With some molecules the JPlogPDescriptor returns NaN and prints a message in the form of:
"115201 not found"
I assume that this number refers to the atomtype as described in https://doi.org/10.1186/s13321-018-0316-5
I also understand that there are 13 atomtypes missing in the CDK implementation as mentioned in the conclusion of the above article. But in total I found 17 atomtypes to be missing. I will list them below with the corresponding SMILES for which the atomtype is missing
- 115210 P1N(N=C(C=1)C)C(OCC)=O
- 115202 P(OCCCCCCC)=O
- 115201 P(OC(=O)CC)=C=C
- 115100 PC(N(C1=CC=CC=C1)C2=CC=CC=C2)C
- 153202 I(N(CN=IO)C)=NCO
- 153201 I(=O)C1=CC=C(C=C1)C(F)(F)F
- 153403 I1(OC(=O)C2=C1C(=CC(OC)=C2)C)(O)=O
- 109102 FC1OC(=O)CC1C#CF
- 116503 S(O)(O)(=O)(CC(F)(F)F)C1C(F)=CC=CC=1
- 116500 S(F)(F)(F)(=C(C)C)C
- 116504 S(S(O)(=O)=O)(O)(O)(=O)CC
- 116502 S(OCCCCC)(O)(CCC)(C=C)=C
- 116501 S(F)(F)(F)(NCCN1C(=O)C=CC1=O)=C(OC)O
- 116604 S(O)(O)(O)(N)(CCO)CCO
- 116400 S(F)(F)(F)C
- 105200 N2C1N=CC=CC=1B=C2C
- 105100 S(C1=CC=C(C=C1)B)CC
Is there any way to prevent this, so that JPlogPDescriptor doesn't return NaN, but instead uses a different atomtype to approximate the missing one?
With some molecules the JPlogPDescriptor returns NaN and prints a message in the form of:
"115201 not found"
I assume that this number refers to the atomtype as described in https://doi.org/10.1186/s13321-018-0316-5
I also understand that there are 13 atomtypes missing in the CDK implementation as mentioned in the conclusion of the above article. But in total I found 17 atomtypes to be missing. I will list them below with the corresponding SMILES for which the atomtype is missing
Is there any way to prevent this, so that JPlogPDescriptor doesn't return NaN, but instead uses a different atomtype to approximate the missing one?