pubs_plotting_code

script.py

import pickle
import seaborn as sns
sns.set_style("whitegrid")
%pylab --no-import-all
%matplotlib inline
flatui = ["#9b59b6", "#3498db", "#95a5a6", "#e74c3c", "#34495e", "#2ecc71"]

data = pickle.load(open("allele_dic.pkl", "rb"))
translate = pickle.load(open("translate.pkl", "rb"))
aminotonumber = pickle.load(open("aminotonumber.pkl", "rb"))

aa_to_barcodes = {}
pos_to_barcodes = {}
for barcode, value in data.items():
    pos, codon = value[0].split("_")
    aa = translate[codon.replace("T", "U")]
    if aa not in aa_to_barcodes:
        aa_to_barcodes[aa] = []
    if pos not in pos_to_barcodes:
        pos_to_barcodes[pos] = []
    aa_to_barcodes[aa].append(barcode)
    pos_to_barcodes[pos].append(barcode)

scores = []
for key, value in aa_to_barcodes.items():
    scores.append((aminotonumber[key], len(value), key))
scores = list(zip(*scores))
sns.set_palette(sns.color_palette(flatui))
plt.bar(scores[0], scores[1])
plt.xlim(0, len(scores[0]))
sns.axlabel("Amino acids", "Barcode count")
plt.xticks(np.arange(len(scores[0]))+0.5, scores[2])
plt.title("Number of barcodes mapped to each amino acid")
plt.show()

scores = []
for pos, barcodes in pos_to_barcodes.items():
    scores.append((int(pos), len(barcodes)))
scores = list(zip(*scores))
sns.set_palette(sns.color_palette(flatui[1:]))
plt.bar(scores[0], scores[1], width=1.0, edgecolor="none")
plt.xlim(2, len(scores[0]))
sns.axlabel("Position", "Barcode count")
plt.title("Number of barcodes mapped to Ubiquitin AA position")
plt.show()

poses = np.zeros((len(aminotonumber), 77))
for barcode, value in data.items():
    pos, codon = value[0].split("_")
    aa = aminotonumber[translate[codon.replace("T", "U")]]
    poses[int(aa), int(pos) - 1] += 1
    
plt.figure(figsize=(10, 4))
sns.heatmap(poses,
            xticklabels=5, yticklabels=aas)
sns.axlabel("Ubiquitin AA position", "Amino Acid")
plt.show()