How do you know this Q-MAP™ platform will work (for designing and developing novel drugs)?

 
Consider designing and building a top-level Formula 1 race car. You don’t just slap some tires and an engine on a chassis then enter it into the Grand Prix. Each component of the car is meticulously designed and extensively tested before being assembled into something that can roll. Even once assembled, engineers test the car as an integrated whole in various scenarios and under various conditions. Only after full testing and evaluation does that race car make it to the track vying to go all out to cross that finish line ahead of the others.
 
Spektron’s Q-MAP™ is very much like a race car. Different components, built separately, fit together to make a complete platform. We can engineer and test these components individually before we integrate them. Then, we can assemble and test the platform as a whole. For example, consider our system for creating a molecular design space as the steering system of the race car. Can we use it to point a chemist in the right direction given our targets and anti-targets? We will test that in various scenarios and for various therapeutic targets. Consider our machine learning models as the high-powered engine. We provide it with high-octane fuel in the form of data, curated and annotated, and tune it precisely for the race. We stress test each model ensuring that it is robust and tuned explicitly for its purpose.  Do the models provide the right inferrential and predictive power?
 
We assemble these components into the platform as we test them. For us, the real rollout test is the transition from pure in silico work to physical, pre-clinical testing. Our test track, so to speak, is a small-scale synthesis, yielding enough material for initial, low-cost testing. If the platform passes the test there, we are off to the races!
 
We also take the approach of building prototypes and iteratively developing the platform. Platform development may never be complete as there are always tweaks to make! Much like the experimental components of a race car get Designed, Developed, Tested, and Evaluated so do we use this DDT&E methodology when building our platform. We’ve already cycled through the first iteration of end-to-end and applied proper testing on each component. We have assembled them into a Version 1 release and, figuratively, rolled it the shop floor. We have curated training and test data, built some key machine learning models, shaped an initial design space for Atypical Antipsychotics with lower safety and tolerance issues, and designed some Q-MAP™ optimized leads.
 
Before we take this race car out to the track, in the form of small-scale synthesis and initial testing, we want to have a second iteration under our belt. We have been building in features to the platform to increase confidence in each component and in the product that it creates.
 
Our initial test results suggest that the platform will produce leads to pre-clinical testing and clinical trials that have a strong success profile. Another way to look at it is that our race car has the power, the handling, and the drive to cross under the checkered flag.