Multi-strategy modified sparrow search algorithm for hyperparameter optimization in arbitrage prediction models.

Deep learning models struggle to effectively capture data features and make accurate predictions because of the strong non-linear characteristics of arbitrage data. Therefore, to fully exploit the model performance, researchers have focused on network structure and hyperparameter selection using various swarm intelligence algorithms for optimization. Sparrow Search Algorithm (SSA), a classic heuristic method that simulates the sparrows' foraging and anti-predatory behavior, has demonstrated excellent performance in various optimization problems. Hence, in this study, the Multi-Strategy Modified Sparrow Search Algorithm (MSMSSA) is applied to the Long Short-Term Memory (LSTM) network to construct an arbitrage spread prediction model (MSMSSA-LSTM). In the modified algorithm, the good point set theory, the proportion-adaptive strategy, and the improved location update method are introduced to further enhance the spatial exploration capability of the sparrow. The proposed model was evaluated using the real spread data of rebar and hot coil futures in the Chinese futures market. The obtained results showed that the mean absolute percentage error, root mean square error, and mean absolute error of the proposed model had decreased by a maximum of 58.5%, 65.2%, and 67.6% compared to several classical models. The model has high accuracy in predicting arbitrage spreads, which can provide some reference for investors.

Emotion Recognition from Multiband EEG Signals Using CapsNet.

Emotion recognition based on multi-channel electroencephalograph (EEG) signals is becoming increasingly attractive. However, the conventional methods ignore the spatial characteristics of EEG signals, which also contain salient information related to emotion states. In this paper, a deep learning framework based on a multiband feature matrix (MFM) and a capsule network (CapsNet) is proposed. In the framework, the frequency domain, spatial characteristics, and frequency band characteristics of the multi-channel EEG signals are combined to construct the MFM. Then, the CapsNet model is introduced to recognize emotion states according to the input MFM. Experiments conducted on the dataset for emotion analysis using EEG, physiological, and video signals (DEAP) indicate that the proposed method outperforms most of the common models. The experimental results demonstrate that the three characteristics contained in the MFM were complementary and the capsule network was more suitable for mining and utilizing the three correlation characteristics.

Colonization of Morus alba L. by the plant-growth-promoting and antagonistic bacterium Burkholderia cepacia strain Lu10-1.

BACKGROUND: Anthracnose, caused by Colletotrichum dematium, is a serious threat to the production and quality of mulberry leaves in susceptible varieties. Control of the disease has been a major problem in mulberry cultivation. Some strains of Burkholderia cepacia were reported to be useful antagonists of plant pests and could increase the yields of several crop plants. Although B. cepacia Lu10-1 is an endophytic bacterium obtained from mulberry leaves, it has not been deployed to control C. dematium infection in mulberry nor its colonization patterns in mulberry have been studied using GFP reporter or other reporters. The present study sought to evaluate the antifungal and plant-growth-promoting properties of strain Lu10-1, to clarify its specific localization within a mulberry plant, and to better understand its potential as a biocontrol and growth-promoting agent. RESULTS: Lu10-1 inhibited conidial germination and mycelial growth of C. dematium in vitro; when applied on leaves or to the soil, Lu10-1 also inhibited the development of anthracnose in a greenhouse, but the effectiveness varied with the length of the interval between the strain treatment and inoculation with the pathogen. Strain Lu10-1 could survive in both sterile and non-sterile soils for more than 60 days. The strain produced auxins, contributed to P solubilization and nitrogenase activity, and significantly promoted the growth of mulberry seedlings. The bacteria infected mulberry seedlings through cracks formed at junctions of lateral roots with the main root and in the zone of differentiation and elongation, and the cells were able to multiply and spread, mainly to the intercellular spaces of different tissues. The growth in all the tissues was around 1-5 × 105 CFU per gram of fresh plant tissue. CONCLUSIONS: Burkholderia cepacia strain Lu10-1 is an endophyte that can multiply and spread in mulberry seedlings rapidly and efficiently. The strain is antagonistic to C. dematium and acts as an efficient plant-growth-promoting agent on mulberry seedlings and is therefore a promising candidate as a biocontrol and growth-promoting agent.

Shotgun proteomic analysis of mulberry dwarf phytoplasma.

BACKGROUND: Mulberry dwarf (MD), which is caused by phytoplasma, is one of the most serious infectious diseases of mulberry. Phytoplasmas have been associated with diseases in several hundred plant species. The inability to culture phytoplasmas in vitro has hindered their characterization at the molecular level. Though the complete genomes of two phytoplasmas have been published, little information has been obtained about the proteome of phytoplasma. Therefore, the proteomic information of phytoplasmas would be useful to elucidate the functional mechanisms of phytoplasma in many biological processes. RESULTS: MD phytoplasmas, which belong to the 16SrI-B subgroup based on the 16S DNA analysis, were purified from infected tissues using a combination of differential centrifugation and density gradient centrifugation. The expressed proteome of phytoplasma was surveyed by one-dimensional SDS-PAGE and nanocapillary liquid chromatography-tandem mass spectrometry. A total of 209 phytoplasma proteins were unambiguously assigned, including the proteins with the functions of amino acid biosynthesis, cell envelope, cellular processes, energy metabolism, nucleosides and nucleotide metabolism, replication, transcription, translation, transport and binding as well as the proteins with other functions. In addition to these known function proteins, 63 proteins were annotated as hypothetical or conserved hypothetical proteins. CONCLUSIONS: Taken together, a total of 209 phytoplasma proteins have been experimentally verified, representing the most extensive survey of any phytoplasma proteome to date. This study provided a valuable dataset of phytoplasma proteins, and a better understanding of the energy metabolism and virulence mechanisms of MD phytoplasma.

Oxidative folding of hirudin in human serum.

Human serum contains factors that promote oxidative folding of disulphide proteins. We demonstrate this here using hirudin as a model. Hirudin is a leech-derived thrombin-specific inhibitor containing 65 amino acids and three disulphide bonds. Oxidative folding of hirudin in human serum is shown to involve an initial phase of rapid disulphide formation (oxidation) to form the scrambled isomers as intermediates. This is followed by the stage of slow disulphide shuffling of scrambled isomers to attain the native hirudin. The kinetics of regenerating the native hirudin depend on the concentrations of both hirudin and human serum. Quantitative regeneration of native hirudin in undiluted human serum can be completed within 48 h, without any redox supplement. These results cannot be adequately explained by the existing oxidized thiol agents in human serum or the macromolecular crowding effect, and therefore indicate that human serum may contain yet to be identified potent oxidase(s) for assisting protein folding.