iDRUG - intelligent drug discovery

iDrug -The Age of Intelligent Drug Discovery

Sean Ekins

collabchem.com

• Being ahead of the curve• Internet of Things– lab is wireless, control

everything from office• Personalized drug discovery from home• Beyond Apps – drug discovery workflows • One drug – many diseases = new blockbuster• Portals – data and predictions together• Who needs robots – can we automate more• Anyone can do a this – even you!

http://goo.gl/UujRX

Ballel et al., Fueling Open-Source drug discovery: 177 small-molecule leads against tuberculosis ChemMedChem 2013.

GSK screened 2M compounds – 3 yrs beforeBayesian predictions for 14,000 cpds exposed 11 / 15 (73%) correct when paper was published Further prospective validation example

http://goo.gl/UujRX

http://goo.gl/UujRX

Bayesian models identified GSK TB hits 3 years earlier

Predicted targets of GSK TB hits months earlier using TB Mobile

GSK report hits Dec 2012

24th Jan 2013 http://goo.gl/9LKrPZ

GSK predict targets Oct 2013

The Internet of Things – your house is under control from anywhere – but what about your lab?

What if our databases did more than house data – controlling its creation

• Enable connections to vendors– Assay Depot, CROs etc

• Facilitate outsourcing, insourcing data, cpds• Control lab equipment remotely, data upload• Control lab staffing, resources, plan useage• Why own the lab when someone else can –

but you control it (wherever it is)

DYOD – design your own drugs

• Thanks to our genome screening all will have an idea of what enzymes, transporters we are deficient in

• We will know which drugs are metabolized by which enzymes and which transporters and involved

• Why not tailor drugs • What tools do we need?• How to predict enzymes, transporters?

Transporters are the next big thing

All Apped out?Prepare for apps to be around for a long time– Its what we do with them that matters

• Drug discovery Workflows– Connectivity– Shareability– Ease– Use across devices

• New drug discovery tools may go straight to apps and ignore desktop– E.g. green solvents app

• Desktop – diminishing importance– We better prepare for that

Clark et al.,Submitted

App workflows

One drug– many diseases – repurposing

• Disease A may have a market of $400M• Disease B may have a market of $600M• Alone they may not be big enough to entice a

pharma• Together it’s a $1bn dollar drug• Can we find examples 1 drug – many diseases

TB and malaria

• 2 different diseases – combined deaths ~3-4M/yr R&D budget ~$1Bn

• 1 mycobacteria• 1 parasite• Shared targets• Screening data in TB• Actives vs different diseases• What if people are unaware of compounds with

dual activity?

A computational window into data and models

Should there be more ?

Make them accessible to anyone

Optimal Human properties

Optimal Mouse properties

Optimal TB entry properties

Drug discovery is repetitive and there are 1000s of diseases

Drug discovery is high risk

Do we need robots or just smarter programs that discover the ideas we test?

• Imagine having time / resources to mine datasets

• Imagine having time / resources to keep repeating vs different diseases

• Challenge is not creating the data but finding hidden value

• Anyone can do this, if the tools are available – experience not necessary

En route to treasure?

iDrug Discovery