from numpy import average
from scipy import interpolate
from scipy.stats import pearsonr, spearmanr, sem, ks_2samp,wilcoxon,mannwhitneyu
import pandas as pd
from numpy import *
from matplotlib.pyplot import *

protein_data = pd.read_csv('./aryal2011_proteomics_data_relative_isotope_abundance_timecourse.csv')
timepoints = protein_data.columns[4:17]
nC = len(protein_data.columns)
nT = len(timepoints)


protein_data['D1-1-L0-0'] = protein_data['D1-1']
protein_data['L1-1-D1-1'] = (protein_data['L1-1']-protein_data['D1-1'])
protein_data['D2-1-L1-1'] = (protein_data['D2-1']-protein_data['L1-1'])
protein_data['L2-1-D2-1'] = (protein_data['L2-1']-protein_data['D2-1'])
protein_data['D3-1-L2-1'] = (protein_data['D3-1']-protein_data['L2-1'])

diff_labels = protein_data.columns[-5:]

##calculate light and dark changes, using 37-25 (mostly L) vs 25-13 (mostly D)
light_mean = mean(protein_data['D2-1-L1-1'].dropna())
dark_mean = mean(protein_data['L1-1-D1-1'].dropna())

out_data = protein_data[['ORF**','D2-1-L1-1','L1-1-D1-1']]
out_data.columns = ['ORF**','Incorporation in the light','Incorporation in the dark']

out_data.to_csv('Cyanothece_dark_vs_light_protein_synthesis.csv',index=False)