyellow',
             'burlywood', 'seashell', 'mediumspringgreen', 'fuchsia', 'papayawhip', 'blanchedalmond', 'chartreuse',
             'dimgray', 'black', 'peachpuff', 'springgreen', 'aquamarine', 'white', 'orange', 'brown', 'ivory',
             'dodgerblue', 'peru', 'lawngreen', 'chocolate', 'crimson', 'forestgreen', 'slateblue', 'cyan',
             'mintcream', 'silver', 'antiquewhite', 'mediumorchid', 'skyblue', 'gray', 'goldenrod', 'floralwhite',
             'moccasin', 'saddlebrown', 'grey', 'mediumvioletred', 'slategrey', 'red', 'deeppink', 'limegreen',
             'palegoldenrod', 'plum', 'turquoise', 'lavender', 'maroon', 'yellowgreen', 'sandybrown', 'thistle',
             'violet', 'navy', 'magenta', 'dimgrey', 'tan', 'rosybrown', 'olivedrab', 'blue', 'ghostwhite',
             'honeydew', 'cornflowerblue', 'linen', 'powderblue', 'seagreen', 'snow', 'sienna', 'mediumblue',
             'royalblue', 'green', 'mediumpurple', 'midnightblue', 'cornsilk', 'paleturquoise', 'bisque',
             'slategray', 'khaki', 'wheat', 'teal', 'deepskyblue', 'salmon', 'steelblue', 'palevioletred',
             'aliceblue', 'orchid', 'gainsboro', 'mediumseagreen', 'mediumturquoise', 'lemonchiffon', 'cadetblue',
             'lavenderblush', 'coral', 'purple', 'aqua', 'whitesmoke', 'mediumslateblue', 'mediumaquamarine',
             'beige', 'blueviolet', 'azure', 'oldlace']

fig = plt.figure()
ax = fig.add_subplot(111)
labels = {}


    

In [5]:

    

# plot the scatter plot
# data is unique kmers, percent 16S, percent phage, percent prok, percent prok + phage

for e in ['WGS', 'AMPLICON', 'CLONE', 'OTHER', 'RNA-Seq', 'WGA']:
    col = allcolors.pop(0)
    print(e + "\t" + str(len(experimentlibraries[e])))

    prok = []
    phage = []
    labels[col] = e

    for i in range(1000):
        r = random.randint(0, len(experimentlibraries[e]) - 1)
        prok.append(data[experimentlibraries[e][r]][3])
        phage.append(data[experimentlibraries[e][r]][2])

    ax.scatter(phage, prok, label=e, c=col)


    

    




WGS	44651
AMPLICON	160247
CLONE	407
OTHER	6720
RNA-Seq	2525
WGA	1265

In [6]:

    

# generate the legend
# this is to get the legend on a 3D plot
scatterproxy = []
labeltexts = []
for color in labels:
    scatterproxy.append(matplotlib.lines.Line2D([0], [0], linestyle="none", c=color, marker='o'))
    labeltexts.append(labels[color])

ax.legend(scatterproxy, labeltexts, numpoints=1)


    

    
Out[6]:





<matplotlib.legend.Legend at 0x7f1e08c5b8d0>

In [7]:

    

# show the figure
ax.set_xlabel('Percent phage')
ax.set_ylabel('Percent prokaryote')
ax.set_ylim([0,100])
ax.set_xlim([0, 100])
plt.show()


    

In [ ]: