
---
title: "Built by Patient Zero"
description: "When the founder is Patient Zero, engineering timelines become biological countdowns. Discover how 33MED is stripping academic bloat to build a lean, autonomous computational immunology engine to halt Ankylosing Spondylitis."
date: 2023-10-24
author: "33MED Engineering"
category: "The \"Patient Zero\" AS Cure Methodology"
---
Built by Patient Zero: The Architecture of Absolute Urgency
In traditional biotech, timelines are measured in fiscal quarters, regulatory phases, and decade-long R&D cycles. But when the founder is Patient Zeroβactively fighting the irreversible progression of Ankylosing Spondylitis (AS) and syndesmophyte spinal fusionβtime is not a metric; it is a biological countdown.
At 33MED, our mission isn't abstract. We are building the foundational Phase 1 engine, ImmuNav, to autonomously identify molecular targets that halt spine fusion. To achieve this, we cannot rely on the sluggish, bloated infrastructure typical of academic research. There is no time for complex RabbitMQ message brokers, cumbersome Python wrappers, or convoluted database layers.
Urgency demands ruthless efficiency. We have stripped away the excess to execute a highly performant, lean Node.js, MongoDB, and React ecosystem. Here is the comprehensive engineering blueprint of how 33MED is turning raw computational immunology into a life-saving commercial SaaS platform.
𧬠Core Tool Matrix: 100% Free & Commercial-Safe
To safely build a commercial platform without restrictive licensing pitfalls, we architected our Node.js backend to bypass encumbered tools (like AlphaFold 3's non-commercial wrapper). Instead, ImmuNav orchestrates a suite of fully open, commercial-safe alternatives using native child_process CLI execution.
| Tool | Focus Area | Licensing / Commercial Access | Node.js Integration Strategy |
|---|---|---|---|
| Chai-1 / Boltz-1 | Multimodal Molecular & Structure Prediction | Free (Apache 2.0 / MIT). Chai-1 shifted weights to Apache 2.0, enabling full commercial SaaS utilization. | Execute via standalone pre-compiled CLI binaries, passing dynamically generated JSON/SMILES inputs. |
| GROMACS | Molecular Dynamics & Binding Thermodynamics | Free (GNU LGPL). | Spawn native gmx commands via Node child_process.spawn() to monitor real-time thread progress. |
| Bioconductor | Genomic / Transcriptomic Pathway Mapping | Free (Open Source). | Execute standalone R scripts via Rscript CLI to parse RNA/expression data sets. |
| Chaste | Tissue-Scale Cellular Calcification Modeling | Free (BSD License). | Fire the compiled C++ binary, passing an agent-generated XML payload mapping cellular parameters. |
| OpenSim CLI | Musculoskeletal Joint & Muscle Load Simulation | Free (Apache 2.0). | Programmatically execute opensim-cmd to validate physical mechanical stresses. |
ποΈ The Native Node/Mongo Swarm Engine
We rejected the dogma that AI swarms require dedicated message brokers. By utilizing MongoDB as a durable, atomic task queue, we maintain extreme horizontal scalability without the infrastructure overhead.
1. Atomic Schema Design
Our database architecture relies on two foundational collections. This structure natively tracks agent parameters and manages our robust task pipeline:
// Collection: agents
{
_id: ObjectId("66654a..."),
role: "BioinformaticsOperator",
status: "active",
globalObjective: "Identify molecular targets to halt syndesmophyte growth and spine fusion.",
humanGuidanceOverride: "Focus explicitly on small-molecule inhibitors targeting IL-17 pathways.",
lastPulse: ISODate("2026-06-08T22:00:00Z")
}
// Collection: tasks (The Lightweight Queue)
{
_id: ObjectId("66654b..."),
agentId: ObjectId("66654a..."),
tool: "gromacs",
status: "pending_review",
payload: {
inputFiles: ["target_sequence.fasta"],
md_params: { nsteps: 50000, dt: 0.002, integrator: "md" } // Validated by Zod
},
humanSignOff: {
isApproved: false,
actionedAt: null
},
logs: [],
outputMetadata: { pdbId: "1AKI", finalEnergy: -420000 }
}
2. The Background Processing Loop
Operating on Port 4000, our Simulation Worker runs a lightweight, recursive polling loop. It directly queries MongoDB for actionable jobs, strictly respecting a global pause state and mandatory human approvals before touching the silicon.
async function processQueue() {
// 1. Structural Check: Ensure swarm isn't globally paused
const globalStatus = await db.collection('settings').findOne({ key: 'swarm_status' });
if (globalStatus?.isPaused) {
setTimeout(processQueue, 5000);
return;
}
// 2. Fetch the next executable task (Atomic update)
const task = await db.collection('tasks').findOneAndUpdate(
{ status: 'queued', 'humanSignOff.isApproved': true },
{ $set: { status: 'processing', startedAt: new Date() } },
{ sort: { createdAt: 1 }, returnDocument: 'after' }
);
if (!task) {
setTimeout(processQueue, 2000); // Back-off if queue is empty
return;
}
try {
// 3. Native binary execution via Node Child Processes
const binaryProcess = child_process.spawn(getToolBinary(task.tool), task.payload.cliArgs);
binaryProcess.stderr.on('data', (data) => {
// Stream runtime logs directly into the MongoDB document
db.collection('tasks').updateOne({ _id: task._id }, { $push: { logs: data.toString() } });
});
binaryProcess.on('close', async (code) => {
if (code === 0) {
await handleTaskSuccess(task);
} else {
await db.collection('tasks').updateOne({ _id: task._id }, { $set: { status: 'failed' } });
}
processQueue(); // Instantly jump to the next job
});
} catch (error) {
await db.collection('tasks').updateOne({ _id: task._id }, { $set: { status: 'failed', error: error.message } });
processQueue();
}
}
π‘ Data Collection Engine: Harvesting the Cure
To cure a disease, you must first ingest the world's knowledge of it. Our Data Ingestion Agents operate as automated background tasks, harvesting raw scientific literature and deploying LLMs to structure biological insights.
- PubMed Sourcing (NCBI Entrez): Leveraging a free NCBI API key, our Node backend utilizes
axiosto query explicit keyword pairings (e.g.,Ankylosing Spondylitis AND syndesmophyte) at an accelerated rate of 10 requests/second. - arXiv Sourcing: We continuously poll the open arXiv Atom/XML feed (
cat:q-bio.*+AND+all:ankylosing), bypassing the need for authentication while capturing cutting-edge preprints. - Gemini Pipeline Integration: Raw XML text is piped directly to the Google Gemini SDK. By enforcing strict JSON formatting, we turn unstructured literature into a queryable biological database.
const systemPrompt = `You are an expert computational immunologist. Analyze the provided scientific text.
Extract any explicit molecular targets, binding affinities, genetic expressions, or pathways related to Ankylosing Spondylitis.
Return the output strictly in the following JSON format:
{
targetProtein: "string",
associatedCytokines: ["string"],
observedEffect: "string",
confidenceScore: 0.00
}`;
const result = await geminiModel.generateContent([systemPrompt, rawPaperText]);
const structuredData = JSON.parse(result.response.text());
await db.collection('knowledge_base').insertOne(structuredData);
ποΈ Command & Control: The HITL React Dashboard
AI autonomy is powerful, but when the stakes are biological survival, absolute executive control is mandatory. Our React Frontend (Port 5173) serves as a Human-In-The-Loop (HITL) Swarm Management Dashboard.
ββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββ
β IMMUNAV AUTOMATION HUB [ SYSTEM STATUS: RUNNING ] β
ββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββ€
β [ π GLOBAL SWARM PAUSE ] [ π RESUME SYSTEM ] [ βοΈ EDIT GLOBAL OBJECTIVE ] β
ββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββ€
β ACTIVE AGENTS β
β π€ Agent_Ingestion_1 β Status: [ WORKING ] β Task: Scraping arXiv q-bio β
β π€ Agent_Operator_3 β Status: [ BLOCKED ] β Task: Awaiting Config Approvalβ
ββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββ€
β π¨ AGENT CONFIGURATION APPROVAL GATEWAY β
β Task ID: #98322 β Tool Target: GROMACS Molecular Dynamics β
β Target Interaction: Small-molecule inhibitor binding to IL-17 Receptor β
β β
β [ View Agent-Suggested md_params Payload ] β
β { β
β "integrator": "md", β
β "nsteps": 50000, β
β "dt": 0.002 β
β } β
β β
β βοΈ [ Manually Edit JSON Parameters Inline ] β
β ββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββ β
β [ β REJECT CONFIGURATION ] [ β οΈ APPROVE & RUN JOB ] β
ββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββ
- The Global Interrupter: Toggling the Global Pause updates MongoDB to
{ isPaused: true }. The backend worker checks this boolean before processing, safely halting execution without crashing tasks mid-calculation. - The Guidance Injector: Directional pivots are captured via a clean text interface and appended to the agent's database document. The next Gemini prompt seamlessly merges your real-time strategic shifts into the system instructions.
- The Mandatory Approval Gate: Agents formulate hypotheses and design simulations, but they cannot execute without human sign-off. Tasks sit in
pending_reviewuntil parameters (validated viaZod) are manually verified, tweaked, and authorized on the dashboard.
π§ͺ Micro-Dataset Testing Protocol: Iterating with Safety
Developing computational physics architecture on local machines requires strict sandboxing to prevent API exhaustion or hardware throttling. ImmuNav is built on a rigid micro-testing protocol:
- Data Sourcing Sandboxing: We hardcode retrieval boundaries during development. Appending
&retmax=3to PubMed and&max_results=3to arXiv ensures our scraping, regex sanitization, and Gemini extraction pipelines process flawlessly on micro-batches before the data floodgates open. - Molecular Modeling Sandbox (1AKI): We do not benchmark Chai-1 or GROMACS scripts on massively complex spinal proteins. Instead, we pull Hen Egg White Lysozyme (PDB ID: 1AKI). Running an energy minimization sequence on this tiny, standard structure takes under 2 minutes, creating an ultra-fast feedback loop to verify
stdoutparsing and exit codes. - Mechanical Modeling Sandbox: Full biomechanical rigs are bypassed during UI development. We utilize the simple single-joint pendulum templates found within the OpenSim source files to ensure our React dashboards accurately render thermodynamic charts and parameter adjustments without bogging down the CPU.
When you are Patient Zero, building a biotech company isn't an academic exercise; it's a survival mechanism. By engineering out the bloat and maintaining a laser focus on high-performance execution, 33MED is transforming the pursuit of a cure from a distant hope into a programmable reality.
