SMILES in PSets

[MomenehForoutan]

Hi team,
I have been checking the SMILES for several drugs from PRISM PSet manually and they do not seem to be consistent with those from PubChem. I also tried retrieving SMILES using the webchem package but the one from webchem is also different from the other two for the same drug. Checking Costello et al paper, it seems that the SMILES in the PRISM PSet is coming from their supplementary table, is that right? If so, I was just wondering if you know what would be the source of this discrepancy across SMILES from different sources, and if you have done (or will do) any curation on the SMILES? Also, how did you retrieve the SMILES for other PSets?

Below is an example using drug Terfenadine to show you my point. The SMILES for this drug on PubChem website (when looking manually) is CC(C)(C)C1=CC=C(C=C1)C(CCCN2CCC(CC2)C(C3=CC=CC=C3)(C4=CC=CC=C4)O)O, which is different from what I get from PSet and webchem (see below).

# From PSet:
pi <- drugInfo(PRISM)
pi$smiles[pi$treatmentid == "Terfenadine"]
# "CC(C)(C)c1ccc(cc1)C(O)CCCN1CCC(CC1)C(O)(c1ccccc1)c1ccccc1, CC(C)(C)c1ccc(cc1)C(O)CCCN1CCC(CC1)C(O)(c1ccccc1)c1ccccc1, CC(C)(C)c1ccc(cc1)C(O)CCCN1CCC(CC1)C(O)(c1ccccc1)c1ccccc1"

# From webchem:
webchem::cir_query("Terfenadine", "smiles")
# A tibble: 1 × 2
#  query       smiles                                                   
 # <chr>       <chr>                                                    
# Terfenadine CC(C)(C)c1ccc(cc1)C(O)CCCN2CCC(CC2)C(O)(c3ccccc3)c4ccccc4

Many thanks!
Sepideh