Drug encapsulation is a crucial technology in pharmaceuticals and medicine. It involves enclosing active pharmaceutical ingredients within a carrier material. A wide range of materials and technologies are employed for encapsulation, including biopolymers, artificial polymers, and nanoparticles (NPs). NPs are particularly effective in drug encapsulation, as they can enhance the delivery and effectiveness of almost all types and classes of drugs. Some of the key advantages of drug encapsulation include: Protection (shield drugs from degradation by environmental factors, such as light, oxygen, and moisture); Targeted Delivery (allow precise targeting of drugs to specific organs or tissues, reducing side effects and increasing efficacy); Controlled Release (sustained release to ensure consistent therapeutic effect over time); Stability (extended shelf life and effectiveness); and Bioavailability (improved solubility and absorption of drugs, increasing their bioavailability and therapeutic impact).

Despite the many advantages of drug encapsulation, there are limited computational tools available to help “Smart” design of nanoparticles. Designing an encapsulated system requires careful consideration of several factors. The properties of the cargo play a significant role, as different cargos suit different encapsulation techniques. The physicochemical properties and degradation routes of the cargo influence the encapsulation method. The route of delivery and final formulation are also crucial, impacting encapsulation requirements. 

Here we propose to lay the groundwork for a tool that captures this knowledge, enabling AI/ML approaches to efficiently select the best encapsulation methods and dramatically accelerate the implementation of novel therapeutics to combat infectious diseases
The PhD will require a mix of techniques and skills spanning AI and laboratory based work.

Key project aims

The key project aims are:

• curate existing data for encapsulation decision making (in partnership with AI experts on data collection): When designing an encapsulated system there are several factors to be considered, including but not limited to the Properties of the cargo, Route of delivery/formulation, Controlled release, Solubility/dispersibility and size. This information is in the public domain but also protected by industry. The workshop will determine what are the key characteristic required to populate a framework and eventually an AI/ML tool. The first steps will be using open-source literature to populate the framework. 
• collect new experimental data towards closing gaps and confirming AI/ML recommendations. We will identify areas where targeted experimental studies will have the largest impact
• validate decision making tool: Ultimately, the framework and AI/ML tool will identify the most appropriate encapsulation tool for a specific drug cargo. This will need to be confirmed using traditional wet lab techniques. Formulation will be generated with encapsulation feasibility, efficiency, release profile and stability evaluated. This data will be fed back into the AI/ML tool with iterative testing required. Encapsulation efficiency and stability will be determined via development of suitable detection methodologies (such as HPLC, LCMS, ninhydrin). Characterisation of vesicles such as size, zeta potential and polydispersity index will also be determined (such as Malvern Zetasizer). In addition, stability/release of cargo load will be measured using in vitro models and microfluidic systems (such as USP-4)