Results

As a first-year iGEM team, all of our members dedicated themselves to tackling the urgent challenge of antimicrobial resistance (AMR). However, our journey was full of ups and downs. With limited experience in wet lab work, restricted access to lab equipment, and a school schedule that often left us with little lab time, we faced many obstacles. On top of that, our lab was not always available, and as a brand-new team, we were constantly encountering experimental processes and unexpected outcomes that were entirely new to all of us. Despite all that, we still gave our very best, pouring in our time and effort just to bring our ideas to life in the lab. Even though we did not manage to finish every experiment, what we gained was just as valuable. The experience of trying, learning, and growing together as a team. The setbacks did not stop us, they reminded us why we started, and gave us the determination to keep moving forward.

To construct our donor plasmid, we used the pSB1C3 backbone with mCherry and Chloramphenicol resistance as the base. Into this backbone, we aimed to insert the BLIP-I and BLIP-II sequences, allowing the plasmid to serve as a carrier of BLIP in our donor bacterium. This construct would later enable us to test conjugation and observe whether BLIP can inhibit β-lactamase activity in recipient strains.

Step 1. Sequence optimization

To ensure reliable expression of our BLIP constructs in E. coli, we reverse-translated the protein sequences of BLIP-I and BLIP-II obtained from the NIH database and performed codon optimization specifically for E. coli expression. The resulting gene sequences were adjusted to satisfy the codon usage and sequence complexity requirements of GenScript and Twist Bioscience, while preserving the original amino acid sequences. Figure 1 presents the finalized nucleotide sequences submitted for gene synthesis following optimization.

Figure 1. Sequence optimization for gene synthesis

Step 2. PCR amplification + Gel purification

The BLIP gene fragments were successfully amplified by PCR, producing clear bands of the expected size on agarose gel electrophoresis (Figure 2). The amplified products were then excised and purified using the GeneDireX PCR Clean-Up and Gel Extraction Kit to remove residual primers, nucleotides, and enzymes that could interfere with downstream reactions. (Figure 3)

After purification, the resulting high-purity DNA was directly used for restriction digestion with EcoRI and PstI to prepare the inserts for ligation into the pSB1C3 backbone. This purification step ensured clean, intact DNA suitable for cloning and efficient enzymatic digestion.

Figure 2. Gel electrophoresis results of the PCR amplifications of BLIP-I and II

Figure 3. Purification of PCR products by excision and gel purification kit

Step 3. Transformation of DH5ɑ with PSB1C3-mCherry (distribution kit #O11)

To begin constructing our donor plasmid, we first transformed E. coli DH5α with the pSB1C3-mCherry backbone (distribution kit #O11). This backbone carried Chloramphenicol resistance and the mCherry reporter, which would later allow us to track and confirm our constructs visually. When we checked the transformation plates, only a single colony appeared.

Figure 4. Resuspended pSB1C3-mCherry backbone (distribution kit #O11)

Figure 5. Transformation result of DH5ɑ_PSB1C3_mCherry. The outgrowth culture was plated on a Ch-LB plate and cultured at 37 ℃ overnight.

Step 4. Subculturing the DH5ɑ_PSB1C3_mCherry

After we transformed DH5α with pSB1C3-mCherry, we were excited to see a colony appear, but also nervous, because it was the only colony on the plate. Losing it would mean starting over from scratch. To avoid this risk, we first streaked the colony onto a Ch-LB plate to expand our supply. We also inoculate it to a LB broth to allow it to grow without antibiotic pressure. Once it produced more colonies, we subcultured it onto both a Ch-LB plate and a Ch-LB plate . This way, we could maintain selective pressure while also ensuring we didn’t lose our only positive clone.

This careful step gave us enough material to continue with plasmid preparation and preserved our successful transformation for the next stages of plasmid construction.

Figure 6. Subculture result of DH5ɑ_PSB1C3_mCherry in Ch-LB plate

Step 5. Making glycerol stocks of DH5ɑ_PSB1C3_mCherry

Once we confirmed that our DH5α_pSB1C3_mCherry clone was stable, we prepared glycerol stocks to preserve it. Since this plasmid was the backbone for our donor construction, we wanted to make sure we would not lose the strain after future subculturing or experiments. By freezing glycerol stocks at –80 °C, we created a long-term, reliable source of our construct that we could return to at any time. This step gave us security and ensured that even if future transformations failed, we could always recover our starting material.

Figure 7. Glycerol stocks of DH5ɑ_PSB1C3_mCherry

Step 6. Miniprep of PSB1C3_mCherry

After successfully growing DH5α carrying the pSB1C3-mCherry backbone, we performed a plasmid miniprep to extract and purify the plasmid DNA. This step was essential because it provided us with a clean supply of the backbone needed for our ligation with BLIP-I and BLIP-II. Having purified DNA not only allowed us to move forward with restriction digestion but also gave us a reliable stock that could be used repeatedly without depending on fresh bacterial cultures each time.

Figure 8. Pellet of bacteria collected during the centrifugation step in the miniprep protocol

Step 7. Restriction digestion

With purified pSB1C3-mCherry plasmid DNA (from step 6) and BLIP genes (step 2) in hand, we carried out restriction digestion with EcoRI and PstI to prepare the backbone for ligation with BLIP-I and BLIP-II inserts. Our initial digests produced very faint or sometimes missing bands on the gel, and the DNA concentrations after purification were often too low for efficient cloning. (Figure 9)

Figure 9. Gel electrophoresis of restriction digestion products of purified DNA samples using the initial purification protocol. Weak DNA band intensity indicates low recovery efficiency when ddH₂O was used for elution from the silica column.

Before identifying the main cause of our low DNA recovery, we conducted several troubleshooting steps to improve yield during gel extraction and purification.

First, we verified the preparation of all reagents and discovered that ethanol had not been added to Buffer W2, a crucial component for DNA precipitation and binding to the silica column. After correcting this oversight, we observed only slight improvement in DNA retention. Second, we increased the DNA loading concentration for gel electrophoresis by combining multiple samples into a single wide well to visualize faint bands more clearly. Third, we attempted back-elution of the silica column to recover additional DNA that might have remained bound after the initial elution.

However, none of these adjustments significantly improved recovery. The DNA yield remained low, and gel bands were still faint. It was only then that we considered the potential effect of pH on DNA elution. We realized that ddH₂O gradually becomes slightly acidic over time (Figure 10), which compromises DNA stability and recovery efficiency.

Figure 10. The ddH₂O used for the elution was found to be too acidic. According to the manual, the optimal pH for the eluent should be around 7.0~8.5

By switching to Tris buffer for elution, we achieved clearer digested bands and substantially higher DNA yield. (Figure 11) This inspiring result marked an important engineering success in our workflow. The adjustment provided sufficient high-quality backbone DNA for the next steps of our cloning process. The gels containing the desired DNA fragments were carefully excised, extracted, and purified using the GeneDireX Gel Extraction Kit. The resulting purified DNA samples were then used directly in the subsequent ligation and transformation steps.

Figure 11. Gel electrophoresis of the restriction digestion results using purified BLIP-I & II sequences and pSB1C3 backbone (labelled as C) using modified protocol (Tris buffer elution).

Step 8. Ligation and transformation of DH5ɑ with PSB1C3_BLIP-I and PSB1C3_BLIP-II

After preparing both our BLIP inserts and the pSB1C3-mCherry backbone, we performed ligation to assemble the donor plasmids and transformed them into E. coli DH5α.

At first, the transformation did not yield colonies after standard overnight incubation. (Figure 12) When reviewing our reference protocol provided by NYCU Formosa, we realized that the amount of plasmid DNA we used during transformation was too low, 3 µL instead of the intended 10 µL, which likely reduced the efficiency of the transformation. Instead of discarding the attempt, we decided to extend incubation to 36 hours in hopes of recovering transformants.

This patience paid off. After the extended incubation, one colony began to appear on the Ch-LB plate spread with DH5α_pSB1C3_BLIP-I-mCherry. (Figure 13) The colony showed the distinct pink coloration from the mCherry reporter, which we affectionately referred to as our “Pinky Juniors.”

For DH5α_pSB1C3_BLIP-I-mCherry, no colony showed up on the Ch-LB plates even after prolonged incubation; however, the remaining outgrowth culture we added to the Ch-LB broth showed bacterial growth. (Figure 14)

The fluorescence confirmed that our ligations had succeeded and that DH5α was now carrying the BLIP constructs. This was a critical milestone: despite setbacks, we demonstrated that both BLIP-I and BLIP-II could be successfully cloned into our donor backbone and maintained in bacteria.

Figure 12. Transformation result of DH5α_pSB1C3_BLIP-I-mCherry (Left) and DH5α_pSB1C3_BLIP-II-mCherry (Right). No bacterial growth after spreading the outgrowth culture on Ch-LB plates.

Figure 13. Transformation result of DH5α_pSB1C3_BLIP-I-mCherry

Figure 14. Transformation result of DH5α_pSB1C3_BLIP-II-mCherry. Bacterial growth in Ch-supplemented LB broth.

Step 9. Subculturing of DH5ɑ_PSB1C3_BLIP-I and DH5ɑ_PSB1C3_BLIP-II

After the breakthrough of obtaining colonies carrying our BLIP constructs, our next priority was to secure and expand them. We carefully picked colonies of DH5α_pSB1C3_BLIP-I and DH5α_pSB1C3_BLIP-II and subcultured them into fresh Ch-LB broth and onto selective plates.

This step ensured that our precious transformants would not be lost and gave us enough material for downstream experiments. Watching the pink-fluorescent cultures grow stronger was a reassuring sign that our plasmids were stable and ready for further testing. By establishing these subcultures, we created reliable working stocks of both BLIP-I and BLIP-II donor plasmids, marking an important checkpoint in our construction process.

Figure 15. Subculture of DH5α_pSB1C3_BLIP-I-mCherry

Figure 16. Subculture of DH5α_pSB1C3_BLIP-II-mCherry

Step 10. Miniprep of pSB1C3_BLIP-I and pSB1C3_BLIP-II

With stable subcultures of DH5α carrying our BLIP constructs, we performed a plasmid miniprep to extract and purify the DNA. This step was crucial, as it provided us with a clean and concentrated supply of the pSB1C3_BLIP-I and pSB1C3_BLIP-II plasmids for further use.

The purified plasmids were intended for transformation into our donor strain (E. coli S17-1), where they would act as the vehicles for conjugation experiments. By completing this step, we successfully created ready-to-use BLIP plasmids that could be stored, shared, and tested in our conjugation system.

Figure 17 Miniprepped pSB1C3_BLIP-I and pSB1C3_BLIP-II plasmids

To test the effect of our BLIP donor plasmids, we needed a recipient plasmid carrying ampicillin resistance. This plasmid would serve as the counterpart in our conjugation system, allowing us to observe whether BLIP could inhibit β-lactamase and resensitize resistant bacteria.

Step 1. Resuspension of samples from the distribution kit.

To begin constructing our recipient plasmid, we first resuspended several samples from the iGEM distribution kit, including BBa_J435300, BBa_J43550, and BBa_J435320. These plasmids carried different versions of ampicillin resistance, and we prepared them so they could be used directly in our transformation experiments. By having these plasmids ready, we ensured that the next steps, transforming E. coli DH5α, would proceed without delays.

Interestingly, while the iGEM distribution kit resuspension is normally expected to turn red due to the cresol red dye, our samples appeared brownish immediately after resuspension. This unexpected color shift was unusual and suggested that the dye or plasmid preparation may have behaved differently from what we anticipated.

Figure 18. Resuspended Product of BBa_J435300 from Distribution Kit #G7, Plate #2

Figure 19. Resuspended Product of BBa_J43550 from Distribution Kit #O6, Plate #2

Figure 20. Resuspended Product of BBa_J435320 from Distribution Kit #O9, Plate #2

Step 2. Transformation of DH5ɑ with BBa_J435300 (Distribution Kit #G7, Plate #2)

We attempted to transform E. coli DH5α with BBa_J435300, the plasmid carrying ampicillin resistance, to create our recipient bacteria. This step was essential for building the counterpart in our conjugation system.

However, the transformation did not succeed. After reviewing the process, we suspected that the issue may have come from the resuspended distribution kit sample, which had been stored in the freezer for too long. Despite this setback, the attempt helped us identify potential sources of error and prepared us for our subsequent transformations.

Figure 21. Transformation of DH5ɑ with BBa_J435300

Step 3. Mistaken transformation of DH5ɑ with BBa_J43550 (Distribution Kit #O6, Plate #2)

The plasmid from O6 instead of O9. This plasmid, BBa_J43550, carried an ampicillin resistance backbone but was originally designed for use in yeast. Surprisingly, after transformation into E. coli DH5α, we observed growth on our plates. At first, this result left us puzzled, since we had not expected a yeast-related plasmid to support growth in a bacterial system.

Curious about this unexpected outcome, we looked up the plasmid's details online and discovered that while it indeed contained an ampicillin resistance gene, this alone could not explain why it persisted in E. coli. The most likely explanation is that BBa_J43550 also carries an E. coli-compatible origin of replication, which allowed it to replicate in DH5α. Combined with the AmpR gene, this enabled colonies to survive and expand.

This accidental transformation not only surprised us but also taught us the importance of focus and accuracy in every step of the lab process. What began as a small oversight became an opportunity to understand plasmid compatibility more deeply and reminded us that careful attention to detail is essential in synthetic biology.

Figure 22. Transformation of DH5ɑ with BBa_J43550

Step 4. Transformation of DH5ɑ with BBa_J435320 (The Correct Distribution Kit #O9, Plate #2)

We then performed the transformation using the correct recipient plasmid, BBa_J435320 (O9). Initially, we saw no colonies on the LB agar plate with ampicillin, which suggested that the transformation had failed. (Figure 23) However, when we checked the LB broth with ampicillin where the transformants had also been cultured, we observed visible growth. (Figure 24) This result indicated that the transformation had, in fact, worked, just at a very low efficiency that did not immediately appear on plates.

To further confirm the transformation, we plated the culture again onto fresh ampicillin plates, and this time, colonies successfully appeared. We then streaked these colonies for single isolates to prepare them for subculturing.

We also noted that the overall growth rate of our DH5α transformants was slower than expected. Based on repeated observations across different transformations, we suspected that this was due to the low quality of our DH5α competent cells, which had gone through stressful shipping and handling. This may have weakened the cells and reduced their transformation efficiency, leading to delayed or reduced colony formation.

Figure 23. Transformation of DH5ɑ with BBa_J435320 on Amp-LB plate

Figure 24. DH5α _BBa_J435320 grown in Amp-LB broth. The third and fourth tubes, inoculated with concentrated cells following heat-shock transformation, exhibit visible turbidity, indicating successful bacterial growth.

Miniprep of BBa_J435320 (Distribution Kit #O9, Plate #2)

After confirming growth of the O9 transformants, we performed a miniprep to extract plasmid DNA. This step was essential to obtain a clean and concentrated supply of the recipient plasmid, which would later be used in our conjugation experiments. The purified plasmid ensured that we could move forward without depending solely on live cultures.

Figure 25 Miniprepped BBa_J435320 plasmids

For the donor part of our conjugation system, we chose E. coli S17-1, a strain well-suited for plasmid transfer because of its built-in conjugation functions. Into this strain, we aimed to introduce our constructed plasmids PSB1C3_BLIP-I and PSB1C3_BLIP-II, enabling S17-1 to serve as the vehicle for delivering BLIP into recipient bacteria.

Unlike our familiar DH5α strain, working with S17-1 presented new challenges, especially since we did not have streptomycin available until mid-October and had to maintain the strain without antibiotics under strict aseptic conditions. This marked the beginning of our efforts to establish a functional donor strain for conjugation.

Step 1. Rehydration and initial growth of S17-1

We began by rehydrating the lyophilized E. coli S17-1 strain, carefully reviving it to prepare for downstream use as our donor bacteria.

At this stage, we faced an additional challenge: our lab did not yet have access to streptomycin, which is normally used to maintain S17-1 cultures.

As a result, we had to grow the strain without antibiotics, relying instead on strict aseptic conditions to prevent contamination. This step required extra caution, but it allowed us to successfully establish the initial growth of S17-1 and set the foundation for its role as our donor strain.

Figure 26. Initial growth of E. coli S17-1

Step 2. Making glycerol stock

To preserve our revived E. coli S17-1 strain for long-term use, we prepared glycerol stocks and stored them at –80 °C. This ensured that we could always return to a stable, viable culture for future experiments without the need to repeat the rehydration process. By securing these stocks early, we created a reliable backup that safeguarded our progress and gave us confidence to move forward with subsequent transformations.

Figure 27. Glycerol stocks of S17-1

Step 3. First attempt of Heat-shock transformation

This was our very first time attempting to transform E. coli S17-1. Without realizing that the cells were not chemically competent, we proceeded with the transformation as if they were ready. Unsurprisingly, the attempt failed, and no colonies were observed on our plates. Although it was disappointing at the moment, this mistake became an important learning experience for us. It taught us the critical role of preparing competent cells before transformation and reminded us that each step in the workflow must be carefully checked before moving forward.

Figure 28. Heat-shock transformation of S17-1 with pSB1C3_BLIP-1_mCherry (1st attempt)

Step 4. Second attempt of heat-shock transformation

During our first attempt at preparing chemical competent cells from E. coli S17-1, we struggled to obtain a sufficient pellet after centrifugation. This was likely due to too short of an incubation time, which resulted in very low cell density. The pellet we collected was extremely small, and as a result, the cells might not be successfully made competent. In the end, we still decided to store these as glycerol stocks.

Figure 29. The extremely small pellet obtained during our first attempt at preparing competent E. coli S17-1 cells

In our second attempt of preparing chemical competent cells, we extended the growth time and managed to obtain a good amount of pellet, indicating healthier biomass. However, instead of using the cells immediately, we stored both attempts as glycerol stocks at –80 °C.

We then used the competent E. coli S17-1 that we had prepared earlier and stored as glycerol stock at –80 °C to perform a heat shock, but it failed to produce any colonies. We suspected that this was due to the S17-1 competent cells losing competency after being frozen for too long. S17-1 appears to be much more sensitive, and using freshly prepared competent cells would likely have increased our success rate.

Figure 30. Heat-shock transformation of pSB1C3-BLIP-I into E. coli S17-1

Figure 31. Heat-shock transformation of pSB1C3_BLIP-II into E. coli S17-1

Step 5. Third attempt of heat-shock transformation

In our third attempt, we prepared E. coli S17-1 competent cells and, instead of storing them, we used them fresh for transformation. This change proved crucial, the freshly prepared competent cells successfully took up the BLIP plasmid, confirming that S17-1 could be transformed once handled under the right conditions. This outcome demonstrated the importance of using fresh competent cells with S17-1, as storage in glycerol stocks appeared to compromise transformation efficiency. By using this approach, we ensured that future heat-shock transformations would not fail due to the use of non-fresh S17-1 cells.

For the plasmid source, we used the plasmids (pSB1C3_BLIP-I and pSB1C3_BLIP-II) that had been successfully miniprepped from our DH5α transformants. This DH5α backup was created as mentioned earlier in the “Construction of Plasmid 1” section. By introducing these plasmids into fresh competent S17-1, the transformation was successful, establishing S17-1 as a donor strain for our conjugation experiments.

Figure 32. Transformation results of S17-1_pSB1C3_BLIP-I-mCherry

Figure 33. Transformation results of S17-1_pSB1C3_BLIP-II-mCherry

Step 1. Subculturing of DH5ɑ_BBa_J435320 after the transformation

Figure 34. Subculture results of DH5ɑ_BBa_J435320

Step 1. Solid Conjugation

We For solid conjugation, we co-cultured the donor strain (E. coli S17-1 carrying pSB1C3-BLIP) and the recipient strain (E. coli DH5α carrying BBa_J435320) on LB agar plates and LB agar supplemented with X-gal. The plates were only supplemented with X-gal as we could not get IPTG to our lab in time. After incubation at 37 °C for six hours, the plates exhibited visible pink and blue coloration, corresponding to the characteristic pigmentation of the donor (mCherry expression) and the recipient (lacZ expression), respectively. The presence of both colors indicated that both strains were actively growing and in close contact, a necessary condition for conjugation to occur on solid media.began by rehydrating the lyophilized E. coli S17-1 strain, carefully reviving it to prepare for downstream use as our donor bacteria.

Figure 35. Solid conjugation result after 6 hours incubation at 37 °C

Figure 36. Solid conjugation result (X-gal added) after 6 hours incubation at 37 °C

Step 2. Validation and Final Results

After several days of preparation, we finally obtained our conjugation results. They worked perfectly and fully supported our hypotheses. Despite our extremely tight schedule (with the wiki freeze just hours away), we were able to capture the key experimental outcomes that confirm our system’s functionality. To provide context for these final results, we first outline the potential outcome patterns we anticipated, followed by our actual observed results.

Following incubation for conjugation, we gently scraped the mixed bacterial growth from the plate surfaces and resuspended the cells in 1 mL of sterile PBS. From this suspension, 100 µL aliquots were plated onto chloramphenicol (Ch), ampicillin (Amp), and dual-antibiotic (Ch + Amp) LB plates, each supplemented with X-gal, and incubated overnight for selection and validation.

To evaluate both conjugation success and BLIP functionality, we combined antibiotic selection with blue-white screening.

We carefully considered possible plate outcomes to interpret our conjugation data effectively:

- Pink and blue colonies on Ch plates; no growth on Ch + Amp plates — ideal outcome: Complete proof of concept. This pattern would confirm that conjugation successfully transferred the BLIP plasmid from E. coli S17-1 to E. coli DH5α and that BLIP effectively inhibited β-lactamase activity.
- Pink colonies on Ch plates; blue colonies on Amp plates; no growth on Ch + Amp plates — partial or failed conjugation:Indicates no plasmid transfer between donor and recipient. The donor and recipient populations remain separate, and BLIP's inhibitory effect cannot be verified.
- Pink and blue colonies on Ch plates; growth on Ch + Amp plates — successful conjugation, inactive BLIP: Suggests plasmid transfer occurred, but BLIP failed to inhibit β-lactamase activity in the recipient strain.

We carefully considered possible plate outcomes to interpret our conjugation data effectively:

Figure 37. Conjugation results between E. coli S17-1_pSB1C3-BLIP-I-mCherry and E. coli DH5α_BBa_J435320.
Left: Pink and blue colonies on the Ch-LB plate indicate donor and transconjugant cells.
Middle: Blue colonies on the Amp-LB plate show recipient and transconjugant strains carrying lacZ.
Right: Only one pink colony on the dual Ch-Amp-LB plate demonstrates effective BLIP-mediated inhibition of β-lactamase activity.

Figure 38. Conjugation results between E. coli S17-1_pSB1C3-BLIP-II-mCherry and E. coli DH5α_BBa_J435320.
Left: Pink and blue colonies on the Ch-LB plate indicate donor and transconjugant cells.
Middle: Blue colonies on the Amp-LB plate show recipient and transconjugant strains carrying lacZ.
Right: No colonies on the dual Ch-Amp-LB plate demonstrate complete BLIP-mediated inhibition of β-lactamase activity.

At the time of the wiki freeze, our final conjugation plates were still incubating overnight, and we had very limited time to conduct additional verification steps. Ideally, we would have miniprepped the transconjugants and performed targeted PCR or restriction digestion to confirm the plasmid sequences. However, due to the rapidly approaching deadline, we were unable to carry out these follow-up validations.

Nevertheless, the observable phenotypes on our selection plates already provide clear functional evidence supporting successful conjugation and BLIP activity, fulfilling our Proof of Concept for the PlasMission Spy system.

We began with PCR amplification of BLIP-I and BLIP-II, but the gel electrophoresis showed heavy smearing and low band intensities. (Figure 39)The issue was later traced to a primer ordering mistake. As we ran out of VR and VF2 primers, we ordered a new batch; however, instead of the primer sequences (VR and VF2), the primer binding site sequences were mistakenly ordered. In addition, the restriction digestion of BBa_J435320 produced no visible bands, indicating that the miniprepped plasmid likely contained little or no DNA, possibly due to poor miniprep practice.
By the time these issues were identified, we had successfully constructed the donor strain S17-1_pSB1C3_BLIP-1-mCherry for conjugation, making this direct cloning approach unnecessary. The plan was therefore discontinued.

Figure 39. Gel electrophoresis result of the PCR products with erroneous VR and VF2 primers

This project has been a truly memorable experience for every single member of our wet lab team. Each of us feels deeply grateful to have had the opportunity to participate in this year’s iGEM competition, to learn, to experiment, and to grow together. Even though our journey was far from perfect—filled with mistakes, unexpected outcomes, and results that did not always align with our expectations, it was these very challenges that made the experience meaningful.

What we gained was more than just data or plasmids, it was the resilience to keep trying, the courage to embrace failure, and the joy of discovery when things finally worked. Every setback taught us something new, and every small success felt like a victory shared by the whole team.

While we may not have achieved every goal we set out to accomplish, we are proud of the progress we made and the lessons we will carry forward. This project reminded us that science is not only about perfect results, but also about the journey of persistence, teamwork, and creativity. Above all, it showed us the power of collaboration and the excitement of being part of something larger than ourselves.

As we close this chapter, we do so with gratitude, for the guidance we received, for the friendships we strengthened, and for the chance to contribute, in our own way, to addressing the urgent challenge of antimicrobial resistance.

This is not the end, but the beginning of our journey as young scientists.