Search

Collective intelligence is an important component of the new generation of artificial intelligence (AI). It plays a decisive role in stimulating and converging innovative forces as well as coupling and integrating large-scale intelligent systems. It is of great significance for promoting deep integration of AI and traditional industries and enabling the sustainable development of the national economy. This study summarizes the overall technical framework of collective intelligence and its major research areas, including:multi-agent systems and optimal decision-making, unmanned swarm systems, open source collective intelligence software, and federated learning. Moreover, it analyzes how these core technologies can be applied in industrial scenarios, in order to establish intelligent processing loops of perception‒cognition‒decision‒action, to support platform economy with distributed intelligence, and to reshape industrial development and digital economy ecosystems. Based on the subjects and application modes of the technical framework, this study analyzes the core industries related to collective intelligence, particularly the software service industry, the smart city industrial cluster, and the intelligent agriculture and port industries based on unmanned swarm systems, by highlighting their significant requirements and empowerment approaches for collective intelligence technologies. Furthermore, this study presents suggestions on how to utilize collective intelligence technologies to foster development of rated industries. It is suggested that we should continuously promote the establishment of open source communities of collective intelligence, enhance the intellectual core of the AI technological innovation ecosystem, and accelerate the domestic substitute of unmanned swarm systems through integrated system research.

The effect of methanol addition on the heterologous expression of isoprenyl transferase NovQ was studied in Pichia pastoris Gpn12, with menadione and isopentenol as precursors to catalyze vitamin K2 (MK-3) synthesis. The expression of NovQ increased by 36% when 2% methanol was added every 24 h. The influence of initial pH, temperature, methanol addition, precursors (menadione, isopentenol) addition, catalytic time and cetyltrimethyl-ammonium bromide (CTAB) addition were explored in the P. pastoris whole-cell catalytic synthesis process of MK-3 in shaking flask. Three significant factors were then studied by response surface method. The optimal catalytic conditions obtained were as follows: catalytic temperature 31.56 ℃, menadione 295.54 mg/L, catalytic time 15.87 h. Consistent with the response surface prediction results, the optimized yield of MK-3 reached 98.47 mg/L in shaking flask, 35% higher than that of the control group. On this basis, the production in a 30-L fermenter reached 189.67 mg/L when the cell catalyst of 220 g/L (dry weight) was used to catalyze the synthesis for 24 h. This method laid the foundation for the large-scale production of MK-3 by P. pastoris Gpn12.

Bacillus subtilis JA produces a broad spectrum of bioactive lipopeptides with great potential applications in agriculture and industry. Crude lipopeptides were extracted with 100% methanol from the precipitate which was obtained by adding 6 mol/L HCl to the cell-free culture broth. The crude extract was separated on a Diamonsil C18 column (5 microm, 250 mm x 4.6 mm) in reversed-phase high performance liquid chromatography (HPLC) system to separate the lipopeptide homologues. Five peaks were eluted from HPLC. Electrospray ionization mass spectrometry (ESI-MS) was used to analyze each HPLC fraction. The results showed three series of ion peaks. According to the values of m/z, the three series of ions were classified into surfactin, iturin and fengycin homologues, which were well-known biosurfactants produced by B. subtilis strains. The major ions were structurally characterized using tandem mass spectrometry. This study suggested a reasonable method for the isolation and identification of lipopeptides produced by B. subtilis strains.

Lipopeptides produced by Bacillus subtilis JA antagonized a broad spectrum of plant fungal pathogens. The purification and identification of the lipopeptide antibiotics plays an important role for further research. Crude lipopeptides were extracted with methanol from the precipitate, which was obtained by adding 6mol/L HCl to the cell-free culture broth and then stored at 4 degrees C overnight. The crude extract was run on reversed-phase HPLC system with a Diamonsil C18 column (250 mm x 4.6 mm, Dikma) to separate the lipopeptides. Two antifungal compounds, which had strong inhibitory activity against various plant fungal pathogens, such as Fusarium graminearum, were purified. The molecular weights of two compounds were determined by electrospray ionization mass spectrometry (ESI/MS). Two compounds, with molecular weights of 1042.4 Da and 1056.5 Da, were homologues differed by a structure of -CH2. ESI collision induced dissociation mass spectrometry analysis was used to sequence the structure of purified compounds. Typical b- and y- type fragments showed that compound 1 (with a molecular weight of 1042.4 Da) had a primary structure of Pro-Asn-Tyr-PAA-Asn-Tyr-Asn-Gln (PAA represented beta-amino acid), which was consistent with lipopeptide iturin A. Compound 2 was a homologue of iturin A.