This year, our team has made multifaceted contributions aimed at supporting and inspiring future iGEM teams through reliable biological parts and innovative hardware tools. Guided by iGEM’s spirit of creativity and generosity, we have developed and rigorously characterized novel genetic constructs for lipid droplet targeting in Saccharomyces cerevisiae, created an open-source smart monitoring system for real-time process tracking in carotenoid extraction and beyond. Each effort is thoroughly documented and designed to provide reusable, scalable, and practical resources that lower barriers and accelerate research for the global iGEM community.

1. Contribution to the existing parts

yeGFP (BBa_K4866003)

yeGFP has consistently proven to be a reliable reporter in Saccharomyces cerevisiae. This year, we used yeGFP to screen signal peptides for lipid droplet targeting and to evaluate whether fusion at the N- or C-terminus of HD2 would be more effective. Colocalization results from yeGFP and Nile red confirmed that HD2 is the most efficient lipid droplet-targeting signal peptide, with C-terminal fusion yielding the best performance.

We also continued using the commonly adopted promoters (pTDH3, BBa_K2637011) and terminators (tADH1, BBa_K4121040 and tCYC1, BBa_K5370003) for S. cerevisiae. Although these regulatory elements are well-established, our work this year adds another layer of evidence for their functionality. While their individual contributions can only be fully assessed through composite part characterization e.g. POX1-PTEF1-CarB-T ADH1+ PTDH3-CarRP-T CYC1-POX1 (BBa_25JCA3OI). The characterization of these promotors and terminators was detailed in the new composite part.

2. Contribution to new parts in the Registry

This year we have 8 new basic parts and 11 new composite parts. All the new parts contributions are listed below:

For all the new parts contributions, we will only provide examples of basic and composite parts here. The full list is shown in the table above.

For the basic parts, we have specifically tested four different signal peptides for lipid droplet targeting: PTDH3-yeGFP- HD2-TCYC1, PTDH3-yeGFP- HD1-TCYC1 (BBa_258VK0BD), PTDH3-yeGFP- HD3-TCYC1 (BBa_254ROFU6), PTDH3-yeGFP- HD4-TCYC1 (BBa_25BWBIEW). It is a good reference for future studies which is interesting in lipid droplets targeting.

For the composite parts, we are so proud of the developing and optimizing of the lipid droplets targeting toolkit BBa_25SXFI1M: POX1-PTDH3-CarRP-HD2-TCYC1 + PTEF1-CarB-linker(G4S)2-HD2-TADH1-POX1.**

Here is how we did that:

To identify the best lipid droplet-targeting signal peptide, we screened four different candidates. Since we used a linker to fuse the proteins with the signal peptides, we also optimized the positioning of the linker. Through experimentation, we determined that only fusion at the C-terminus of the protein yielded effective lipid droplet targeting.

We further optimized the configurations for different proteins, resulting in CarRP without a linker and CarB with a linker. Finally, we docked the substrate and enzyme (CarB and CarRP) to assess their spatial arrangement and calculated the Kcat values to evaluate the feasibility of protein fusion.

3.Contribution to Experiment: Smart Monitoring System

We have developed an image-based smart monitoring system for carotene extraction, consisting of a hardware timed photography device and a software intelligent reasoning module.

Beyond carotene extraction, this system has broad applicability in other biological experiment scenarios—such as microbial fermentation (for antibiotics or enzyme production) and biochemical product purification—where real-time, non-invasive monitoring of color/morphological changes and process parameters is needed.

The iGEM ambassador Xiaohan Zhang, was impressed with our image-based carotene detector, praising its practical design. After confirming that our method uses color data from standard photos rather than complex infrared spectroscopy, he emphasized its advantages: the process is simple, affordable, and has promising viability for broad real-world implementation.

For the hardware part, the core is a user-friendly timed photography device equipped with a high-definition USB camera, a Raspberry Pi controller, and a 4G wireless router. It can automatically capture images every 30 minutes, store data locally, and sync images to Cloud OBS in real time—no complex wiring is required, and it supports 24-hour non-intrusive monitoring. This device effectively solves the pain points of traditional carotene extraction processes, such as lagging manual sampling, error-prone offline detection, and inability to track key process nodes in real time.

For the software part, we are proud of its three integrated functional modules: the HSV Change Analyzer, the Production Model Analyzer, and the Combined Inference module. Here is how we built and validated it:

First, we trained a device runtime model by analyzing the temporal changes of HSV color features (hue, saturation, brightness) in collected images using 4th-degree polynomial regression, establishing an accurate mapping between visual data and extraction time. Second, we constructed a yield and solubility prediction model by applying polynomial regression to historical runtime-yield-solubility datasets, enabling quantification of extraction efficiency. Finally, during testing, the software deduced real-time runtime from new image HSV values and predicted yield/solubility with high accuracy (R² close to 1), forming a closed-loop of “image capture - feature analysis - parameter inference”.

More detailed information about our monitoring system could be read in our hardware wiki page.