Table 10 Predicted SMILES with lower Tanimoto similarity indices compared with the original SMILES
From: DECIMER 1.0: deep learning for chemical image recognition using transformers
| No. | Original SMILES | Predicted SMILES | Tanimoto similarity index |
|---|---|---|---|
| 1 | P#CP=PP=PP=PP=PP=PP=PP=PP=PP=PP=PP=P | N#CC=NSSSSSSSSSSSSSSSSC=N | 0 |
| 2 | N1=NOO1 | C=1=NOC1 | 0.14 |
| 3 | OC1OC(C=2C(F)=C(F)C(F)=C(F)C21)C)C | O=C(OCOCCC(F)=C(F)C(F)=C(F)C)NC | 0.35 |
| 4 | OCC(C)(CO)C12CCC(C1)C3SSSC32 | OCC(C)(CO)C1C=2SSSC2CCC1C | 0.59 |
| 5 | O=C1N=CC2=CC(=O)C=CC2=N1 | O=C1N=CC2=NC(=O)C=CC2=N1 | 0.81 |