Meeting 2 - GeorgeIniatis/Blood_Brain_Barrier_Drug_Prediction GitHub Wiki

Meeting Minutes

I gave a short status report of what I did the past week and asked the supervisor the difference between compounds and drugs

The supervisor gave a thorough explanation which can be summarised as:

• Compounds can be any combination of chemicals
• Drugs are chemicals which have been proved to have a clinical benefit

We both agreed that compounds should also be included in our dataset

The supervisor then mentioned that we could create a system, once we have our model, that accepts various drugs and outputs a confidence score of whether the drug supplied passes the BBB and how much confident the system is. An extension of the system could also highlight to the user why it reached that decision and maybe flag some descriptors of the drug that played a significant role in that decision, positive or not.

The supervisor encouraged me to keep reading papers and collecting any data I can along the way that could be useful later on. Once we have the data we should then check for duplicates and if those duplicates are contradicting each other. This could make for interesting discussion in the dissertation

A discussion then followed about drug creation/discovery and delivery and some of the vocabulary that can be expected to been seen while going through papers

Action Plan

• Continue looking at the literature and making notes
• Try some automated google searches using DrugBank and the Drug Synonyms database
• Look at SMILES format as well