Closed LukasChoi closed 5 years ago
Hello @LukasChoi, thank you for your interest in our work. You can find the datasets under data folder. For each dataset: ligands.txt -> it contains the sequences of drugs (SMILES) proteins.txt -> it contains the sequences of proteins in the data affinity.txt -> contains the binding affinity matrix in txt form (drugs x proteins in the same order given in the above text files Y -> contains the binding affinity matrix in pickle form (drugs x proteins in the same order given in the above text files)
There are a few questions:
In the /data/davis folder: 1) there are two similarities, drug-drug and target-target, what's the criteria of similarity? 2) what's the meaning of numbers in the target-target_similarities_WS.txt? I don't think the number means probability. 3) what's the standard of affinity? I don't understand the meaning of numbers in drug-atrget_interaction_affinities_Kd__Davis_et_al.2011v1.txt 4) there are six test_fold_setting files under "folds" directory, what's the role of these ones? Can you explain the meaning of numbers in the files? For example, [34121, 51548, 12611,...]
In the /data/kiba folder: 5) there two versions of affinity files, kiba_binding_affinity and kiba_binding_affinity_v2, what's the difference? 6) and what's the meaning of 'nan' in the affinity files? 7) there are ligands and proteins files with kiba prefix and without one, kiba_proteins and proteins, what's the difference? 8) what's the role of sim files (kiba_drug_sim and kiba_target_sim)?
In understanding the datasets, your explanation will be very helpful to me.
Thanks in advance.
Hi @LukasChoi, I made a readme file here and removed some duplicate data. https://github.com/hkmztrk/DeepDTA/blob/master/data/README.md
Could you please refer to this read me and then check out the updated datasets? If there is anything unclear, let me know.
I'm implementing DeepDTA with TensorFlow. By the way, I couldn't download KIBA and Davis Datasets. Where can I download Kinese KIBA and Davis datasets?
Thanks in advance.