Aldonza-et-al._BMC-Biology_Fig-2F

import csv
import matplotlib.pyplot as plt
from wordcloud import WordCloud

# Get cell lines for each drug pair
Gef_Pac_cls = []

input_file = open("CLs/Gef_Pac.txt", "r")
for row in input_file.readlines():
	Gef_Pac_cls.append(row.strip())

# Get corresponding gene names of cosmic IDs
cell_id = {}

with open("datasets/cosmic_id.csv", newline="") as csvfile:
	reader = csv.DictReader(csvfile)
	for row in reader:
		cell_id[row['COSMIC_ID']] = row['Name']

# Perform computation
wanted = ["PDGFRA", 'PDGFRB', 'KDR', 'KIT', 'FLT3', 'MET', 'ALK', 'CSF1R', 'RET', 'FLT1', 'FLT4', 'AXL', 'CAMK1', 'EPHB4', 'DDR1', 'NTRK1', 'JAK2', 'FGFR1', 'FGFR3', 'EGFR', 'ERBB2', 'ADCK4']

gene_list = []
dataset = [] 

with open('datasets/expression_input.csv', newline='') as csvfile:
	reader = csv.DictReader(csvfile)
	for row in reader:
		if row['\ufeffGENE_SYMBOLS'] in wanted:
			gene_list.append(row['\ufeffGENE_SYMBOLS'])
			del row['\ufeffGENE_SYMBOLS']
			dataset.append(row)

	gene_value = {}

	for index, row in enumerate(dataset):
		for data in row:
			if data[5:] in cell_id:
				cell = cell_id[data[5:]]
				if cell in Gef_Pac_cls:
					if gene_list[index] not in gene_value.keys():
						gene_value[gene_list[index]] = float(row[data])
					else:	
						gene_value[gene_list[index]] += float(row[data])

	
	write_gene_value = open("plot_values/expression/wordcloud/wordcloud_plot_values.csv", "w+")

	for key, value in gene_value.items():
		gene_value[key] = value/len(gene_value)
		write_gene_value.write(str(key)+","+str(value)+"\n")

	wordcloud = WordCloud(background_color="white")
	wordcloud.generate_from_frequencies(frequencies=gene_value)
	plt.figure()
	plt.imshow(wordcloud, interpolation="bilinear")
	plt.axis("off")
	plt.savefig("images/expression/wordcloud/wordcloud_image.png", dpi=1200)
	plt.close()