robert.elder commited on
Commit
3a67458
·
1 Parent(s): 4e5a7f9

bug fixes for chemicals package, density lookup

Browse files
Files changed (2) hide show
  1. ChemID.py +31 -7
  2. custom_chemicals_db.tsv +3 -0
ChemID.py CHANGED
@@ -4,10 +4,15 @@ import pandas as pd
4
  import numpy as np
5
  import pubchempy as pcp
6
  import cirpy
 
7
  import chemicals
 
 
 
8
 
9
  import bs4
10
  import urllib
 
11
  import json
12
 
13
  from io import BytesIO
@@ -476,17 +481,19 @@ def is_cas(cas):
476
  except:
477
  return False
478
 
479
- def string2density(string):
480
  rho, rho_origin = None, None
481
  # try to scrape from PubChem
482
  if not rho:
483
  try:
484
  content = None
485
- compounds = pcp.get_compounds(string, namespace='name')
486
  c = compounds[0]
487
  cid = c.cid
488
- fid = urllib.request.urlopen(f'https://pubchem.ncbi.nlm.nih.gov/rest/pug_view/data/compound/{cid}/JSON')
489
- webpage = fid.read().decode('utf-8')
 
 
490
  content = json.loads(webpage)
491
  except:
492
  pass
@@ -501,12 +508,29 @@ def string2density(string):
501
  for ii in k['Information']:
502
  try:
503
  rho_string = ii['Value']['StringWithMarkup'][0]['String']
504
- tmp_rho = re.match('(?:\d+(?:\.\d*)?|\.\d+)',rho_string)
505
- tmp_rho = float(tmp_rho.group())
506
- rho_list.append(tmp_rho)
 
 
 
 
 
 
 
 
 
507
  except:
508
  continue
509
  if rho_list:
 
 
 
 
 
 
 
 
510
  rho = np.mean(rho_list)
511
  rho_origin = 'pubchem'
512
  else:
 
4
  import numpy as np
5
  import pubchempy as pcp
6
  import cirpy
7
+
8
  import chemicals
9
+ ## add custom chemical definitions (i.e., to correct confusion between methane and carbon)
10
+ db = chemicals.identifiers.get_pubchem_db()
11
+ db.load('custom_chemicals_db.tsv')
12
 
13
  import bs4
14
  import urllib
15
+ import requests
16
  import json
17
 
18
  from io import BytesIO
 
481
  except:
482
  return False
483
 
484
+ def string2density(name):
485
  rho, rho_origin = None, None
486
  # try to scrape from PubChem
487
  if not rho:
488
  try:
489
  content = None
490
+ compounds = pcp.get_compounds(name, namespace='name')
491
  c = compounds[0]
492
  cid = c.cid
493
+ url = f'https://pubchem.ncbi.nlm.nih.gov/rest/pug_view/data/compound/{cid}/JSON'
494
+ #fid = urllib.request.urlopen(url)
495
+ #webpage = fid.read().decode('utf-8')
496
+ webpage = requests.get(url).text
497
  content = json.loads(webpage)
498
  except:
499
  pass
 
508
  for ii in k['Information']:
509
  try:
510
  rho_string = ii['Value']['StringWithMarkup'][0]['String']
511
+ rho_string = rho_string.replace('Relative density (water = 1): ', '')
512
+ #print(rho_string)
513
+ #tmp_rho = re.match('(?:\d+(?:\.\d*)?|\.\d+)',rho_string)
514
+ m = re.match('((?:\d+(?:\.\d*)?|\.\d+))(?:-((?:\d+(?:\.\d*)?|\.\d+)))?',rho_string)
515
+ groups = m.groups()
516
+ if len(groups):
517
+ for g in groups:
518
+ try:
519
+ tmp_rho = float(g)
520
+ rho_list.append(tmp_rho)
521
+ except:
522
+ continue
523
  except:
524
  continue
525
  if rho_list:
526
+ ## remove outliers using interquartile range (IQR)
527
+ rho_list = np.array(rho_list)
528
+ q75,q25 = np.percentile(rho_list,[75,25])
529
+ intr_qr = q75-q25
530
+ hi = q75+(1.5*intr_qr)
531
+ lo = q25-(1.5*intr_qr)
532
+ mask = (rho_list < hi) & (rho_list > lo)
533
+ rho_list = rho_list[mask]
534
  rho = np.mean(rho_list)
535
  rho_origin = 'pubchem'
536
  else:
custom_chemicals_db.tsv ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:c46fe917b6bf96481b6d3dfd4613b60103affab34ede2863ecf16b8425385316
3
+ size 7805