A new class of drugs called oligonucleotide therapeutics (OTs) can be tailored to a patient’s specific genetic sequence, but getting them into the right tissues in sufficient amounts remains the biggest hurdle. Developers add complex chemical modifications to improve delivery, but those same modifications make it difficult to track the drugs in the body, slowing down safety and effectiveness testing. Nanovery has built a prototype test called RNAmeter, based on nucleic acid nanorobotics (NANs), that measures OTs accurately in bodily fluids and tissues. It requires 80% fewer steps and is 75–90% faster than current methods. The problem is that designing NANs for each new OT currently relies on trial and error, which is slow and expensive. This project will use machine learning to replace that trial-and-error process. The team will generate training data, build predictive models, and create software that automates NAN design. Their goal is to cut turnaround time from 12 weeks to 2 weeks. If successful, this would let developers assess OT performance faster, accelerate drug development programmes, and ultimately get treatments to patients with genetic disorders more quickly.
View original technical description
A rising class of drugs 'Oligonucleotide Therapeutics'(OTs) offer the potential of treatments tailored to specific genetic sequences of patients, alleviating unique genetic disorders and hard-to-treat diseases. The biggest challenge OTs face is delivery to target tissues in sufficient quantities. To overcome this, developers have implemented complex chemical modifications on first-generation OTs that enhance delivery and performance. Simultaneously, the tools used to track second-generation OTs in the body are impacted by these chemistries, which makes assessing their effectiveness and safety during development difficult and prolonged. Nanovery has built a prototype test 'RNAmeter' based on Nucleic Acid Nanorobotics (NANs) that can measure OTs accurately and easily in a range of bodily fluids and tissues. RNAmeter provides top-level performance in detecting and quantifying all types of OTs, requiring 80% fewer steps and being 75-90% faster. The project will focus on improving the design process for NAN design for OTs, currently conducted by trial-and-error approaches due to their chemical complexity. This will be achieved by: 1\. Generation of training data sets 2\. Predictive model development based on ML approaches 3\. Software engineering to create an integrated computer pipeline for NAN design We aim to reduce turnaround and cost for new NANs for OTs from 12 weeks to 2 weeks. This will allow us to get data to developers faster and service more drug development programmes with RNAmeter for the ultimate benefit of patients.
Plain English summaries and category classifications on this site are generated by AI and may not perfectly reflect the original research.
Is something wrong? Let us know