python/salus/dfcorr/diffout.py

   1 #!/usr/bin/env python3
   2
   3 import numpy as np
   4 import matplotlib as mpl
   5 import matplotlib.pyplot as plt
   6 from matplotlib.backends.backend_pdf import PdfPages
   7 mpl.rcdefaults()
   8 mpl.rc('ps', fonttype=42, papersize='figure')
   9 mpl.rc('pdf', fonttype=42, compression=9)   #pdf.fonttype: 3  # Output Type 3 (Type3) or Type 42 (TrueType)
  10 mpl.rc('figure', figsize=(129/25.4, 129/25.4), dpi=600) # autolayout=True
  11 mpl.rc('savefig', dpi='figure') # bbox='tight'
  12
  13 import pandas as pd
  14 import seaborn as sns
  15
  16 def qw(s):
  17     return tuple(s.split())
  18
  19 prefix = '/share/result/spatial/test_huxs/prj/cmpmethod/testicles2/c2ltest/'
  20
  21 diffout = pd.read_csv(f'{prefix}diffout.csv.zst', index_col=0)
  22 diffcorr=diffout.corr()
  23
  24 # micromamba install zstandard
  25 factors = qw('Elongating Endothelial InnateLymph Leydig Macrophage Myoid SPG STids Scytes Sertoli Unknown')
  26
  27 pdf = PdfPages(f'{prefix}diffout.pdf')
  28 for onefactor in factors:
  29     colids = []
  30     colids.append(f"{onefactor}_1")
  31     colids.append(f"{onefactor}_2")
  32     onecorr = diffcorr[colids[0]][colids[1]]
  33     print(f'{onefactor}: {onecorr}')
  34     tmpDF=pd.DataFrame()
  35     tmpDF['rank1']=diffout[colids[0]].rank()
  36     tmpDF['rank2']=diffout[colids[1]].rank()
  37     tmpDF[colids[0]]=diffout[colids[0]]
  38     tmpDF[colids[1]]=diffout[colids[1]]
  39     plt.figure()
  40     sns.jointplot(data=tmpDF, x=colids[0], y=colids[1],kind="scatter",marginal_ticks=True,s=1)
  41     plt.suptitle(f'{onefactor}, Corr: {onecorr}')
  42     pdf.savefig()
  43     plt.close()
  44 pdf.close()
  45 print(f'.\n[i]Done.', flush=True)