-
Notifications
You must be signed in to change notification settings - Fork 4
/
analyze_reg.py
1707 lines (1451 loc) · 77.6 KB
/
analyze_reg.py
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
282
283
284
285
286
287
288
289
290
291
292
293
294
295
296
297
298
299
300
301
302
303
304
305
306
307
308
309
310
311
312
313
314
315
316
317
318
319
320
321
322
323
324
325
326
327
328
329
330
331
332
333
334
335
336
337
338
339
340
341
342
343
344
345
346
347
348
349
350
351
352
353
354
355
356
357
358
359
360
361
362
363
364
365
366
367
368
369
370
371
372
373
374
375
376
377
378
379
380
381
382
383
384
385
386
387
388
389
390
391
392
393
394
395
396
397
398
399
400
401
402
403
404
405
406
407
408
409
410
411
412
413
414
415
416
417
418
419
420
421
422
423
424
425
426
427
428
429
430
431
432
433
434
435
436
437
438
439
440
441
442
443
444
445
446
447
448
449
450
451
452
453
454
455
456
457
458
459
460
461
462
463
464
465
466
467
468
469
470
471
472
473
474
475
476
477
478
479
480
481
482
483
484
485
486
487
488
489
490
491
492
493
494
495
496
497
498
499
500
501
502
503
504
505
506
507
508
509
510
511
512
513
514
515
516
517
518
519
520
521
522
523
524
525
526
527
528
529
530
531
532
533
534
535
536
537
538
539
540
541
542
543
544
545
546
547
548
549
550
551
552
553
554
555
556
557
558
559
560
561
562
563
564
565
566
567
568
569
570
571
572
573
574
575
576
577
578
579
580
581
582
583
584
585
586
587
588
589
590
591
592
593
594
595
596
597
598
599
600
601
602
603
604
605
606
607
608
609
610
611
612
613
614
615
616
617
618
619
620
621
622
623
624
625
626
627
628
629
630
631
632
633
634
635
636
637
638
639
640
641
642
643
644
645
646
647
648
649
650
651
652
653
654
655
656
657
658
659
660
661
662
663
664
665
666
667
668
669
670
671
672
673
674
675
676
677
678
679
680
681
682
683
684
685
686
687
688
689
690
691
692
693
694
695
696
697
698
699
700
701
702
703
704
705
706
707
708
709
710
711
712
713
714
715
716
717
718
719
720
721
722
723
724
725
726
727
728
729
730
731
732
733
734
735
736
737
738
739
740
741
742
743
744
745
746
747
748
749
750
751
752
753
754
755
756
757
758
759
760
761
762
763
764
765
766
767
768
769
770
771
772
773
774
775
776
777
778
779
780
781
782
783
784
785
786
787
788
789
790
791
792
793
794
795
796
797
798
799
800
801
802
803
804
805
806
807
808
809
810
811
812
813
814
815
816
817
818
819
820
821
822
823
824
825
826
827
828
829
830
831
832
833
834
835
836
837
838
839
840
841
842
843
844
845
846
847
848
849
850
851
852
853
854
855
856
857
858
859
860
861
862
863
864
865
866
867
868
869
870
871
872
873
874
875
876
877
878
879
880
881
882
883
884
885
886
887
888
889
890
891
892
893
894
895
896
897
898
899
900
901
902
903
904
905
906
907
908
909
910
911
912
913
914
915
916
917
918
919
920
921
922
923
924
925
926
927
928
929
930
931
932
933
934
935
936
937
938
939
940
941
942
943
944
945
946
947
948
949
950
951
952
953
954
955
956
957
958
959
960
961
962
963
964
965
966
967
968
969
970
971
972
973
974
975
976
977
978
979
980
981
982
983
984
985
986
987
988
989
990
991
992
993
994
995
996
997
998
999
1000
import os
from os.path import isfile, join
# import pyodbc
import pandas as pd
# import joblib
import numpy as np
import matplotlib.pyplot as plt
import matplotlib
import seaborn as sns
# import matplotlib._color_data as mcd
# import matplotlib.patches as mpatch
################ For Volcano
from matplotlib.colors import ListedColormap
from matplotlib.ticker import Locator
from bioinfokit import analys, visuz
import matplotlib.patches as mpatch
from collections import OrderedDict
################ For Volcano
# import plotly.express as px
# import plotly
# import plotly.figure_factory as ff
# import plotly.graph_objs as go
# import plotly.io as pio
#
# from sklearn.decomposition import PCA
# from sklearn import preprocessing
# from sklearn.manifold import TSNE
# import umap
# import json
#
# import scipy.spatial as sp
# from collections import OrderedDict
# import networkx as nx
# import re
def draw_color_text_box(overlap, figsize, title, title_font_size=16, color_fontsize=12,
title_font_y=1.1, fontfamily=None):
"""
Source: https://matplotlib.org/tutorials/colors/colors.html
:param overlap:
:return:
"""
# import matplotlib._color_data as mcd
# import matplotlib.patches as mpatch
# overlap = {name for name in mcd.CSS4_COLORS
# if "xkcd:" + name in mcd.XKCD_COLORS}
# overlap = {'p1': 'g', 'p2': 'b', 'p3': 'y', 'p4': 'r'}
# fig = plt.figure(figsize=[4.8, 2])
# figsize = [2, 2]
fig = plt.figure(figsize=figsize)
fig.suptitle(title, fontsize=title_font_size, y=title_font_y, fontfamily=fontfamily)
ax = fig.add_axes([0, 0, 1, 1])
# for j, n in enumerate(sorted(overlap.keys(), reverse=True)):
for j, key in enumerate(reversed(list(overlap.keys()))):
weight = None
# cn = mcd.CSS4_COLORS[n]
cn = overlap[key]
# xkcd = mcd.XKCD_COLORS["xkcd:" + n].upper()
# xkcd = n
# if cn == xkcd:
# weight = 'bold'
r1 = mpatch.Rectangle((0, j), 1, 1, color=cn)
# r2 = mpatch.Rectangle((1, j), 1, 1, color=xkcd)
txt = ax.text(figsize[0] / 2, j + 0.5, ' ' + key, va='center', fontsize=color_fontsize,
weight=weight, fontfamily=fontfamily)
ax.add_patch(r1)
# ax.add_patch(r2)
ax.axhline(j, color='k')
pass
# ax.text(.5, j + 1.5, 'Color', ha='center', va='center')
# ax.text(1.5, j + 1.5, 'xkcd', ha='center', va='center')
ax.set_xlim(0, figsize[0])
ax.set_ylim(0, j + (figsize[1] / 2) + 0.2)
ax.axis('off')
return fig
"""
if row["p_val"] == 0:
star = '***'
elif row["p_val"] <= 0.001:
star = '**'
elif row["p_val"] <= 0.01:
star = '**'
elif row["p_val"] <= 0.05:
star = '*'
elif row["p_val"] <= 0.1:
star = '.'"""
def p_val_sig_level_symbol(p_val):
star = ""
# 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1
if p_val == 0:
star = '***'
elif p_val <= 0.001:
star = '**'
elif p_val <= 0.01:
star = '**'
elif p_val <= 0.05:
star = '*'
elif p_val <= 0.1:
star = '.'
return star # x
def desc_reg(r_df, columns_reg_desc, rename_covars_dicts):
def describe_coef(row):
star = p_val_sig_level_symbol(row["p_val"])
return "{} ({} {})".format(
round(row["coef"], 3),
round(row["p_val"], 3),
star
)
pass
r_df['code_desc'] = r_df.apply(lambda row: describe_coef(row), axis=1)
if len(r_df[~r_df["resp_var"].isnull()]) == 0:
# print("[Ignored Tab] empty r_df is given.")
return None
# resp_var_type
spl_char = '$!@'
if True:
r_df['resp_var'] = r_df['resp_var'] + \
spl_char + r_df["resp_var_type"].astype(str) + \
spl_char + r_df["N"].astype(str) + \
spl_char + r_df["NA_count"].astype(str) + \
spl_char + r_df["num_covars"].astype(str) + \
spl_char + r_df["unique_val_counts"].astype(str) + \
spl_char + r_df["value_counts"].astype(str)
q = r_df.pivot(index='resp_var', columns='covariate', values=['code_desc', "coef", "p_val", "summary"])
else:
q = r_df.pivot_table(index=['resp_var', 'resp_var_type'], columns='covariate',
values=['code_desc', "coef", "p_val", "summary"], aggfunc=lambda x: x)
q = q.rename(columns=rename_covars_dicts)
q.columns = q.columns.map('|'.join).str.strip('|')
rename_cols_dicts = {}
rename_cols_dicts.update({f'code_desc|{c}': c for c in columns_reg_desc})
rename_cols_dicts.update({f'coef|{c}': f'{c} coef' for c in columns_reg_desc})
rename_cols_dicts.update({f'p_val|{c}': f'{c} p_val' for c in columns_reg_desc})
rename_cols_dicts.update({f'summary|{c}': f'{c} summary' for c in columns_reg_desc})
q = q.rename(columns=rename_cols_dicts)
if False:
q = q.rename(columns={
'code_desc|HEI-15': 'HEI-15', 'code_desc|RW.WFDPI': 'RW.WFDPI', 'code_desc|WCDPI': 'WCDPI',
'code_desc|WFDPI': 'WFDPI',
#
'coef|HEI-15': 'HEI-15 coef', 'coef|RW.WFDPI': 'RW.WFDPI coef', 'coef|WCDPI': 'WCDPI coef',
'coef|WFDPI': 'WFDPI coef',
#
'p_val|HEI-15': 'HEI-15 p_val', 'p_val|RW.WFDPI': 'RW.WFDPI p_val', 'p_val|WCDPI': 'WCDPI p_val',
'p_val|WFDPI': 'WFDPI p_val',
#
'summary|HEI-15': 'HEI-15 summary', 'summary|RW.WFDPI': 'RW.WFDPI summary',
'summary|WFDPI': 'WFDPI summary',
'summary|WCDPI': 'WCDPI summary'
})
pass
q = q.reset_index()
def restore_static_columns(row):
values_spl = row["resp_var"].split(spl_char)
row["resp_var"] = values_spl[0]
row["resp_var_type"] = int(values_spl[1])
row["N"] = int(values_spl[2])
row['NA_count'] = int(values_spl[3])
row['num_covars'] = int(values_spl[4])
row['unique_val_counts'] = int(values_spl[5])
if values_spl[6] == 'nan':
row['value_counts'] = None
else:
row['value_counts'] = values_spl[6]
return row
q = q.apply(restore_static_columns, axis=1)
# q["resp_var_type"] = q["resp_var"].apply(lambda x: int(x.split(spl_char)[1]))
# q["N"] = q["resp_var"].apply(lambda x: x.split(spl_char)[2])
# q['NA_count'] = q['resp_var'].apply(lambda x: int(x.split(spl_char)[3]))
# q['num_covars'] = q['resp_var'].apply(lambda x: int(x.split(spl_char)[4]))
# q['unique_val_counts'] = q['resp_var'].apply(lambda x: int(x.split(spl_char)[5]))
# q['value_counts'] = q['resp_var'].apply(lambda x: x.split(spl_char)[6])
# # This must be last!
# q["resp_var"] = q["resp_var"].apply(lambda x: x.split(spl_char)[0])
if False:
print('@@@@@@@@', q.columns)
print('+++++++', ["resp_var", "resp_var_type", "N",
'NA_count', 'num_covars', 'unique_val_counts', 'value_counts'] +
[col for col_list in [(f'{m}', f'{m} coef', f'{m} p_val', f'{m} summary') for m in columns_reg_desc] for
col
in
col_list])
pass
q = q[
["resp_var", "resp_var_type", "N",
'NA_count', 'num_covars', 'unique_val_counts', 'value_counts'] +
[col for col_list in [(f'{m}', f'{m} coef', f'{m} p_val', f'{m} summary') for m in columns_reg_desc] for col in
col_list]
]
# LBXSCH - Cholesterol (mg/dL) #'HEI-15 Both', 'HEI-15 D1'
return q
# return q[
# ["resp_var", "resp_var_type", "N",
# 'NA_count', 'num_covars', 'unique_val_counts', 'value_counts'] +
# [col for col_list in [(f'{m}', f'{m} coef', f'{m} p_val') for m in columns_reg_desc] for col in col_list]
# # 'HEI-15', 'HEI-15 coef', 'HEI-15 p_val'
# # 'RW.WFDPI', 'RW.WFDPI coef', 'RW.WFDPI p_val',
# # 'WCDPI', 'WCDPI coef', 'WCDPI p_val',
# # 'WFDPI', 'WFDPI coef', 'WFDPI p_val',
# # 'HEI-15 summary', 'RW.WFDPI summary', 'WFDPI summary', 'WCDPI summary'
# ]
def count_sig(path, columns_reg_desc, rename_covars_dicts, encoding="ISO-8859-1"):
# print("XX ", path)
desc_df = desc_reg(
r_df=pd.read_csv(path, index_col=0, encoding=encoding),
columns_reg_desc=columns_reg_desc,
rename_covars_dicts=rename_covars_dicts
)
if desc_df is None:
return None, None, None
for col in columns_reg_desc:
desc_df[col + " sig"] = desc_df[col].apply(lambda r: 1 if '*' in r else 0)
pass
file_name = path.split('/')[-1].replace('reg_analysis_boxcox_', '').replace('_', ' ').title()[:-4]
stat1 = [
(file_name, "Num vars", len(desc_df)),
]
stat2 = {
"Module": file_name,
"Num Vars": len(desc_df)
}
desc_df.insert(loc=0, column='module', value=file_name)
return desc_df, stat1, stat2
def helped_is_binary(x):
if str(x) == "nan":
return None
if (len(x)) != 1:
raise Exception()
return list(x)[0]
# print("q2", len(reg_analysis_boxcox_all[reg_analysis_boxcox_all['var'] == 'LBXV1A']))
def is_categorical(row):
ret = 0
if row['HEI-15 coef'] is None or np.isnan(row['HEI-15 coef']):
ret = None
if row['is_categorical tested'] == 1:
ret = 1
if row['is_binary'] == 1 and row['is_categorical tested'] == 0:
ret = 1
# if '/mL' not in row['var_desc']:
# ret = 1
# else:
# print("Var {} [{}] is not binary!!".format(row['var'], row['var_desc']))
pass
if str(row['is_ordinal']) == 'nan' or str(row['categorical_levels']) == 'nan':
ret = None
if str(row['is_ordinal']) != 'nan':
if len(row['is_ordinal']) > 1:
print(row)
raise Exception('len[is_ordinal] > 1')
is_ordinal = list(row['is_ordinal'])
if len(is_ordinal) == 1 and is_ordinal[0] != 0:
ret = 1
if str(row['categorical_levels']) != 'nan':
if len(row['categorical_levels']) > 1:
print(row)
raise Exception('len[categorical_levels] > 1')
if len(row['categorical_levels']) > 0:
ret = 1
if str(row['category']) != 'nan':
if len(row["category"]) > 1:
print(row)
raise Exception('len[category] > 1')
if str(row['sub_category']) != 'nan':
if len(row["sub_category"]) > 1:
print(row['var'], ": has more than one sub category ", row["sub_category"])
if str(row['is_ordinal']) != 'nan':
if len(row["is_ordinal"]) > 1:
print(row)
raise Exception('len[is_ordinal] > 1')
if str(row["N Patel"]) != 'nan':
if len(row["N Patel"]) > 1:
print(row)
raise Exception('len[N Patel] > 1')
return ret
def merge_all_regs(VarDescription, survey_year, survey_year_code_to_direction_name,
path_reg_analysis,
columns_reg_desc,
rename_covars_dicts,
only_work_on_module=None):
"""
"""
path = os.path.abspath("{}/{}".format(path_reg_analysis, survey_year_code_to_direction_name[survey_year]))
if os.path.exists(path) == False:
print('Path does not exists: ', path)
return None
onlyfiles = [f for f in os.listdir(path) if isfile(join(path, f)) and f.startswith("reg_analysis_boxcox_")]
stats1 = []
stats2 = []
all_desc_df = []
for box_cox_analysis in onlyfiles:
if only_work_on_module is not None:
if only_work_on_module.lower() not in box_cox_analysis.lower():
continue
# print("Working on {}".format(box_cox_analysis))
# if "Dioxins".lower() not in box_cox_analysis.lower():
# continue
desc_df, stat1, stat2 = count_sig(path=path + '/' + box_cox_analysis,
columns_reg_desc=columns_reg_desc,
rename_covars_dicts=rename_covars_dicts
)
if desc_df is None:
# print("File {} is empty!".format(
# path + '/' + box_cox_analysis
# ))
continue
all_desc_df.append(desc_df)
# if 'LBXV1A' in list(desc_df.index):
# print(box_cox_analysis)
# q1 = desc_df.reset_index()
# print(len(q1[q1['resp_var'] == 'LBXV1A']))
# pass
stats1 += stat1
stats2.append(stat2)
# break
pass
stats1_df = pd.DataFrame(stats1, columns=["Module", 'Index', 'Count'])
stats2_df = pd.DataFrame(stats2)
# stats1_df = stats1_df[stats1_df["Index"].isin(["Num vars", "DPI Any sig", "HEI-15 sig"])]
# stats1_df = stats1_df[~stats1_df["Module"].isin(["Custom"])]
# sns.set(font_scale=1.5)
"""Merge All"""
reg_analysis_boxcox_all = pd.concat(all_desc_df).reset_index(drop=True)
reg_analysis_boxcox_all = reg_analysis_boxcox_all.rename(
columns={"resp_var": "var", "resp_var_type": "var_type"}
)
# print("q1", len(reg_analysis_boxcox_all[reg_analysis_boxcox_all['var'] == 'LBXV1A']))
"""
Add variable descriptions and save results for all modules.
"""
def organize_series(s):
res = [v for v in s]
res.sort()
return res
# this will fix inconsistant values for a variables
VarDescription_category = VarDescription[
["var", "var_desc", "category", "sub_category", "is_binary", "is_ordinal", "categorical_levels", "N", "series"]
].groupby(["var", "var_desc"]).agg({
"category": lambda x: set([v for v in x if str(v) != "nan"]),
"sub_category": lambda x: set([v for v in x if str(v) != "nan"]),
"is_binary": lambda x: set([v for v in x if str(v) != "nan"]),
"is_ordinal": lambda x: set([v for v in x if str(v) != "nan"]),
"categorical_levels": lambda x: set([v for v in x if str(v) != "nan"]),
# avariable can exist in multiple years but iassigned to different modules! like URXOP4
# "series": lambda x: organize_series(x),
"N": lambda x: set([v for v in x if str(v) != "nan"])
}).reset_index()
VarDescription_category = VarDescription_category.rename(columns={"N": "N Patel"})
def find_var_series_in_all_modules(var):
# avariable can exist in multiple years but iassigned to different modules! like URXOP4
var_series = VarDescription[VarDescription["var"] == var][["var", "series"]].groupby("var").agg({
"series": lambda x: organize_series(x)
})
return var_series["series"].values[0]
VarDescription_category["series"] = VarDescription_category["var"].apply(
lambda var: find_var_series_in_all_modules(var))
VarDescription_category["num series"] = VarDescription_category["series"].apply(lambda x: len(x))
reg_analysis_boxcox_all = pd.merge(
reg_analysis_boxcox_all,
VarDescription_category,
on="var", how="left")
reg_analysis_boxcox_all['is_binary'] = reg_analysis_boxcox_all['is_binary'].apply(helped_is_binary)
print('========', reg_analysis_boxcox_all.columns)
reg_analysis_boxcox_all['is_categorical tested'] = (
reg_analysis_boxcox_all['HEI-15 summary'].str.startswith("factor")
.apply(lambda x: None if x is None else (1.0 if x is True else 0.0))
)
# print("q2", len(reg_analysis_boxcox_all[reg_analysis_boxcox_all['var'] == 'LBXV1A']))
reg_analysis_boxcox_all['is_categorical'] = reg_analysis_boxcox_all.apply(is_categorical, axis=1)
def helper_remove_set(x):
if str(x) == 'nan':
return ''
if len(x) > 1:
return ', '.join(list(x))
elif len(x) == 1:
return list(x)[0]
else:
return ''
reg_analysis_boxcox_all["category"] = reg_analysis_boxcox_all["category"].apply(helper_remove_set)
reg_analysis_boxcox_all["sub_category"] = reg_analysis_boxcox_all["sub_category"].apply(helper_remove_set)
reg_analysis_boxcox_all["categorical_levels"] = reg_analysis_boxcox_all["categorical_levels"].apply(
helper_remove_set)
reg_analysis_boxcox_all["categorical_levels"] = reg_analysis_boxcox_all["categorical_levels"].apply(
helper_remove_set)
reg_analysis_boxcox_all["N Patel"] = reg_analysis_boxcox_all["N Patel"].apply(helper_remove_set)
# reg_analysis_boxcox_all["N"] = reg_analysis_boxcox_all["N"].apply(helper_remove_set)
if 'series' not in reg_analysis_boxcox_all.columns:
reg_analysis_boxcox_all.insert(1, 'series', survey_year)
else:
reg_analysis_boxcox_all['series'] = survey_year
return {
"all_merged": reg_analysis_boxcox_all,
"stats1_df": stats1_df,
"stats2_df": stats2_df,
"survey_year": survey_year
}
##################### VOLCANO
class MinorSymLogLocator(Locator):
"""
https://stackoverflow.com/questions/20470892/how-to-place-minor-ticks-on-symlog-scale
Dynamically find minor tick positions based on the positions of
major ticks for a symlog scaling.
"""
def __init__(self, linthresh):
"""
Ticks will be placed between the major ticks.
The placement is linear for x between -linthresh and linthresh,
otherwise its logarithmically
"""
self.linthresh = linthresh
def __call__(self):
'Return the locations of the ticks'
majorlocs = self.axis.get_majorticklocs()
# iterate through minor locs
minorlocs = []
# handle the lowest part
for i in range(1, len(majorlocs)):
majorstep = majorlocs[i] - majorlocs[i - 1]
if abs(majorlocs[i - 1] + majorstep / 2) < self.linthresh:
ndivs = 10
else:
ndivs = 9
minorstep = majorstep / ndivs
locs = np.arange(majorlocs[i - 1], majorlocs[i], minorstep)[1:]
minorlocs.extend(locs)
return self.raise_if_exceeds(np.array(minorlocs))
def tick_values(self, vmin, vmax):
raise NotImplementedError('Cannot get tick locations for a '
'%s type.' % type(self))
# # https://reneshbedre.github.io/blog/volcano.html
def volcano(df="dataframe", lfc=None, pv=None, lfc_thr=1, pv_thr=0.05, color=("green", "grey", "red"), valpha=1,
geneid=None, genenames=None, gfont=8, dim=(5, 5), r=300, ar=90, dotsize=8, markerdot="o",
sign_line=False, gstyle=1, show=False, figtype='png', axtickfontsize=9,
axtickfontname="Arial", axlabelfontsize=9, axlabelfontname="Arial", axxlabel=None,
axylabel=None, xlm=None, ylm=None, plotlegend=False, legendpos='best',
figname='volcano', legendanchor=None,
legendlabels=['significant up', 'not significant', 'significant down'],
x_log_scale=False, xlim=None, custom_color_column=None):
_x = r'$ log_{2}(Fold Change)$'
_y = r'$ -log_{10}(P_{value})$'
color = color
if custom_color_column is not None:
color = list(df[custom_color_column].unique())
# for i in range(0, 3 - len(color)):
# color.append('red')
# print(color)
# check if dataframe contains any non-numeric character
assert visuz.general.check_for_nonnumeric(df[lfc]) == 0, 'dataframe contains non-numeric values in lfc column'
assert visuz.general.check_for_nonnumeric(df[pv]) == 0, 'dataframe contains non-numeric values in pv column'
# this is important to check if color or logpv exists and drop them as if you run multiple times same command
# it may update old instance of df
df = df.drop(['color_add_axy', 'logpv_add_axy'], axis=1, errors='ignore')
# allow having less than 3 colors
if custom_color_column is None:
assert len(set(color)) == 3, 'unique color must be size ometrics_sings_not_changed_cohortsf 3'
df.loc[(df[lfc] >= lfc_thr) & (df[pv] < pv_thr), 'color_add_axy'] = color[0] # upregulated
df.loc[(df[lfc] <= -lfc_thr) & (df[pv] < pv_thr), 'color_add_axy'] = color[2] # downregulated
df['color_add_axy'].fillna(color[1], inplace=True) # intermediate
pass
else:
df['color_add_axy'] = df[custom_color_column]
pass
df['logpv_add_axy'] = -(np.log10(df[pv]))
# plot
assign_values = {col: i for i, col in enumerate(color)}
color_result_num = [assign_values[i] for i in df['color_add_axy']]
# Allow having two colors!
if False:
assert len(set(
color_result_num)) == 3, 'either significant or non-significant genes are missing; try to change lfc_thr or ' \
'pv_thr to include both significant and non-significant genes'
fig = plt.figure()
plt.subplots(figsize=dim)
if plotlegend:
s = plt.scatter(df[lfc], df['logpv_add_axy'], c=color_result_num, cmap=ListedColormap(color), alpha=valpha,
s=dotsize,
marker=markerdot)
assert len(legendlabels) == 3, 'legendlabels must be size of 3'
plt.legend(handles=s.legend_elements()[0], labels=legendlabels, loc=legendpos,
bbox_to_anchor=legendanchor)
else:
plt.scatter(df[lfc], df['logpv_add_axy'], c=color_result_num, cmap=ListedColormap(color), alpha=valpha,
s=dotsize,
marker=markerdot)
# plt.axes().set_xscale('log')
if sign_line:
plt.axhline(y=-np.log10(pv_thr), linestyle='--', color='#7d7d7d', linewidth=1)
plt.axvline(x=lfc_thr, linestyle='--', color='#7d7d7d', linewidth=1)
plt.axvline(x=-lfc_thr, linestyle='--', color='#7d7d7d', linewidth=1)
# visuz.gene_exp.geneplot(df, geneid, lfc, lfc_thr, pv_thr, genenames, gfont, pv, gstyle)
gene_exp.geneplot(df, geneid, lfc, lfc_thr, pv_thr, genenames, gfont, pv, gstyle)
if x_log_scale is True:
# https://stackoverflow.com/questions/43372499/plot-negative-values-on-a-log-scale
# Here's a function to make range split into bins for symlog scale: gist.github.com/artoby/0bcf790cfebed5805fbbb6a9853fe5d5. – artoby Jul 15 at 20:53
plt.xscale("symlog", linthreshy=1e-1)
if xlim is not None:
plt.xlim(xlim)
"""
Just to add minor ticks
https://stackoverflow.com/questions/20470892/how-to-place-minor-ticks-on-symlog-scale
"""
xaxis = plt.gca().xaxis
xaxis.set_minor_locator(MinorSymLogLocator(1e-1))
""""""
pass
if axxlabel:
_x = axxlabel
if axylabel:
_y = axylabel
# visuz.general.axis_labels(_x, _y, axlabelfontsize, axlabelfontname)
# visuz.general.axis_ticks(xlm, ylm, axtickfontsize, axtickfontname, ar)
# visuz.general.get_figure(show, r, figtype, figname)
general.axis_labels(_x, _y, axlabelfontsize, axlabelfontname)
general.axis_ticks(xlm, ylm, axtickfontsize, axtickfontname, ar)
general.get_figure(show, r, figtype, figname)
return fig
def select_best_coef_pval(metric_coef_pval, strategy):
if metric_coef_pval is None or str(metric_coef_pval) == 'nan':
return None
coef_pval_list = [cp for cpl in metric_coef_pval.values() for cp in cpl]
if strategy == 'coef largest':
coef_pval_list = sorted(coef_pval_list, key=lambda x: abs(x[0]), reverse=True)
elif strategy == 'pvalue smalles':
coef_pval_list = sorted(coef_pval_list, key=lambda x: abs(x[1]), reverse=False)
else:
raise Exception('bad strategy')
# print(coef_pval_list)
if len(coef_pval_list) > 0:
return coef_pval_list[0]
else:
return None
# import pandas as pd
# import matplotlib.pyplot as plt
from mpl_toolkits.mplot3d import Axes3D
# import numpy as np
import matplotlib.cm as cmc
# import seaborn as sns
from matplotlib_venn import venn3, venn2
from random import sample
from functools import reduce
import sys
from matplotlib.colors import ListedColormap
from adjustText import adjust_text
def involcano(table="dataset_file", lfc="logFC", pv="p_values", lfc_thr=1, pv_thr=0.05, color=("green", "red"),
valpha=1,
geneid=None, genenames=None, gfont=8):
general.depr_mes("bioinfokit.visuz.gene_exp.involcano")
def ma(table="dataset_file", lfc="logFC", ct_count="value1", st_count="value2", lfc_thr=1):
general.depr_mes("bioinfokit.visuz.gene_exp.ma")
def corr_mat(table="p_df", corm="pearson"):
general.depr_mes("bioinfokit.visuz.stat.corr_mat")
def screeplot(obj="pcascree"):
y = [x * 100 for x in obj[1]]
plt.bar(obj[0], y)
plt.xlabel('PCs', fontsize=12, fontname="sans-serif")
plt.ylabel('Proportion of variance (%)', fontsize=12, fontname="sans-serif")
plt.xticks(fontsize=7, rotation=70)
plt.savefig('screeplot.png', format='png', bbox_inches='tight', dpi=300)
plt.close()
def pcaplot(x="x", y="y", z="z", labels="d_cols", var1="var1", var2="var2", var3="var3"):
for i, varnames in enumerate(labels):
plt.scatter(x[i], y[i])
plt.text(x[i], y[i], varnames, fontsize=10)
plt.xlabel("PC1 ({}%)".format(var1), fontsize=12, fontname="sans-serif")
plt.ylabel("PC2 ({}%)".format(var2), fontsize=12, fontname="sans-serif")
plt.tight_layout()
plt.savefig('pcaplot_2d.png', format='png', bbox_inches='tight', dpi=300)
plt.close()
# for 3d plot
fig = plt.figure()
ax = fig.add_subplot(111, projection='3d')
for i, varnames in enumerate(labels):
ax.scatter(x[i], y[i], z[i])
ax.text(x[i], y[i], z[i], varnames, fontsize=10)
ax.set_xlabel("PC1 ({}%)".format(var1), fontsize=12, fontname="sans-serif")
ax.set_ylabel("PC2 ({}%)".format(var2), fontsize=12, fontname="sans-serif")
ax.set_zlabel("PC3 ({}%)".format(var3), fontsize=12, fontname="sans-serif")
plt.tight_layout()
plt.savefig('pcaplot_3d.png', format='png', bbox_inches='tight', dpi=300)
plt.close()
def hmap(table="dataset_file", cmap="seismic", scale=True, dim=(4, 6), clus=True, zscore=None, xlabel=True, ylabel=True,
tickfont=(10, 10)):
general.depr_mes("bioinfokit.visuz.gene_exp.hmap")
def venn(vennset=(1, 1, 1, 1, 1, 1, 1), venncolor=('#00909e', '#f67280', '#ff971d'), vennalpha=0.5,
vennlabel=('A', 'B', 'C')):
fig = plt.figure()
if len(vennset) == 7:
venn3(subsets=vennset, set_labels=vennlabel, set_colors=venncolor, alpha=vennalpha)
plt.savefig('venn3.png', format='png', bbox_inches='tight', dpi=300)
elif len(vennset) == 3:
venn2(subsets=vennset, set_labels=vennlabel, set_colors=venncolor, alpha=vennalpha)
plt.savefig('venn2.png', format='png', bbox_inches='tight', dpi=300)
else:
print("Error: check the set dataset")
class gene_exp:
def __init__(self):
pass
def geneplot(d, geneid, lfc, lfc_thr, pv_thr, genenames, gfont, pv, gstyle):
if genenames is not None and genenames == "deg":
for i in d[geneid].unique():
if (d.loc[d[geneid] == i, lfc].iloc[0] >= lfc_thr and d.loc[d[geneid] == i, pv].iloc[0] < pv_thr) or \
(d.loc[d[geneid] == i, lfc].iloc[0] <= -lfc_thr and d.loc[d[geneid] == i, pv].iloc[0] < pv_thr):
if gstyle == 1:
plt.text(d.loc[d[geneid] == i, lfc].iloc[0], d.loc[d[geneid] == i, 'logpv_add_axy'].iloc[0], i,
fontsize=gfont)
elif gstyle == 2:
plt.annotate(i, xy=(
d.loc[d[geneid] == i, lfc].iloc[0], d.loc[d[geneid] == i, 'logpv_add_axy'].iloc[0]),
xycoords='data', xytext=(5, -15), textcoords='offset points', size=gfont,
bbox=dict(boxstyle="round", alpha=0.1),
arrowprops=dict(arrowstyle="wedge,tail_width=0.5", alpha=0.1, relpos=(0, 0)))
else:
print("Error: invalid gstyle choice")
sys.exit(1)
elif genenames is not None and type(genenames) is tuple:
for i in d[geneid].unique():
if i in genenames:
if gstyle == 1:
plt.text(d.loc[d[geneid] == i, lfc].iloc[0], d.loc[d[geneid] == i, 'logpv_add_axy'].iloc[0], i,
fontsize=gfont)
elif gstyle == 2:
plt.annotate(i, xy=(
d.loc[d[geneid] == i, lfc].iloc[0], d.loc[d[geneid] == i, 'logpv_add_axy'].iloc[0]),
xycoords='data', xytext=(5, -15), textcoords='offset points', size=gfont,
bbox=dict(boxstyle="round", alpha=0.1),
arrowprops=dict(arrowstyle="wedge,tail_width=0.5", alpha=0.1, relpos=(0, 0)))
else:
print("Error: invalid gstyle choice")
sys.exit(1)
elif genenames is not None and type(genenames) is dict:
for i in d[geneid].unique():
if i in genenames:
if gstyle == 1:
plt.text(d.loc[d[geneid] == i, lfc].iloc[0], d.loc[d[geneid] == i, 'logpv_add_axy'].iloc[0],
genenames[i], fontsize=gfont)
elif gstyle == 2:
plt.annotate(genenames[i], xy=(
d.loc[d[geneid] == i, lfc].iloc[0], d.loc[d[geneid] == i, 'logpv_add_axy'].iloc[0]),
xycoords='data', xytext=(5 + 10, -15 + 25), textcoords='offset points', size=gfont,
bbox=dict(boxstyle="round", alpha=0.1 + 0.05),
arrowprops=dict(arrowstyle="wedge,tail_width=0.5", alpha=0.1 + 0.5, # arrow
relpos=(0, 0)))
else:
print("Error: invalid gstyle choice")
sys.exit(1)
def volcano(df="dataframe", lfc=None, pv=None, lfc_thr=1, pv_thr=0.05, color=("green", "grey", "red"), valpha=1,
geneid=None, genenames=None, gfont=8, dim=(5, 5), r=300, ar=90, dotsize=8, markerdot="o",
sign_line=False, gstyle=1, show=False, figtype='png', axtickfontsize=9,
axtickfontname="Arial", axlabelfontsize=9, axlabelfontname="Arial", axxlabel=None,
axylabel=None, xlm=None, ylm=None, plotlegend=False, legendpos='best',
figname='volcano', legendanchor=None,
legendlabels=['significant up', 'not significant', 'significant down']):
_x = r'$ log_{2}(Fold Change)$'
_y = r'$ -log_{10}(P-value)$'
color = color
# check if dataframe contains any non-numeric character
assert general.check_for_nonnumeric(df[lfc]) == 0, 'dataframe contains non-numeric values in lfc column'
assert general.check_for_nonnumeric(df[pv]) == 0, 'dataframe contains non-numeric values in pv column'
# this is important to check if color or logpv exists and drop them as if you run multiple times same command
# it may update old instance of df
df = df.drop(['color_add_axy', 'logpv_add_axy'], axis=1, errors='ignore')
assert len(set(color)) == 3, 'unique color must be size of 3'
df.loc[(df[lfc] >= lfc_thr) & (df[pv] < pv_thr), 'color_add_axy'] = color[0] # upregulated
df.loc[(df[lfc] <= -lfc_thr) & (df[pv] < pv_thr), 'color_add_axy'] = color[2] # downregulated
df['color_add_axy'].fillna(color[1], inplace=True) # intermediate
df['logpv_add_axy'] = -(np.log10(df[pv]))
# print(df[df['color']==color[0]].count(), 'zzzz')
# plot
assign_values = {col: i for i, col in enumerate(color)}
color_result_num = [assign_values[i] for i in df['color_add_axy']]
assert len(set(
color_result_num)) == 3, 'either significant or non-significant genes are missing; try to change lfc_thr or ' \
'pv_thr to include both significant and non-significant genes'
plt.subplots(figsize=dim)
if plotlegend:
s = plt.scatter(df[lfc], df['logpv_add_axy'], c=color_result_num, cmap=ListedColormap(color), alpha=valpha,
s=dotsize,
marker=markerdot)
assert len(legendlabels) == 3, 'legendlabels must be size of 3'
plt.legend(handles=s.legend_elements()[0], labels=legendlabels, loc=legendpos,
bbox_to_anchor=legendanchor)
else:
plt.scatter(df[lfc], df['logpv_add_axy'], c=color_result_num, cmap=ListedColormap(color), alpha=valpha,
s=dotsize,
marker=markerdot)
if sign_line:
plt.axhline(y=-np.log10(pv_thr), linestyle='--', color='#7d7d7d', linewidth=1)
plt.axvline(x=lfc_thr, linestyle='--', color='#7d7d7d', linewidth=1)
plt.axvline(x=-lfc_thr, linestyle='--', color='#7d7d7d', linewidth=1)
gene_exp.geneplot(df, geneid, lfc, lfc_thr, pv_thr, genenames, gfont, pv, gstyle)
if axxlabel:
_x = axxlabel
if axylabel:
_y = axylabel
general.axis_labels(_x, _y, axlabelfontsize, axlabelfontname)
general.axis_ticks(xlm, ylm, axtickfontsize, axtickfontname, ar)
general.get_figure(show, r, figtype, figname)
def involcano(df="dataframe", lfc="logFC", pv="p_values", lfc_thr=1, pv_thr=0.05, color=("green", "grey", "red"),
valpha=1, geneid=None, genenames=None, gfont=8, dim=(5, 5), r=300, ar=90, dotsize=8, markerdot="o",
sign_line=False, gstyle=1, show=False, figtype='png', axtickfontsize=9,
axtickfontname="Arial", axlabelfontsize=9, axlabelfontname="Arial", axxlabel=None,
axylabel=None, xlm=None, ylm=None, plotlegend=False, legendpos='best',
figname='involcano', legendanchor=None,
legendlabels=['significant up', 'not significant', 'significant down']):
_x = r'$ log_{2}(Fold Change)$'
_y = r'$ -log_{10}(P-value)$'
color = color
assert general.check_for_nonnumeric(df[lfc]) == 0, 'dataframe contains non-numeric values in lfc column'
assert general.check_for_nonnumeric(df[pv]) == 0, 'dataframe contains non-numeric values in pv column'
# this is important to check if color or logpv exists and drop them as if you run multiple times same command
# it may update old instance of df
df = df.drop(['color_add_axy', 'logpv_add_axy'], axis=1, errors='ignore')
assert len(set(color)) == 3, 'unique color must be size of 3'
df.loc[(df[lfc] >= lfc_thr) & (df[pv] < pv_thr), 'color_add_axy'] = color[0] # upregulated
df.loc[(df[lfc] <= -lfc_thr) & (df[pv] < pv_thr), 'color_add_axy'] = color[2] # downregulated
df['color_add_axy'].fillna(color[1], inplace=True) # intermediate
df['logpv_add_axy'] = -(np.log10(df[pv]))
# plot
assign_values = {col: i for i, col in enumerate(color)}
color_result_num = [assign_values[i] for i in df['color_add_axy']]
assert len(set(
color_result_num)) == 3, 'either significant or non-significant genes are missing; try to change lfc_thr or ' \
'pv_thr to include both significant and non-significant genes'
plt.subplots(figsize=dim)
if plotlegend:
s = plt.scatter(df[lfc], df['logpv_add_axy'], c=color_result_num, cmap=ListedColormap(color), alpha=valpha,
s=dotsize, marker=markerdot)
assert len(legendlabels) == 3, 'legendlabels must be size of 3'
plt.legend(handles=s.legend_elements()[0], labels=legendlabels, loc=legendpos,
bbox_to_anchor=legendanchor)
else:
plt.scatter(df[lfc], df['logpv_add_axy'], c=color_result_num, cmap=ListedColormap(color), alpha=valpha,
s=dotsize, marker=markerdot)
gene_exp.geneplot(df, geneid, lfc, lfc_thr, pv_thr, genenames, gfont, pv, gstyle)
plt.gca().invert_yaxis()
if axxlabel:
_x = axxlabel
if axylabel:
_y = axylabel
general.axis_labels(_x, _y, axlabelfontsize, axlabelfontname)
if xlm:
print('Error: xlm not compatible with involcano')
sys.exit(1)
if ylm:
print('Error: ylm not compatible with involcano')
sys.exit(1)
general.axis_ticks(xlm, ylm, axtickfontsize, axtickfontname, ar)
general.get_figure(show, r, figtype, figname)
def ma(df="dataframe", lfc="logFC", ct_count="value1", st_count="value2", lfc_thr=1, valpha=1, dotsize=8,
markerdot="o", dim=(6, 5), r=300, show=False, color=("green", "grey", "red"), ar=90, figtype='png',
axtickfontsize=9,
axtickfontname="Arial", axlabelfontsize=9, axlabelfontname="Arial", axxlabel=None,
axylabel=None, xlm=None, ylm=None, fclines=False, fclinescolor='#2660a4', legendpos='best',
figname='ma', legendanchor=None, legendlabels=['significant up', 'not significant', 'significant down'],
plotlegend=False):
_x, _y = 'A', 'M'
assert general.check_for_nonnumeric(df[lfc]) == 0, 'dataframe contains non-numeric values in lfc column'
assert general.check_for_nonnumeric(df[ct_count]) == 0, \
'dataframe contains non-numeric values in ct_count column'
assert general.check_for_nonnumeric(
df[st_count]) == 0, 'dataframe contains non-numeric values in ct_count column'
# this is important to check if color or A exists and drop them as if you run multiple times same command
# it may update old instance of df
df = df.drop(['color_add_axy', 'A_add_axy'], axis=1, errors='ignore')
assert len(set(color)) == 3, 'unique color must be size of 3'
df.loc[(df[lfc] >= lfc_thr), 'color_add_axy'] = color[0] # upregulated
df.loc[(df[lfc] <= -lfc_thr), 'color_add_axy'] = color[2] # downregulated
df['color_add_axy'].fillna(color[1], inplace=True) # intermediate
df['A_add_axy'] = (np.log2(df[ct_count]) + np.log2(df[st_count])) / 2
# plot
assign_values = {col: i for i, col in enumerate(color)}
color_result_num = [assign_values[i] for i in df['color_add_axy']]
assert len(
set(
color_result_num)) == 3, 'either significant or non-significant genes are missing; try to change lfc_thr' \
' to include both significant and non-significant genes'
plt.subplots(figsize=dim)
# plt.scatter(df['A'], df[lfc], c=df['color'], alpha=valpha, s=dotsize, marker=markerdot)
if plotlegend:
s = plt.scatter(df['A_add_axy'], df[lfc], c=color_result_num, cmap=ListedColormap(color),
alpha=valpha, s=dotsize, marker=markerdot)
assert len(legendlabels) == 3, 'legendlabels must be size of 3'
plt.legend(handles=s.legend_elements()[0], labels=legendlabels, loc=legendpos,
bbox_to_anchor=legendanchor)
else:
plt.scatter(df['A_add_axy'], df[lfc], c=color_result_num, cmap=ListedColormap(color),
alpha=valpha, s=dotsize, marker=markerdot)
# draw a central line at M=0
plt.axhline(y=0, color='#7d7d7d', linestyle='--')
# draw lfc threshold lines
if fclines:
plt.axhline(y=lfc_thr, color=fclinescolor, linestyle='--')
plt.axhline(y=-lfc_thr, color=fclinescolor, linestyle='--')
if axxlabel:
_x = axxlabel
if axylabel:
_y = axylabel
general.axis_labels(_x, _y, axlabelfontsize, axlabelfontname)
general.axis_ticks(xlm, ylm, axtickfontsize, axtickfontname, ar)
general.get_figure(show, r, figtype, figname)
def hmap(df="dataframe", cmap="seismic", scale=True, dim=(4, 6), rowclus=True, colclus=True, zscore=None,
xlabel=True,
ylabel=True, tickfont=(10, 10), r=300, show=False, figtype='png', figname='heatmap'):
# df = df.set_index(d.columns[0])
# plot heatmap without cluster
# more cmap: https://matplotlib.org/3.1.0/tutorials/colors/colormaps.html
# dim = dim
fig, hm = plt.subplots(figsize=dim)
if rowclus and colclus:
hm = sns.clustermap(df, cmap=cmap, cbar=scale, z_score=zscore, xticklabels=xlabel, yticklabels=ylabel,
figsize=dim)
hm.ax_heatmap.set_xticklabels(hm.ax_heatmap.get_xmajorticklabels(), fontsize=tickfont[0])
hm.ax_heatmap.set_yticklabels(hm.ax_heatmap.get_ymajorticklabels(), fontsize=tickfont[1])
general.get_figure(show, r, figtype, figname)
elif rowclus and colclus is False:
hm = sns.clustermap(df, cmap=cmap, cbar=scale, z_score=zscore, xticklabels=xlabel, yticklabels=ylabel,
figsize=dim, row_cluster=True, col_cluster=False)
hm.ax_heatmap.set_xticklabels(hm.ax_heatmap.get_xmajorticklabels(), fontsize=tickfont[0])
hm.ax_heatmap.set_yticklabels(hm.ax_heatmap.get_ymajorticklabels(), fontsize=tickfont[1])
general.get_figure(show, r, figtype, figname)
elif colclus and rowclus is False:
hm = sns.clustermap(df, cmap=cmap, cbar=scale, z_score=zscore, xticklabels=xlabel, yticklabels=ylabel,
figsize=dim, row_cluster=False, col_cluster=True)
hm.ax_heatmap.set_xticklabels(hm.ax_heatmap.get_xmajorticklabels(), fontsize=tickfont[0])
hm.ax_heatmap.set_yticklabels(hm.ax_heatmap.get_ymajorticklabels(), fontsize=tickfont[1])
general.get_figure(show, r, figtype, figname)
else:
hm = sns.heatmap(df, cmap=cmap, cbar=scale, xticklabels=xlabel, yticklabels=ylabel)
plt.xticks(fontsize=tickfont[0])
plt.yticks(fontsize=tickfont[1])
general.get_figure(show, r, figtype, figname)
class general: