Meeting 14 - GeorgeIniatis/Blood_Brain_Barrier_Drug_Prediction GitHub Wiki

Meeting Minutes

This meeting mainly focused on discussing error analysis and the PCA plots showcased

Comments made by supervisor:

• There does seem to be some grouping suggesting that there is some pattern that could be looked at in more detail but it would require some in depth chemical knowledge which we do not have
• Systematic weakness to the classifier that could be negated by further exploring the errors Exploring methods for explainability
• Could help chemist making use of this tool to better understand the model's results

Dissertation comments:

• State your hypothesis early on and the all things you want to test
• Make it easy for the marker to follow

Action Plan

• Explore LIME library
• Start working on the dissertation
• Convert excel files to csv (Optional/End of the project)
• Allow user to enter a PubChem ID instead of entering the chemical properties manually (Optional/End of the project)