Closed xuzhang5788 closed 5 years ago
Hello @xuzhang5788 , thanks a lot for your interest in our paper. I should clarify that in DeepDTA, proteins are represented with their sequences. However, in another model, we use the SMILES string of the interacting ligands of the protein, here. The repo has more detail explanation in README. If you have specific questions regarding to that study, please let me know.
DeepDTA here, uses SMILES only for ligand description.
@hkmztrk Thank you for your response. You said, "proteins are represented with their sequences". About the sequences, do you use DNA sequences or amino acid sequences? When you used sequences to represent proteins, you lost structural information. Even you use the interacting ligands' smiles to represent proteins, still, you lost the structural information. Any explanation?
Hello @xuzhang5788, here we used amino acid sequences. I agree with you about the lost structure information, however not for all proteins the structure is determined although sequence is always available.
Using interacting ligands' SMILES, we aimed to capture function information of protein, because we know that similar proteins tend to bind similar ligands.
Thank you for your post. I read your paper, but I am not very clear about how to represent proteins using ligands' smiles. Would you like to explain it again? Many thanks