Platform & Expertise
At Denovicon Therapeutics, the overall drug discovery process begins and ends with our cutting edge computational platform and proprietary molecular modeling-AI algorithm. This means everything from the strategic identification of viable biological targets to the design of small-molecule clinical candidates is triaged using the computational platform. This along with our combined decades’ worth of drug discovery expertise (resulting in several clinical candidates and another currently moving through clinical trials) gives us our edge in modern-day drug discovery.
biological pathways & target selection
From the very start, our biologists and computational chemists work closely together to make sure that therapeutically-interesting biological pathways and targets are viable for small-molecule interrogation. In this way, targets that are unlikely to be druggable are avoided and those with the highest chance of success become the immediate focus.
Traditional drug discovery often starts with a high-throughput screen (HTS) of an in-house compound collection containing some millions of compounds. The overall process takes several months to complete and usually results in very low hit rates and poor quality hits. Our strategy involves screening a virtual compound collection of billions of compounds. No molecule gets synthesized or tested in an assay until it goes through the computational platform. In this way, quicker turnaround times, lower costs, and hit rates that are orders of magnitude higher than traditional HTS are ensured. Further, the resulting hits provide higher quality chemical matter starting points.
From Hits to Leads
A key aspect of our platform involves a sophisticated design strategy, which strategically transforms a given hit into a strong lead compound with novel patent position (IP) and initial drug properties. Once again, this is realizable on the same large scale with quick turnaround times as during the hit phase.
optimization towards Clinical candidacy
When optimizing a lead molecule, rather than focusing on a single property at a time, the multiple properties that are needed to move it towards drug candidacy are simultaneously addressed. This is done using molecular modeling and AI calculations along with experimental data. And as more data is generated, the calculations are iteratively refined and the models improved. This “iterative multiparameter optimization” scheme is a very unique aspect of our drug discovery platform.
While traditional drug discovery tends to spend 5–7 years in the pre-clinical stage with the synthesis of 5,000–10,000 compounds, we anticipate being able to lower the timeline to around 1–2 years with only ~50 compounds synthesized in total. Moreover, compounds moving into clinical trials are expected to endure a much lower failure rate (currently, the success rate in clinical trials is only 13.8%).