SCUT-China-L team was established and held the first offline in-person meeting.

Based on the experience from the team's field investigation to the red algae factory and the advisor's suggestions, the project direction was finalized: remodeling the S. cerevisiae to produce rare ginsenoside Rh1 using red algae.

We refined the project details through literature review, formulated the experimental design, selected the cosmetics village after analyzing previous iGEM projects and assigned tasks among team members.

By using FBA algorithm, we wrote code for calculations to verify project feasibility and optimal experimental strategy selection.

Wet lab members underwent lab safety training, learned experimental techniques for primer design, plasmid construction, strain culture, and the use of tools like CHOPCHOP, SnapGene, and NCBI website searches. Determined the use of wild-type S.cerevisiae CEN.PK2-1D for engineering based on literature review and existing lab research.

Dry lab members further analyzed the FBA algorithm and decided to develop related visualization software.

Through literature review and database mining, we screened three agarases from different sources and two neoagarobiose hydrolases from different sources.

We explored platform tools related to protein structure prediction and molecular docking prediction. Based on the five amino acid sequences provided by wet lab members, conducted initial modeling tests, selecting AF2 for structure prediction. Subsequently, we explored the PDB website, used Avogadro and UCSF ChimeraX for computational operations, and learned to use tools like WGLTools, AutoDock Vina, Meeko for molecular docking tests to provide predictive results for wet lab experiments.

Using Gibson Assembly, we successfully constructed six plasmids. After screening and sequencing verification, the plasmids were transformed into yeast cells using the lithium acetate method. Verified and measured the enzyme activities of agarase and neoagarobiose hydrolase, screening the engineered strain Sq-Ag5 for efficient degradation of red algae polysaccharides.

We continued learning from existing model predictions. While receiving verification from wet lab experiments, we conducted further exploratory analysis on other aspects of the predicted enzyme structures.

To increase squalene production, we designed an experiment to optimize the MVA pathway by overexpressing the tHMG1 and IDI1 genes.

We collaborated with the design team members to discuss the visual aesthetics of the software interface, including theme colors and interactive animations.

We learned experimental operations for CRISPR technology. Selected the genomic GAL80 target site and constructed the Cas9-sgRNA plasmid, simultaneously constructing DNA integration fragments for the tHMG1 and IDI1 genes. After yeast gene editing, primary screening, negative screening, and PCR verification, the obtained strain Sq-0 was used for squalene fermentation experiments. Results confirmed the successful squalene production of strain Sq-0, which significantly outperformed the unmodified strain.

We recruited new dry lab members to assist with modeling for a new direction: calculating the economic and environmental benefits of the ginsenoside manufacturing route.

Over these two weeks, we integrated the secretory hydrolase module and the MVA modification module, constructing strain Sq-Ag capable of hydrolyzing agar polysaccharides and producing squalene. Simultaneously optimized culture conditions, red algae liquefaction conditions and the optimal of mixed carbon sources.

In software development, through communication and demos with wet lab members, including internal downloads and usage, we obtained user trial feedback. Then we added an AI agent function to enrich software operations. Meanwhile, we began establishing the synthesis pathway for rare ginsenoside Rh1 in the model. We introduced the rare ginsenoside Rh1 synthesis genes into yeast cells model by constructing high and low copy plasmids.

After introducing the ginsenoside synthesis genes into yeast cells, neither Rh1 nor any intermediate products were detected upon fermentation. Consequently, we planned to redesign the experiment by integrating the genes into the yeast genome and repeating the experiment.

We further enhanced the yield of the precursor squalene. We screened 3 constitutive promoters and 3 inducible promoters for combination, constructing strains Sq-Ag7 to Sq-Ag23, and compared squalene yields.

Simultaneously, we began redesigning the experimental plan, searching for suitable gene integration sites. Using strain Sq-0 as the starting point, we introduced the five genes in batches, conducted experiments, and detected the yields of corresponding products

We recruited new dry lab members to assist in developing the AI agent, conducted online meetings with wet lab members to re-clarify user needs and encountered problems, built a relevant knowledge base, and clarified each member's work.

We analyized promoter characteristics and built a prediction model for optimal promoter combinations.

Following suggestions from Miss Xie and Miss Wang, we further implemented a chimeric intron, by inserting the RPS25Ai intron near the promoter, which further increased squalene yield.

We also conducted experiments and finally successfully detected the yields of the corresponding products.

We used purchased red algae to prepare YPDA medium and cultured Sq-Ag23c, confirming the establishment of the red algae utilization pathway. Meanwhile, constructed strain S.cerevisiae Rh1-Ag for de novo synthesis of Rh1 using red algae polysaccharides as substrate. Verified that introducing the red algae hydrolase genes enabled effective Rh1 production from red algae, and also confirmed that the MVA pathway modifications in yeast cells remained effective after completing the metabolic pathway construction.

We curated data, generated the figures, and wrote the Wiki, while also designed the content for the presentation video.