My name is Ed Addison.
I am a Adjunct Professor with NC State University in the Engineering Department,
and I am also CEO of Cloud Pharmaceuticals which is
a small 20-person company in Raleigh Durham,
North Carolina that does AI-based drug discovery.
My topic is "AI and Big Data for Drug Discovery".
What I am planning to do is to highlight a number of different areas where
AI is being applied industry-wide for drug discovery and drug development.
I'll say in advance that there are many,
many areas where AI can enhance the process and there is
no one magic bullet that designs a drug with just a simple black box.
It just doesn't happen that way.
But there are many, many areas where AI can improve the process,
both in terms of time and cost and accuracy.
So before I begin,
I want to say a couple of things about the process of drug discovery.
Traditionally, drug discovery goes from
the gene to the target identification, to target validation,
to lead generation, lead profiling, lead optimization,
preclinical profiling, eventually clinical trials,
and ultimately approved drug.
Now that is the sort of chemical, structural, biology,
chemistry approach to drug discovery that has been used for many years,
often with wet lab techniques.
As AI comes in to assist with the process,
the process also starts to change.
It no longer is a linear process but it has feedback loops in it.
There is talk in the industry now that
targeted therapeutics are eventually going to go away in favor of modulated biology.
So what does that mean?
That means that the futuristic drug would be using an input,
whether that's chemistry or some other modality,
to drive or change the state of
the cell from state A to state B that's going to be a healthier state.
Whereas today, what we do is we inhibit or excite a single protein,
maybe sometimes two in a cocktail,
that will change or block a pathway and result
in either blocking the disease or making the patient otherwise healthier.