opentrons-protocol-agent

Opentrons Protocol Agent

ID: biomedical.lab_automation.opentrons_agent Version: 1.0.0 Status: Experimental Category: Lab Automation / Robotics

---

Overview

The Opentrons Protocol Agent is an LLM-driven tool designed to bridge the gap between experimental design and physical execution. It translates natural language experimental descriptions (e.g., "Serial dilution of samples A-H") into valid, executable Python protocols for Opentrons OT-2 and Flex robots.

This agent empowers "wet lab" automation by allowing researchers to control liquid handling robots via chat interfaces, reducing the barrier to entry for coding complex automation scripts.

---

Key Capabilities

1. Protocol Generation

• Converts text instructions into opentrons.protocol_api Python scripts.
• Supports common workflows: PCR prep, ELISA, serial dilutions, aliquoting.
• Automatically calculates volumes and well positions.

2. Labware Verification

• Validates that requested labware (plates, tipracks) are compatible and available in the standard library.
• Suggests appropriate pipettes (P20, P300, P1000) based on transfer volumes.

3. Safety & Validation

• Performs "virtual run" simulations to check for errors (e.g., running out of tips, aspirating air).
• Includes safety comments and deck setup instructions in the generated code.

---

Usage

Example Prompt

Create an Opentrons protocol for a serial dilution.
- Source: Column 1 of a 96-well plate (Corning 3635) containing 200uL stock.
- Destination: Columns 2-12 of the same plate.
- Dilution factor: 1:2 (transfer 100uL, mix 3 times).
- Pipette: P300 Single-Channel Gen2.
- Tips: Opentrons 300uL tiprack on slot 1.

LLM Agent Integration

@tool
def generate_opentrons_protocol(
    experiment_type: str,
    source_labware: str,
    dest_labware: str,
    transfer_volume: float,
    replicates: int = 1
) -> str:
    """
    Generates a Python script for an Opentrons robot.
    
    Args:
        experiment_type: 'serial_dilution', 'pcr_prep', 'reformatting'
        source_labware: API name of source plate
        dest_labware: API name of destination plate
        transfer_volume: Volume in uL
    """
    # Logic to construct the protocol string
    pass

---

Dependencies

• opentrons>=7.0.0
• pandas (for sample mapping)

References

• Opentrons Python API V2 Documentation
• "Self-driving laboratories" (Gomes et al., Nature 2023)

---

Author

MD BABU MIA Artificial Intelligence Group Icahn School of Medicine at Mount Sinai md.babu.mia@mssm.edu