Synthetic biology inAgriculture, Chemical, Food, Energy, Medical, Materialsand other fieldsAll have important applications.

For example, in the field of medical drugs, synthetic biology has important applications in vaccine development, molecular diagnostics, cell-based therapies, and the development of new drugs. The advent of low-cost nucleic acid synthesis has enabled synthetic biologists to redesign entire viral genomes using large-scale synonymous mutations. Synthetic biologists purposefully use underrepresented codons to reduce the production of viral proteins in human cells, thereby quickly and reliably producing attenuated viral proteins without detailed knowledge of viral function. The technology has been used in attenuated vaccines such as polio virus, influenza virus, respiratory syncytial virus and dengue virus.

Cell-free Protein Synthesis (CFPS) system is an in vitro gene expression system that uses exogenous DNA or mRNA as a template, artificially adds the required raw materials and energy substances, and uses cell extracts as a condition to synthesize proteins, which can break through cell limitations and express various proteins conveniently and quickly. The first step of CFPS is to lyse the biological cells, extract their organelles, and then mix the energy, raw materials and other substances needed for the synthesis reaction to form a reaction solution. When exogenous DNA or mRNA is added to the reaction solution as a template, a protein expression reaction proceeds in the system.Compared with the traditional protein expression method, CFPS has the advantages of higher product and energy conversion efficiency, more abundant expressible protein, short reaction time, convenient regulation and high synthesis efficiency.

In synthetic biology, CFPS is used inMetabolic engineering, biosensors, genetic circuits, artificial cells and AI-assisted high-throughput protein drug screeningand other fields have application prospects.

With the development of metabolic engineering and synthetic biology, cell-free metabolic engineering (CFME) has been used to produce biomaterials, biofuels and drug precursors. Initially, CFME was performed by assembling metabolic pathways using purified enzymes. However, in vivo protein expression and purification of each pathway protein is laborious and time consuming. The introduction of cell-free protein expression can accelerate the DBT(design-build-test) cycle of metabolic engineering.

Wu and his research team used CFPS for rapid protocol design of 1,4-butanediol (BDO) biosynthesis^[1]. It expresses various metabolic enzymes by means of CFPS, which not only verifies the function of metabolic pathways and various pathway enzymes, but also determines the rate-limiting step of the BDO pathway (4-hydroxybutyrate is converted to downstream metabolites). The team significantly increased the production of BDO by regulating the expression levels of various enzymes and increasing the expression levels of downstream enzymes. This demonstrates that CFPS can serve as a support platform for metabolic engineering and synthetic biology applications. The exploration of BDO metabolic pathway proved that CFPS can rapidly regulate the expression level of pathway enzymes, and screen for enzyme variants with improved catalytic activity to improve product yield. This result demonstrates that CFPS-ME can be directly applied to the development of strains in vivo.

Cell-free biosensors designed by synthetic biology are a promising new tool for the detection of clinically relevant biomarkers.

Wen and his team designed a biosensor based on an E. coli cell-free system, in which a sensing circuit using the acylhomoserine lactone of Pseudomonas aeruginosa was implemented to analyze the effect of sputum samples from cystic fibrosis patients.^[2]The results of this study show that by optimizing the cell-free system and sample extraction method, the molecular 3-oxo-C12-HSL in cystic fibrosis lung sputum samples can be quantitatively measured at the nanomolar level. This study further illustrates the potential of the modular cell-free biosensor as a rapid, low-cost detection method, and also shows that it has the potential to become a platform for detecting and monitoring pathogen biomarkers in human respiratory samples.

At the end of 22 years and the beginning of 23 years, ChatGPT broke the circle!

AI is leading a new revolution in the tech world. In the traditional research and development process, scholars start from the basic knowledge of the subject, through the mutual verification and falsification of assumptions and experiments, put forward new subject knowledge and be recorded and learned by later generations. This process promotes the advancement of subject knowledge. The emergence of AI will greatly accelerate this process. After the new AI is built manually, the researcher will input the experimental conditions to the AI and inform it of the corresponding experimental results. Through large-scale data training, the AI will try to find out the connection between the experimental conditions and the results. Through this connection, AI can independently design experimental conditions and predict the desired experimental results. The experimenter performs the experimental conditions designed by AI and feeds back the experimental results to AI, so that AI corrects and optimizes its own judgment program to achieve the effect of AI guiding the experiment.

In the early 20 s, Google announced that it had developed a AlphaFold that enabled AI to predict protein structural folding, and it made a big difference in the protein prediction competition with scientists. Extremely accelerated the further exploration of protein structure and function. This result reveals the great research potential of AI in the field of scientific research.

In January of this year,A team of researchers from Salesforce Research, Tierra Biosciences, and the University of California synthesized the proteins predicted by the AI model in the lab and found that they were as effective as their natural counterparts.^[3]该研究以「Large language models generate functional protein sequences across diverse families」为题，于2023年1月26日发布在《Nature Biotechnology"on.

His research team was inspired by the success of deep learning-based natural language models to develop ProGen, a protein-based language model. By having the AIProGen deep learn from the training of 0.28 billion original protein sequences from the public sequenced natural protein database, the ability to autonomously design protein sequences to accomplish a desired functional effect is mastered.

The research team let ProGen independently design the sequence of lysozyme, encode the plasmid according to its designed sequence, and use cell-free protein synthesis technology to synthesize hundreds of proteins in vitro at high throughput. The researchers tested their activity and compared their activity to that of an enzyme found in chicken egg white (Hen Egg White Lysozyme,HEWL) during activity verification. The researchers found that two of the artificial enzymes were able to break down bacterial cell walls with activity comparable to HEWL, but their sequences were only about 18 percent identical to each other. These two sequences are approximately 90% and 70% identical to any known protein.

The study is an excellent inspiration for AI-assisted high-throughput protein drug screening practices. Under the brilliance of AI, the importance of cell-free protein synthesis technology cannot be ignored.

High throughput is an important factor in AI-assisted high-throughput protein drug screening. AI training requires a large amount of data, which also means that a large number of experimental results are required to synthesize proteins from sequences. In the traditional method of differential amplification after transfection of plasmids into E. coli, the expression cycle of a protein is generally 4-7 days, which is a great consumption of time and labor cost. The cell-free protein synthesis technology is selected for protein expression, which can be reacted in a 96-well plate, and the reaction time is 2-8h. Greatly reduce the consumption of time and labor costs. In a 96-well plate, 96 reactions can be carried out simultaneously, that is, 96 proteins can be expressed at the same time, which is obvious to improve the experimental throughput. Therefore, it is a wise choice to use cell-free protein synthesis technology to screen AI-assisted macromolecular protein drugs.