Inspiration
The Blood Brain Barrier (BBB) has been known as a huge barrier to drug delivery and therapeutics. Nanoparticles and other drugs have been developed specifically targetting pemeating the BBB to target neurodegenerative diseases and symptoms. After attending the workshop from Jason Rock and discussing with him, we learned that he has been involved with Sunnybrook's latest reserach, being the World's first to enhance drug access to brain areas affected by ALS. Specifically, this research leveraged MRI scans to temporarily deactivate specific sections of the BBB, allowing drugs targetted to certain areas of the brain to be effective when they may have previously been inhibited by the BBB. Our team decided to take this a step further and harness AI to develop Addo Labs: an agentic research pipeline designed to discover existing drugs that previously could not pass the BBB, and use them to prolong life expectancy, reduce symptoms, or even discover a cure to ALS with this new MRI technology.
What it does
Addo Labs was built for clinicians and researchers. Users can upload a patient profile, which parses the information into digestible features that is provided to our Orchastrator Agent. This agent loads in the patient context and develops a research plan which calls on another 5 agents each aiming to discover a new drug targetted towards reducing a symptom, increasing life expectancy, or improving the patient's quality of life. Each goal is unique for every patient's case and medical history, and develops into briefs sent to 5 Research Agents in parallel. The Research Agents are equipped with the NCBI MCP and Pubmed tool calls that allow each agent to gather specific evidence and data to develop a drug profile designed to resolve it's specific brief and goal provided by the Orchastrator Agent. These drug profiles are then matched to it's most similar drug in the Drug Repurposing Knowledge Graph (DRKG) database: a repository containing over 97,000 entities matching drugs to it's affected genes, function, and structure. The top drugs are provided back to the Orchastrator Agent which uses web search functionality and drug database tool calls to rank the top drugs which have the most potential for the particular patient use case. It considers factors such as mechnism, delivery fit with and without the BBB, and safety. A comprehensive report is produced with appropriate references, allowing researchers and clinicians to pursue Addo Labs' findings further. A testing tool has also been developed that allows clinicians and researchers to monitor their patient's symptoms over time, ensuring the success or failures of tests.
How we built it
The frontend is React 19 + TypeScript on Vite, styled with Tailwind v4 and shadcn/ui, with the 3D knowledge-graph scene in three.js. Supabase handles sign-in, saved reports and take-home test results. The agents are a LangGraph pipeline on Claude (Opus orchestrates, Sonnet runs the research sub-agents and the final assessment, Haiku researches each candidate drug), served from a FastAPI backend that streams every stage live to the UI. LangSmith traced every agent's tool calls, thinking, inputs and responses. For matching, each target gene in a drug profile becomes a TransE query in DRKG's 400-dimensional embedding space, and every DrugBank compound is ranked by L2 distance to it. We used k-means to cluster DRKG's 97,000 entities into 24 "islands" so the graph can be explored in 3D. A verification step checks that every PubMed ID and clinical-trial number actually exists before it reaches the report. The take-home test uses a PyTorch facial-expression model (POSTER++), offline speech recognition (Vosk), and a browser tracing test.
Entire
Four of us built Addo Labs in parallel, each driving our own AI coding agents. Entire gave those agents a shared memory of the project, so one person's agent knew what another's had already built and decided. That cut down on duplicate work and kept the codebase organised. [Henri: what it actually did for us + screenshots]
Vultr
Vultr was essential in deploying Addo Labs and making it accessible and live to researchers and clinicians from around the world. By hosting all AI backend endpoints on Vultr's infrastructure, it's high compute allowed for easy deployment and hosting with a live domain (ai-ddo.study)!
Challenges we ran into
With four people each running AI agents in parallel, our biggest problem wasn't bugs; it was duplicated work. At one point we had three separate frontends and two copies of the same agent pipeline. We fixed this by claiming every task in GitHub Issues before writing code and connecting modules through JSON files instead of shared code. A second challenge was keeping the AI honest. Language models default to naming the same few famous drugs, so we made our research agents describe the ideal drug (target genes, mechanism, site of action) without naming any, and built a filter to strip drug names out. Tuning that filter so it didn't also strip proteins like TDP-43 took several iterations. Finally, a 3D picture of a 400-dimensional space is easy to misread, so we rejected our first visualisation (a single blob) and made sure every distance shown on screen is the true high-dimensional number.
Accomplishments that we're proud of
The full pipeline runs live end to end: upload a patient case, and a few minutes later you get a ranked, cited research report with a chat to question it. No citation in that report is invented, because every PubMed ID and trial number is verified before it appears. We're proud of splitting the work so the AI reasons about what a drug should do and a knowledge graph of 97,000 entities decides which drug does it, with candidates scored specifically on how much MRgFUS delivery helps them. We also shipped an at-home test (facial, speech and tracing) that clinicians can send to a patient as a one-time link.
What we learned
The intersection between medicine and technology is a lot closer than we previously imagined. AI has advanced tremendously over the past few years and has proven to be capable in more innovative projects with the potential to discover revolutionary drugs. We also learned how to use Entire and how much easier and organized the development process became when integrating AI tools.
What's next for Addo Labs
Validate results with real clinicians and researchers so that the tool can be used for patients like Alex suffering from ALS or other diseases. The workflow would be further tested, refined, and eventually have the capability to improve the drug discovery process and hopefully discover a cure to terminal diseases like ALS.
