How could I run the FCS algorithm result?

peizhenbai commented 3 years ago

Hi @kexinhuang12345 , I have several questions about your FCS algorithm in MolTrans paper:

What's the dataset that you use to run FCS and get the frequent sub-structure?
How could I run the FCS algorithm? It seems not be in this repo.
About the drug SMILE sequence, did you use the canonical SMILE or just the SMILE format provided by the dataset itself?

This is a great job and inspire me a a lot. Many thanks!

kexinhuang12345 commented 3 years ago

hey, thanks. So FCS is a relatively general algorithm and is discussed in detail in our workshop paper (https://static1.squarespace.com/static/58f7aae1e6f2e1a0f9a56616/t/5e370e2d12092f15876d5753/1580666413389/paper.pdf). It can be applied in any datasets; although we found applying on a larger dataset help identify higher quality substructures. I remembered for MolTrans, we use all drugs in ChEMBL, which can be quickly downloaded here: https://tdcommons.ai/generation_tasks/molgen/#chembl. The FCS algorithm can be found here: https://github.com/kexinhuang12345/ESPF. We use canonical SMILES.

peizhenbai commented 3 years ago

Got it, thanks!

kexinhuang12345 / MolTrans