Week 1 Lab: Pipetting

First time in a wet lab very exciting!

I learned about the different pipette ranges: P20 1–20 µL, P200 20–200 µL, P1000 100–1000 µL and when to use each one appropriately.

I also practiced proper pipetting technique holding the pipette vertically identifying the first and second stops when pressing the plunger and carefully controlling the release to ensure accurate and precise liquid handling.

I got a sneak peek at gel electrophoresis and how it can be used to separate DNA fragments.

Week 2 Lab: DNA Read Write Edit

In Lab 2 I learned how restriction enzymes can be used to cut DNA at very specific sequences, almost like precise molecular scissors. These enzymes recognize short DNA sequences called restriction sites and cleave the DNA at or near those locations, allowing us to deliberately fragment genetic material in a controlled way.

I also learned that restriction enzymes, or endonucleases, naturally come from bacteria. In their original context, they act as a defense mechanism by cutting up invading viral DNA, protecting the bacterial cell from infection. It was interesting to see how a biological immune strategy becomes a foundational lab tool.

We then discussed how CRISPR can be thought of as a generalized or programmable restriction enzyme. Instead of being limited to one fixed recognition site, CRISPR systems can be guided to almost any DNA sequence, making them far more flexible and powerful for gene editing.

I also learned how to use Benchling to simulate restriction enzyme digests in silico. We uploaded DNA sequences and tested different enzyme combinations to see how the DNA would be cut and what fragment sizes we would expect before actually running the gel.

To run a virtual digest, the DNA sequence has to be uploaded in a standard format, usually either FASTA or GenBank.

I learned in more detail how gel electrophoresis works and why DNA moves through the gel the way it does. Because DNA has a negatively charged phosphate backbone, it migrates toward the positive electrode when an electric field is applied. The agarose gel acts like a molecular sieve, so smaller DNA fragments move faster and travel further than larger ones, allowing the fragments to separate by size.

I weighed out agarose and mixed it with 1x TAE buffer to make a 1 percent solution. I microwaved it in short bursts until it fully dissolved, let it cool slightly, added SYBR Safe stain, poured it into the gel tray with a comb inserted, and allowed it to solidify to form wells.

I prepared the DNA digestion mixture by combining lambda DNA, the correct enzyme buffer, the chosen restriction enzyme or enzymes, and nuclease free water. I then incubated the tubes at 37 degrees Celsius so the enzymes could cut the DNA into fragments.

After the gel set, I removed the comb, filled the gel box with 1x TAE buffer, and mixed my DNA samples with loading dye. I carefully loaded each well without puncturing the gel and ran the gel at around 80 to 115 volts for about 45 minutes to separate the DNA fragments by size.

Once the run was complete, I transferred the gel to a blue light transilluminator, and captured an image of the separated DNA bands to analyze the pattern of fragments - there was a lot of noise but the experiment was fun nonetheless.

Week 3 Lab Automation

In Lab 3 I learnt about Opentrons and how lab automation can turn biology into something creative and visual. We used the Opentrons OT-2 pipetting robot to precisely deposit genetically engineered E. coli onto black charcoal agar plates. These bacteria were engineered to express fluorescent proteins in different colors, so when the plates were placed under UV light, the patterns we programmed glowed brightly.

It was a cool mix of automation and biology. Instead of manually pipetting, we let the robot handle the precise liquid handling, which made it possible to create detailed, glowing bio-art designs. It felt like combining coding, synthetic biology, and art into one project, and it gave a glimpse of how automation can scale up much more serious biological experiments too.

We learned how to use the Opentrons Python API to write a protocol, essentially a set of instructions that controls the robot’s pipettes. Instead of manually pipetting, we defined coordinates, volumes, and movement steps in code so the robot could deposit liquid precisely into specific wells to create a defined pattern.

Also we could simulate the protocol before running it on the actual robot. This let us preview how the design would look, check for mistakes, and adjust the pattern in software first.

https://opentrons-art.rcdonovan.com/

One of the coolest parts of this lab was using Opentrons Art, a tool built by TA Ronan that turns lab automation into a creative platform. Instead of writing everything from scratch in Python, this interface dramatically simplifies the workflow for creating agar-based designs. You can literally paint directly onto a virtual plate or upload an image, and the tool converts it into a protocol-ready layout for the robot.

What makes it profound is it’s become a living archive of art created by HTGAA students over time. It transforms a liquid-handling robot into a medium for expression, blending synthetic biology, automation, and visual design!

1. Write a description about what you intend to do with automation tools for your final project. You may include example pseudocode or Python scripts, procedures you may need to automate, 3D printed holders you may need, and more.

I want to use the Opentrons to prototype my bio-self healing blanket idea by automating two core parts of the project. First, I could screen different conditions that encourage biological mineralization or coating formation on scaffold materials. Second, I could test simplified self-healing systems where engineered cells or cell-free reactions deposit repair material in response to specific chemical damage signals. The robot is useful because it can run large combinatorial matrices of pH, ions, nutrients, and precursor concentrations with precision and consistency, and it can repeat dosing, media swaps, and sampling over time without constant manual pipetting.

In the first automated pipeline, I would distribute different mineralization conditions across a multiwell plate containing scaffold coupons. At set times, the Opentrons would refresh media, add precursor doses, and take small aliquots for downstream measurements. In the second pipeline, I would generate gradients of damage cues such as ionic strength or pH and then introduce cells plus repair precursors to see whether deposition localizes to the most damaged regions. This becomes a fast, reproducible way to test both the “architect” build phase and the “maintenance crew” repair phase of the concept.

2. Find and describe a published paper that utilizes the Opentrons or similar automation tools to achieve novel biological applications (eg automated PACE)

The paper DNA-BOT: a low-cost, automated DNA assembly platform for synthetic biology shows how researchers used an Opentrons OT-2 robot to automatically assemble DNA instead of doing everything by hand. They built 88 different genetic constructs in parallel, mixing and matching promoters and genes to explore lots of combinations quickly and cheaply. The big takeaway is that you don’t need an expensive biofoundry anymore a relatively affordable lab robot can handle high-throughput DNA building for everyday research labs.

https://academic.oup.com/synbio/article/5/1/ysaa010/5869449

Week 4 Lab: Protein Design I

Refer to week 4’s homework section for this weeks lab :)

Week 5 Lab: Protein Design II

Refer to week 5’s homework section for this weeks lab :)

Week 6 Lab: PCR & Gibson Assembly

In this two-day lab, we used PCR and Gibson Assembly to engineer mutations in the chromophore region of the purple Acropora millepora chromoprotein (amilCP) in order to generate a range of orange, pink, and blue colour variants. Two separate PCR reactions were performed to generate the DNA fragments required for Gibson Assembly. The insert PCR region extended from 24 base pairs upstream of the chromophore to just beyond the transcription terminator of the gene. The forward primer was specifically designed with an intentional mismatch to introduce a site-directed mutation into the mUAV plasmid DNA. After assembly, the mutated plasmids were transformed into chemically competent E. coli cells for expression and analysis of the resulting colour phenotypes.

We prepared a backbone reaction alongside four color-specific reactions: Blue, Light Pink, Magenta, and Orange.

Setup the following PCR reactions:

After the reaction mixtures were prepared, the tubes were placed into thermocyclers. The plasmid backbone PCR was run using a specialized program, while the colour mutation PCR reactions were run using a separate optimized cycling program.

We purified the PCR products using the Zymo DNA Clean & Concentrator Kit following the Zymo Research protocol based on silica adsorption.

We used Zymo-Spin purification columns to clean up the PCR products, performing two wash steps before eluting the purified DNA for storage. Since the provided protocol was designed for 50 μL PCR reactions, but our PCR reactions had a total volume of 25 μL, we used 20 μL of PCR product for purification while saving 5 μL separately, and scaled the remaining reagent volumes proportionally. Gel electrophoresis was then performed to verify successful DNA amplification.

As you can see from gel electrophoresis, the furthest left lane contained the DNA ladder, while Lane 1 contained the native plasmid control. Lanes 2–5 showed the expected PCR-amplified fragments for the Gibson Assembly, each appearing at approximately 650 bp. The gel results were very convincing, with strong bands at the expected fragment size and little to no evidence of primer dimers or nonspecific bands, indicating good PCR efficiency and high polymerase fidelity. The purified samples were then stored in the fridge until the following lab session.

On day 2, we took our PCR-generated DNA fragments and assembled them using Gibson Assembly. We diverged slightly from the standard protocol by using unpurified PCR products directly for the assembly reaction. This decision was made for a few reasons. First, we realized that we had very limited sample volumes for each colour mutant in both the purified and unpurified conditions, meaning we effectively had to choose between measuring concentration via NanoDrop or using the maximum possible amount of DNA for the assembly itself. In addition, several other groups reported extremely low, almost negligible, DNA concentrations after purification. Given our strong gel electrophoresis results, we felt it was reasonable to assume that the majority of DNA present in the reactions corresponded to the desired amplicon. We therefore proceeded with the unpurified PCR products while following the remainder of the Gibson Assembly protocol as written.

The reaction was incubated at 50°C in the thermocycler for 30 minutes.

We compared two chemically competent E. coli strains: DH5α and 10-beta. After thawing the competent cells on ice, we mixed 20 μL of cells with 4 μL from each Gibson Assembly reaction and incubated the mixtures on ice for 30 minutes to allow the plasmid DNA to associate with the bacterial membranes.

We also prepared an additional transformation reaction using only the native mUAV plasmid in DH5α cells as a positive control for successful transformation. For this control, I used 1 μL of plasmid DNA in an attempt to roughly match the DNA concentration of the Gibson Assembly samples.

Next, we again diverged slightly from the standard protocol. Instead of performing heat shock directly in the original transformation mixture, we transferred each transformation reaction into PCR tubes containing 100 μL of SOC medium and carried out the heat shock step in the thermocycler. The heat shock itself lasted only 45 seconds, after which the cells were immediately returned to ice to stabilize the bacterial membranes.

Following this, the cells were incubated in SOC medium for outgrowth to allow recovery and expression of the antibiotic resistance gene carried by the plasmid. Although the protocol recommended a 60-minute recovery period, our samples were incubated for closer to 45 minutes. Since we did not have access to a standard shaking incubator, we improvised a makeshift shaker using a pipette tip box to keep the cultures gently agitated during incubation.

We plated the entire incubation volume (~124 μL) from each transformation reaction onto LB-agar plates containing chloramphenicol. Glass beads were then used to evenly spread the bacterial suspension across the surface of the plates to promote uniform colony growth.

After 72 hours of incubation, the results were highly successful. We observed the targeted chromophore mutations across both E. coli strains.

The positive control confirmed that the transformation process had worked effectively. While some purple colonies corresponding to the native plasmid were present on all plates, each plate also displayed distinct coloured colonies including orange, pink, blue, and magenta. This indicated successful Gibson Assembly, transformation, and expression of the mutated chromoprotein variants.

Week 7: Neuromorphic Circuits and Mycelium

This two-day lab became a major source of inspiration for my final project!

Using a library of plasmids from the Ron Weiss Lab and HEK293 cells, we designed and built an intracellular artificial neural network (IANN). Unlike traditional synthetic genetic circuits that are largely limited to digital logic, IANNs can perform analog computation and act as universal function approximators, meaning that with enough intracellular artificial neurons they can generate highly complex and tunable cellular responses.

This directly inspired my project’s broader goal of combining externally controlled inputs, such as magnetic-field-guided receptor activation, with neuromorphic circuits to create more adaptive and dynamically programmable cellular systems.

Here were all the components we had available to us for the experiment.

Here is the circuit architecture I designed to run in the neuromorphic wizard:

The "Concentration" column will always be 50 ng/μL and the sum of all numbers in the "DNA wanted (ng)" column should never exceed 800.

On day 2, we visited the Weiss Lab, where Evan initiated the Opentrons workflow used to assemble our neuromorphic circuits. We also observed immortalized human cells under the microscope, which provided a firsthand introduction to experimental mammalian cell biology and the cellular systems underlying these genetic circuits.

Here was the results we obtained:

In the spirit of how to grow, we actually grew something too - mycelium biomaterials following this set of instructions with Ren!

Week 9 Lab: Protein Purification

This lab introduced the fundamentals of protein extraction and purification workflows commonly used in synthetic biology and bioengineering. It was particularly valuable for my final project, since our system required growing and extracting GFP protein before conjugating it to magnetic microparticles. It was also useful to understand how magnetic separation and purification methods can be integrated into biological systems, as my project similarly uses magnets both to purify ligand-conjugated particles and to actively control their interactions with cells. To isolate our protein of interest, we first grew the cells and then lysed them using a combination of B-PER (Bacterial Protein Extraction Reagent) and sonication, producing a lysate solution containing the total protein content of the cells.

To purify the fluorescent proteins from the lysate, we explored two different purification strategies: magnetic bead-based purification and Ni-NTA spin column purification. Both methods selectively isolate His-tagged proteins from the complex lysate mixture, but they rely on different physical mechanisms for separation and recovery.

The first method used functionalized magnetic beads, where magnets were used to immobilize bead-bound proteins during washing and elution steps. The second method used Ni-NTA spin columns, where centrifugal force was instead used to move buffers through a nickel resin that selectively binds His-tagged proteins. Comparing both approaches was particularly valuable for my final project, since it highlighted how magnetic particle-based systems can integrate purification, localization, and external control into a single experimental framework.

This image shows us inspecting the plasmid construct in Benchling, including the location of the His₆-tag (histidine tag) attached to the fluorescent protein coding sequence. The His-tag is important because it acts as a molecular handle that enables protein purification. During purification, the string of histidine amino acids strongly binds to nickel ions on Ni-NTA resin or functionalized magnetic beads, allowing the target protein to be selectively isolated from the rest of the cell lysate through washing and elution steps.

Procedure:

1. Magnetic beads were added to the lysate solution, allowing the tagged proteins to bind to the bead surface.

2. A 500 μL sample of the mWatermelon lysate mixed with magnetic beads was placed onto a magnetic rack, causing the bead-bound proteins to collect tightly against the magnet.

3. The remaining supernatant, containing excess buffer and unbound proteins, was carefully removed by pipetting.

4. The beads were washed with 500 μL of wash buffer containing a low concentration of imidazole (20 mM) to remove non-specific and weakly bound proteins. The sample was mixed and returned to the magnetic rack before the wash solution was removed.

5. To release the purified protein, 200 μL of elution buffer containing a high concentration of imidazole (500 mM) was added to the beads.

6. Once the beads recollected against the magnet, the fluorescent protein-containing liquid was removed and collected as Solution 4.

7. A second elution step was performed with an additional 200 μL of elution buffer to recover remaining fluorescent protein, producing Solution 5.

Procedure:

1. We combined 200 μL of Ni-NTA bead solution with 2 mL of cell lysate and incubated the mixture for approximately 30 minutes, allowing the His-tagged fluorescent proteins to bind to the nickel resin.

2. The mixture was transferred into a spin column and centrifuged at 8,000 RPM for 1 minute, producing a flow-through fraction that was collected for observation.

3. The resin was then washed with 500 μL of wash buffer and centrifuged again at 8,000 RPM for 1 minute to remove unbound and non-specific proteins.

4. To recover the purified protein, 200 μL of elution buffer was added to the column followed by a final centrifugation step at 8,000 RPM for 1 minute.

5. The final eluted fraction was analyzed to confirm the successful purification of the fluorescent protein.

Week 10 Lab: Mass Spectrometry

This week we did a lab at Waters Corp on Liquid Chromatography Mass Spectrometry, one of the core technologies used for modern protein characterization. Their lab was so cool!!

Using enhanced Green Fluorescent Protein (eGFP) as the model system, the lab showed how proteins can be analyzed at multiple levels ranging from overall molecular weight and folding state to their exact amino acid sequence. I found it especially interesting because the workflow progressively “breaks down” the protein from an intact structure into smaller peptide fragments, revealing different layers of biological information at each stage. We also briefly explored Charge Detection Mass Spectrometry (CDMS), which can analyze extremely large biological complexes that are too massive for conventional mass spectrometry techniques.

The lab was split into four rotating stations, each focused on a different stage of protein analysis:

In the first station, we used the Waters Xevo G3 QTof LC–MS system to analyze intact enhanced Green Fluorescent Protein (eGFP). The protein was first buffer-exchanged into ammonium acetate using spin columns before being run under denaturing LC conditions, allowing the mass spectrometer to determine its molecular weight from its mass-to-charge ratio (m/z) and charge states. I thought it was fascinating that proteins can essentially be “weighed” with such extreme precision using electrospray ionization and time-of-flight measurements.

The second part of the station explored how protein folding changes mass spectrometry behaviour. Instead of using chromatography, we directly infused eGFP into the Xevo G3 QTof using a syringe pump so the protein could remain in its native folded state. We then compared this against a denatured version created using formic acid. Folded proteins generate lower charge states because they are compact, while unfolded proteins expose more surface area and produce broader, higher charge state distributions. I found this especially interesting because it showed how mass spectrometry can probe protein structure and conformation, not just molecular weight.

In the second station, we moved into bottom-up proteomics using the Waters BioAccord LC–MS system. The eGFP protein was denatured, reduced, and digested with trypsin, which cuts proteins at lysine and arginine residues to generate smaller peptide fragments. These peptides were then fragmented further inside the mass spectrometer, allowing sections of the amino acid sequence to be reconstructed through peptide mapping. This was probably the most hands-on station and felt almost like molecular reverse engineering, rebuilding the protein sequence from fragmented spectral data.

The final station introduced Charge Detection Mass Spectrometry (CDMS) using the Waters Xevo CDMS system. Unlike conventional mass spectrometry, CDMS can directly measure both the charge and mass-to-charge ratio of individual ions, making it possible to analyze enormous biological assemblies that are too large for standard MS techniques. We used it to analyze Keyhole Limpet Hemocyanin (KLH), a huge multi-megadalton protein complex that exists in different oligomeric states. I thought this was one of the coolest stations because it showed how the same underlying physics can scale from relatively small proteins like GFP all the way up to molecular structures approaching the complexity of biological machines.