Researchers in the biotech, life sciences, and drug discovery industries are making significant strides as a result of developments in cloud computing and artificial intelligence. In order to speed up the discovery and development of novel pharmaceuticals, emerging platforms powered by artificial intelligence and machine learning can quickly sift through enormous amounts of data. The UK-based business Kuano is adding quantum detail to structure-based drug development leveraging computer advances in biomolecular simulation and artificial intelligence, and the company’s preliminary findings point to promising new directions for therapeutic fields including Alzheimer’s disease research.
As cofounder and CTO David Wright, PhD, noted, “While first efforts have provided critical validation with anti-disease Alzheimer’s medications, our platform permits fresh approaches to numerous therapeutic areas.” “We can develop chemicals more quickly and effectively than earlier methods.”
The fundamental physics of how enzymes speed up biochemical processes in the body, particularly a configuration along the reaction route known as the transition state where potential energy is at its highest, is used by Kuano’s approach to drug creation. Kuano employs the particular conformation and chemistry of the transition state to effectively build drug-like compounds that resemble our natural chemistry, in contrast to conventional inhibitor design, which starts with the initial binding.
A PROOF-OF-CONCEPT BASED ON REPRODUCING RESULTS FROM PREVIOUS WORK FOR Purine Nucleoside Phosphorylase (PNP) AND EXPERIMENTAL WORK ON THE PNP REACTION WAS WHAT KUANO FIRST WANTED. FROM VALIDATION TO SIMULATION TO WETLAB. In order to model the reactions and look for stable compounds that matched the transition state features identified, quantum mechanics and molecular mechanics (QMMM) simulations were carried out. Both successfully reproduced processes served as an initial validation of the Kuano approach and platform.
QMMM is used to mimic the enzyme environment and capture the reaction chemistry for the in-silico synthesis of transition states. Advanced sampling methods in conjunction with QMMM simulations enable characterising the entire energy landscape surrounding the event, finding the most likely reaction pathway, and highlighting the critical transition states. Along the reaction route, where a tighter, more complex state occurs, enzyme-catalyzed reactions typically have several transition states. To determine the ideal transition state to target, a huge number of high-quality simulations are needed.
Kuano’s quantum pharmacophore (QP), which captures important aspects of 3D conformation and chemistry in this essential, tightly bound state, is utilised to analyse the structural and dynamic information obtained once a transition state target has been discovered. The kinds of traits that Kuano’s QP finds in the transition stage are what set it apart from other QPs.
Traditional pharmacophores are mostly based on 3D shape; ours goes farther by incorporating quantum characteristics to capture the chemistry in the transition state, according to Wright. After being created, the QP is used to rate various compounds based on how comparable corresponding properties in the candidate compound and transition state are. Through genetic algorithms, machine learning models, or preexisting libraries, candidate chemicals can be found. Then, promising compounds are created and evaluated for effectiveness against the necessary target in a wet lab.
A QUALITY TEMPLATE FOR DRUGS DESIGN APPLIED
Kuano focused on applying the platform to studies on Alzheimer’s where there was a lack of experimental transition state data. Leading collaborators Paul Fish and Fredrik Svensson identified NOTUM as a viable target for Alzheimer’s disease therapy while working with the Alzheimer’s Research UK UCL Drug Discoveries Institute. They were also interested in innovative approaches that held more promise for new discovery.
An enzyme called NOTUM takes chemical tags off of particular proteins in order to control particular signalling pathways in the body. Hundreds of QMMM simulations had to be run in order for the Kuano platform to simulate the enzyme removing the palmitoleic acid (PAM) tag off the Wnt protein. The platform was able to deduce the reaction pathway and the target transition states for each reaction step, identifying the second reaction step as the rate-limiting step and the intended transition state. Without any prior knowledge, the Kuano platform was able to effectively replicate one powerful NOTUM inhibitor and also identified two interesting novel compounds that were synthesised, tested, and found to be powerful NOTUM inhibitors.
When compared to the typical scope of screening for medication candidates, less than 100 chemicals were tested to find these novel ones, according to Wright. One of these two compounds “represents a breakthrough into new chemical space, which indicates the potential to identify novel medications conventional approaches cannot.” One of these two compounds is “totally unlike all known NOTUM inhibitors.”
A GLOBAL COLLABORATORY TEAM APPROACH WITH ORACLE
To enable the research of new drug targets and the discovery of interesting new compounds much more quickly than before, large-scale computation is crucial for drug discovery. A typical transition state search, for instance, consists of hundreds to thousands of individual calculations connected to one another via a dependency tree. Research like this may now be completed in weeks and days rather than years and months because to increased processing capacity.
Kuano has access to cutting-edge computational power, the most recent developments in AI/ML, technical know-how, funding, and more when working with Oracle for Research.
The capacity to quickly deploy massive sets of simulations to find transition states, then effectively screen against large libraries of candidate drugs or generate them using AI models, has been a key component of the success of Wright’s large-scale simulation and screening technique.
Other international partners are also contributing to the work of Wright and his colleagues. “Only with the support of partners will it be feasible to rapidly establish a pipeline of transformational pharmaceuticals,” he stated. “Drug development is a team sport, and we are always seeking for possibilities to engage with academic institutions, biotech businesses, or pharmaceutical companies to develop cutting-edge therapies to treat the world’s most serious diseases.”
THE FUTURE SEEMS WELL-FINANCED AND PROGRESSIVE
Venture capitalists are investing heavily in these revolutionary areas of the living sciences, with more than $35 billion going into biotech firms just in the last two years. Emerging thinkers like Kuano can keep pushing the limits of computational chemistry to progress discoveries more quickly and effectively than earlier methods thanks to substantial entrepreneurial funding and potent computer resources.
“Our next step is to create more improved molecules that are more akin to potential medications. As we proceed with the process of developing a new generation of medications, we are excited to collaborate with Oracle, Wright said.
With a focus on oncology and the development of compounds targeting the NOTUM target as colon cancer treatments, Kuano is currently deployed with a number of new drug design projects.
