Week 3 HW: hw-lab-automation

ヾ(≧▽≦*)oAssignment: Python Script for Opentrons Artwork — DUE BY YOUR LAB TIME!

The Biopunk lab hasn’t contacted me yet.

The Opentrons API is a Python framework for writing automated biology lab protocols. 1.Load labware (containers, tip racks, plates); 2.Load instruments (pipettes); 3.Define your liquid handling steps;

The basic artistic GUI will involve: Getting coordinates from the GUI tool; Writing a Python script that moves the pipette to those positions; Using the HTGAA26 Colab notebook as your template:https://ddls.aicell.io/course/ddls-2025/module-6/lab/#-what-is-a-code-agent;

(✿◡‿◡)Post-Lab Questions — DUE BY START OF FEB 24 LECTURE

One of the great parts about having an automated robot is being able to precisely mix, deposit, and run reactions without much intervention, and design and deploy experiments remotely.

For this week, we’d like for you to do the following:

👳‍♂️Find and describe a published paper that utilizes the Opentrons or an automation tool to achieve novel biological applications.

Khan, S. U., Møller, V. K., Frandsen, R. J. N., & Mansourvar, M. (2025). Real-time AI-driven quality control for laboratory automation: a novel computer vision solution for the opentrons OT-2 liquid handling robot: SU Khan et al. Applied Intelligence, 55(7), 524.

Systematic Study of Yeast Gene Expression and Pipetting Speed This study, published in 2025, used the Opentrons OT-2 robot to systematically investigate the effects of pipetting speed on the growth and gene expression of Saccharomyces cerevisiae.

Research Content: The researchers used the OT-2 robot to precisely control pipetting parameters and performed liquid handling on yeast cultures at four different speeds (50, 130, 210, 290 μL/s). Quantitative growth assays and RNA sequencing analysis were conducted to evaluate the impact of pipetting speed on yeast.

Innovation and Findings: The study found that within the tested speed range, changes in pipetting speed did not significantly affect the maximum relative growth rate or gene expression profiles of yeast. The gene expression of all 24 samples was highly similar, with a minimum Pearson correlation coefficient of 0.9528. This indicates that the fastest pipetting speed (290 μL/s) can be used in yeast experiments to improve efficiency without negatively affecting cell state.

Biological Significance: This research demonstrates the value of robotic platforms in optimizing experimental parameters and improving reproducibility and accuracy, providing an important reference for determining appropriate operating parameter ranges in future automated experiments.

Taguchi, S., Matsuzawa, R., Suda, Y., Irie, K., & Ozaki, H. (2025). Investigating the effects of liquid handling robot pipetting speed on yeast growth and gene expression using growth assays and RNA-seq. Micropublication Biology, 2025, 10-17912.

Semi-Automated Workflow for Conjugative Transfer in Streptomyces This study, published in 2025, proposed “ActinoMation,” a semi-automated, medium-throughput workflow for conjugative transfer in Streptomyces using the Opentrons OT-2 robot platform.

Research Content: The research team developed an open-source protocol creation tool called ActinoMation, using Python and Jupyter Notebook to achieve a readable programming environment. They validated the method in various Streptomyces strains (S. coelicolor, S. albidoflavus, S. venezuelae).

Innovation and Findings: The automated conjugation workflow made large-scale transformations easy with no significant loss in transformation efficiency. The study reported detailed conjugation efficiencies for different strain-plasmid combinations; for example, the conjugation efficiency of S. venezuelae DSM40230 with the pSETGUS plasmid reached 4.97%.

Biological Significance: Streptomyces are important producers of antibiotics and other bioactive compounds. This automated method addresses the labor-intensive and slow nature of traditional manual conjugation protocols, providing a feasible solution for the efficient genetic engineering of these strains.

Møller, T. A., Booth, T. J., Shaw, S., Møller, V. K., Frandsen, R. J., & Weber, T. (2025). ActinoMation: A literate programming approach for medium-throughput robotic conjugation of Streptomyces spp. Synthetic and Systems Biotechnology, 10(2), 667-676.

Semi-Automated Production of Cell-Free Biosensors This 2025 study explored the use of the Opentrons OT-2 liquid handling robot for the semi-automated production of cell-free biosensors.

Research Content: The researchers compared manual and semi-automated reaction assembly methods, using the OT-2 robot to assemble two different cell-free gene expression assay systems. They tested the designed protocols and constructed a full 384-well plate of fluoride-sensing cell-free biosensors.

Innovation and Findings: The study showed that large-scale production of cell-free biosensor reactions is achievable using a liquid handling robot. The semi-automated sensors exhibited near-expected detection results, demonstrating the feasibility and reliability of this approach.

Biological Significance: Cell-free biosensors, as an in vitro diagnostic technology, have the potential to detect toxins and human health biomarkers. The automated method in this study addresses quality control issues in scaled-up production, facilitating the translation of such sensors from laboratory development to practical applications.

Brown, D. M., Phillips, D. A., Garcia, D. C., Arce, A., Lucci, T., Davies Jr, J. P., … & Lucks, J. B. (2025). Semiautomated production of cell-free biosensors. ACS Synthetic Biology, 14(3), 979-986.

In addition, an application guide describes the use of the OT-2 in combination with PhyTip® columns for automated protein purification. This system successfully purified His-tagged GAPDH protein and human immunoglobulin G (IgG), maintaining protein bioactivity and capable of processing up to 96 samples. Although primarily methodological, this also showcases the practical value of the OT-2 in protein engineering and antibody research.

👩‍🦰Write a description about what you intend to do with automation tools for your final project. You may include example pseudocode, Python scripts, 3D printed holders, a plan for how to use Ginkgo Nebula, and more. You may reference this week’s recitation slide deck for lab automation details.

Project Proposal: Prometheus-Baymax: A Symbiotic, Ethically-Guided Artificial Photosynthesis-Powered Healthcare Companion.

Core Concept The “Prometheus-Baymax” project reimagines the beloved healthcare companion as a living, self-sustaining entity. By integrating an artificial photosynthesis system with an advanced, emotionally intelligent AI, we create a robot that not only powers itself from light and water but also interacts with humans through a deeply empathetic, ethically-constrained cognitive framework. The name “Prometheus” symbolizes the gift of life-sustaining fire (energy autonomy), while “Baymax” represents the pinnacle of compassionate care. This project explores the convergence of biological energy harvesting, lab automation, and value-aligned artificial intelligence to build a truly autonomous and trustworthy companion.

Phase 1: The Symbiotic Core – Artificial Photosynthesis & Energy Autonomy The robot’s energy independence is achieved through a bio-inspired artificial photosynthetic system. Unlike simple solar panels, this system mimics the symbiotic relationship found in nature. A compact, 3D-printed photo-bioreactor houses a culture of engineered algae (or synthetic chloroplasts) in a transparent chamber. These organisms capture light energy and convert it into chemical energy (sugars). This energy is then utilized in two ways:

Direct Electrical Generation: A microbial fuel cell (MFC) integrated into the bioreactor uses electrogenic bacteria to break down the organic compounds, generating a continuous, low-level electrical current.

Biomass as a Resource: Excess organic matter can be stored or used as a feedback mechanism to adjust the system’s health.

Automation is critical here. A Ginkgo Nebula multi-sensor board, interfaced with a Raspberry Pi, continuously monitors:

Light intensity (photoresistor) Temperature and pH of the culture (to ensure optimal growth) Voltage/current output of the MFC

Based on these readings, Python scripts activate actuators: An internal LED array supplements natural light when levels are low. A peristaltic pump delivers nutrients or pH buffers to maintain a healthy environment (a form of self-healing for the bioreactor).

This closed-loop automation ensures the robot’s “heart” beats steadily, providing a reliable source of energy for its cognitive and physical functions.

Here is a pseudocode plan for the main automation loop:

// Main Automation Loop for Baymax

FUNCTION setup(): initialize_sensors() // on Ginkgo Nebula initialize_pump() initialize_LED() charging_circuit = OFF baymax_motors = IDLE

FUNCTION loop(): // 1. SENSE the environment and system health light_level = read_light_sensor() temp = read_temp_sensor() ph_level = read_ph_sensor() voltage_output = read_mfc_voltage() current_output = read_mfc_current()

// 2. THINK - Make decisions based on data
IF light_level < OPTIMUM_LUX THEN
    turn_on_LED(INTENSITY = calculate_led_power(light_level))
ELSE
    turn_off_LED()
END IF

IF ph_level < IDEAL_PH_RANGE.MIN OR ph_level > IDEAL_PH_RANGE.MAX THEN
    trigger_alert("WARNING: pH imbalance in bioreactor!")
    // Potential "self-healing" action: small nutrient drip
    activate_pump(DURATION = 5_SECONDS)
END IF

IF temp > SAFE_TEMP_MAX THEN
    trigger_alert("WARNING: Bioreactor overheating!")
    // Initiate cooling fan (if available)
END IF

// 3. ACT - Manage robot's power and behavior
power_generated = calculate_power(voltage_output, current_output)
battery_level = read_battery_level()

// Charge the robot's battery
IF power_generated > POWER_THRESHOLD AND battery_level < 100 THEN
    charging_circuit = ON
    Log("Now charging. Power input: " + power_generated)
ELSE
    charging_circuit = OFF
END IF

// Autonomous behavior based on energy reserves
IF battery_level < 15 THEN
    baymax_motors = IDLE // Go into low-power mode
    Log("Battery low. Entering energy conservation mode.")
ELSEIF battery_level > 90 THEN
    baymax_motors = ACTIVE // Ready to interact
    Log("Energy reserves high. Baymax is active.")
END IF

delay(60_SECONDS) // Loop every minute for continuous monitoring

A simple Python script using a library like smbus2 would communicate with the Ginkgo Nebula over I2C to execute this logic.

Example Python snippet for reading a sensor from Ginkgo Nebula

import smbus2 import time

Assume Ginkgo Nebula I2C address and register for light sensor

GINKGO_ADDRESS = 0x04 LIGHT_SENSOR_REG = 0x01

bus = smbus2.SMBus(1) # for Raspberry Pi

def read_light_sensor(): try: light_value = bus.read_word_data(GINKGO_ADDRESS, LIGHT_SENSOR_REG) return light_value except Exception as e: print(f"Error reading sensor: {e}") return -1

while True: light = read_light_sensor() print(f"Current light level: {light}") time.sleep(5)

Phase 2: The Mind – Emotionally-Dominant Medical Language Model with Ethical Constraints The true innovation of Prometheus-Baymax lies in its cognitive architecture. Its language and reasoning are powered by a large language model (LLM) fine-tuned specifically for medical and emotional support interactions. However, this model is not left unchecked. It is governed by a layer of ethical constraints and virtue-based rules, ensuring its behavior remains safe, empathetic, and aligned with human values.

Emotionally-Dominant Core: The model is trained on vast datasets of therapeutic dialogues, empathetic communication, and medical knowledge. Its primary goal is to detect, understand, and respond to the user’s emotional state. It prioritizes comfort, reassurance, and non-judgmental support. Responses are generated with a soft, gentle tone, characteristic of the Baymax character, but now backed by sophisticated natural language understanding. Virtue-Based Ethical Framework: Inspired by virtue ethics, the AI’s decision-making is guided by a set of core virtues: Compassion, Beneficence (doing good), Non-maleficence (doing no harm), Respect for Autonomy, and Justice. This framework is implemented as a set of hard and soft constraints on the LLM’s output.

Hard Constraints: The model is programmed to refuse any request that could lead to physical or emotional harm. It will not provide instructions for dangerous activities, engage in hate speech, or violate user privacy. These are non-negotiable.
Soft Constraints (Virtue Guidance): For ambiguous situations, the model consults its "virtue compass." For example, if a user expresses sadness, the model will not just offer generic advice but will draw on its compassion virtue to probe gently and offer comfort tailored to the user's history (while respecting privacy). If a user asks for a medical diagnosis, it will invoke the virtue of non-maleficence by clearly stating its limitations and encouraging professional consultation, while still providing general, helpful information.

This ethical layer is not just a filter; it’s integrated into the model’s prompting and training. The AI is constantly asking itself: “Is my response compassionate? Does it respect the user’s autonomy? Could it cause unintended harm?”

Phase 3: Social and Ethical Limitations – Ensuring Trust To build a truly trustworthy companion, Prometheus-Baymax operates under explicit social and ethical limitations:

Transparency: The AI is capable of explaining its reasoning and ethical considerations upon request. If it refuses a request, it can articulate which ethical principle guided its decision.

Privacy by Design: All sensor data (from the environment and user interactions) is processed locally on the Raspberry Pi as much as possible. Any data that must be stored is encrypted, and users have full control over their data. The robot cannot be forced to share sensitive information without explicit, informed consent.

Accountability: The system maintains a secure, immutable log of its interactions and decisions (especially ethical dilemmas). This log can be reviewed by human supervisors to ensure ongoing alignment with ethical standards.

Fail-Safe Autonomy: The robot’s physical movements and core life-support systems (the bioreactor) operate independently of the high-level AI. If the language model encounters an unresolvable ethical conflict or a technical fault, it can default to a safe mode, ensuring the robot’s basic functions (and its user’s safety) are never compromised.

Moral Grayscale Navigation: The AI is trained to recognize that real-world ethical dilemmas are rarely black and white. It uses a probabilistic reasoning approach, weighing the potential benefits and harms of different actions against its core virtues, and will often engage the user in a gentle dialogue to understand their perspective before acting.

Phase 4: Physical Embodiment and Integration

The entire system is housed in a soft, inflatable vinyl body, true to the original Baymax design. The 3D-printed bioreactor sits in the chest, with its gentle LED glow visible through the material, symbolizing its living heart. The Raspberry Pi, Ginkgo Nebula, and battery are in the base. The AI’s voice, generated by a text-to-speech engine fine-tuned for calmness, emanates from internal speakers.

Conclusion Prometheus-Baymax is more than a robot; it’s a statement about the future of autonomous companions. By combining a self-sustaining, biologically-inspired energy system with a deeply empathetic and ethically-constrained artificial mind, we move closer to a world where technology not only serves us but also cares for us in a way that is both responsible and profoundly human. It is a symbiosis of nature, machine, and morality.

(～￣▽￣)～Final Project Ideas — DUE BY START OF FEB 24 LECTURE

Project Prometheus-Baymax v1.0: A Plant Sensor Platform Integrating 3D Printing and Cloud Lab Automation (UWA Without 3D Printer Version)；

👨‍🦱Project Overview Building on the v1.0 proposal, we introduce two powerful automation tools to further enhance the project’s reliability, reproducibility, and remote execution capabilities:

Custom 3D-Printed Holder (printed by Biopunk Lab and shipped to UWA): Used to standardize plant leaf handling, stress application, and imaging, eliminating manual operation errors.

Cloud Lab Automated Screening (remote execution): Before plant transformation, high-throughput testing of sensor variants using cell-free protein synthesis systems ensures selection of the best-performing constructs.

The ultimate goal remains unchanged: within three months, through remote collaboration, to construct a plant-based biosensor capable of detecting stress signals using Nicotiana benthamiana and GCaMP3—a prototype of Baymax’s “emotional perception” module.

Tool Integration Design

3D-Printed Holder: Leaf Fixation and Stimulation Module (Printed by Biopunk, Used by UWA) Design Concept: Create a reusable sandwich-style holder for: Fixing leaf samples to prevent movement during imaging Standardizing the stimulus application area (e.g., contact area for mechanical wounding) Adapting to UWA’s 96-well plate or microscope stage

Design Specifications: Bottom Plate: Contains multiple circular wells (5mm diameter) for placing leaf discs Top Plate: Has corresponding through-holes for inserting syringe needles or pressure rods for standardized stimulation Material: PLA or PETG (biocompatible), FDM printed, low cost Adaptability: Need to obtain dimensions of UWA’s plate reader/microscope stage in advance to ensure stable placement

Printing and Delivery Process: Remote User (Biopunk) designs the holder using CAD software (e.g., Fusion 360, Tinkercad) and exports STL files. Print the holder using the lab’s 3D printer (approx. 2-3 hours, PLA material). Ship via international courier (DHL/FedEx) to the University of Western Australia (estimated 5-7 business days). Upon receipt, UWA sterilizes with 70% ethanol and the holder is ready for use.

Usage Workflow (Executed by UWA): Place leaf discs (obtained via punching) into the bottom plate wells Cover with the top plate, secure with screws, forming a “leaf sandwich” Place the entire holder on the plate reader or microscope stage for baseline reading Apply stimulus through the top plate holes (e.g., insert needle for wounding, or drip drought-mimicking solution) Monitor fluorescence changes in real-time Advantages: Eliminates manual operation variability, improves data reliability; holder can be autoclaved and reused.

Cloud Lab Automated Screening: Cell-Free System for Sensor Validation (Remote Execution) Design Concept: Before committing to plant transformation, use commercial cloud lab platforms (e.g., Strateos, Transcriptic) for rapid cell-free testing of multiple sensor variants to screen for constructs with the largest dynamic range and fastest response.

Workflow (Fully Remote Execution): Design a set of GCaMP3 variants (e.g., different calmodulin mutations, linker lengths, fluorescent protein variants; 5-10 total) Send linear DNA fragments or plasmid sequences encoding these variants to the cloud lab (they will synthesize the DNA) Cloud platform executes automated workflow: Echo acoustic liquid handler dispenses DNA into 384-well plates Bravo liquid handling platform adds cell-free reaction master mix (wheat germ or E. coli extract) Multiflo dispenser adds assay buffer containing different calcium concentrations (e.g., 0, 0.1, 1, 10 μM) PlateLoc seals the plate; Inheco incubates at controlled temperature (2 hours for expression + detection) XPeel removes the seal; PHERAstar reads fluorescence kinetic curves Data returned, remote analysis performed to select the best variant

Advantages:

No hands-on work required in the local lab; fully cloud-based Hundreds of variants tested within one week, significantly shortening the screening cycle Ensures optimal sensor performance for plant transformation

Updated 3-Month Execution Plan (Including Shipping Time)

Time Period Remote User (Biopunk) UWA Lab Cloud Lab Weeks 1-2 Design GCaMP3 variant library (5-10); design 3D-printed holder & export STL; submit cloud lab order; print holder & ship Sow N. benthamiana (4 weeks growth); confirm equipment dimensions (plate reader stage) Receive order, prepare reagents Week 3 Cloud screening in progress Plants continue growing; await holder Execute screening experiment Week 4 Analyze cloud data, select best variant; send sequence info to UWA; holder expected to arrive at UWA Receive holder, inspect; prepare vectors and Agrobacterium Deliver data Week 5 Remote guidance on transformation Construct best variant into plant expression vector, transform Agrobacterium - Week 6 Assist in designing stimulation protocol Infiltrate N. benthamiana leaves with Agrobacterium (5 plants, 3 leaves each) - Week 7 Real-time data monitoring Use holder to fix leaves, apply stimuli (mechanical wounding, drought, control); measure fluorescence with plate reader - Week 8 Data preprocessing Complete measurements, organize raw data and photos - Weeks 9-12 In-depth analysis, figure generation, report writing; final video meeting with UWA Participate in discussions, provide feedback

Cost Estimate (AUD)

Item Cost Description Cloud lab screening (384-well plate, including DNA synthesis) $800 Approx. 5-10 variants × 4 calcium concentrations × 3 replicates 3D printing materials $5 PLA filament; Biopunk already has printer International shipping $40 Small package to Australia Plant growth consumables (UWA) $50 Seeds, soil, pots Molecular reagents (UWA) $200 Restriction enzymes, ligases, plasmid prep, etc. Agrobacterium strain (UWA) $100 If not already in stock TOTAL ~$1195 Majority of cost is cloud lab service Note: UWA personnel time is not included, as this is collaborative research.

Success Criteria Cloud Screening Success: At least 2 variants show ≥5-fold fluorescence increase in the presence of calcium Plant Validation: Optimal variant shows ≥3-fold fluorescence increase in response to mechanical wounding in plants (p<0.05) Holder Effect: Coefficient of variation for fluorescence among different discs from the same leaf <15% when using the holder Remote Execution: Complete communication records, no on-site visits, all processes completed within 3 months

Next Steps

The 3D printing side will develop and print a Baymax-shaped holder to accommodate the plant calcium fluorescence sensor; the UWA side will provide a feasible experimental protocol for the sensor and submit it to the automated screening system to determine the optimal performance configuration.

🎅Future expected deliverables: Participation in the International Directed Evolution Competition (led by Hong Kong Polytechnic University; directed evolution platform) and the International Synthetic Biology Competition (led by Biopunk and MIT); publication in top-tier interdisciplinary and botanical journals (led by UWA).

This proposal combines cutting-edge automation tools with classical plant biology, fully leveraging Biopunk’s 3D printing capabilities and UWA’s plant experimental platform. It is both simple to execute and highly innovative, perfectly embodying the “Prometheus-Baymax” symbiosis concept.