Although we have preliminarily demonstrated that the formate assimilation pathway can function in TD80, its effectiveness is not yet significant enough. Therefore, further optimization will be pursued in the future, which may include providing additional reducing power and ATP, as well as incorporating key microbial components alongside other C1 compounds as carbon sources.
Additionally, given the high proportion of acetate in CDE and the low proportion of formate, a crucial challenge ahead is to optimize Halomonas TD's utilization of formate within CDE, ensuring that formate is used as efficiently as possible alongside acetate.

Successfully building a thermosensitive bio-switch available to the Halomonas TD is exciting, however, we still have a long way to go. After that, we plan to mutate the PR promoter of sfgfp to further broaden the scope of application of our T-switch. >At the same time, we plan to build a reversible thermal regulation, a new version of T-switch. It's a challenging direction, but we're looking forward to embarking on this new journey.

Although we successfully proved that the modified strain Halo6.0 could produce three kinds of products (PHAs, amino acid derivative and protein), we want to further broaden our product catalog in order to further demonstrate the production potential of Halomonas TD and show the world a real "Haloworld!"

Although our Locus screening program has proven effective in identifying high-expression sites, its input is limited to one-dimensional nucleotide sequences, missing some spatial characteristics. We aim to consider additional influencing factors, such as the three-dimensional structure of DNA (major and minor grooves, supercoiling, etc.), to enhance the precision of our screening process.This will be highly challenging, but we believe it is meaningful.

Successfully establishing and operating the GO-term-based ec-GEM is not our main focus. While linear model calculations are rapid, this approach can reduce accuracy. The impact of decreased model precision is not significant for small-scale fermentations but can be critical for large-scale fermentations. Moving forward, we will optimize the classification method of GO terms and explore other clustering approaches. In line with this, the current feeding flow control scheme is still in its preliminary stage; we will attempt to use fuzzy PID control to address the nonlinearities and uncertainties of real cellular states.