EEGGAN-Net: enhancing EEG signal classification through data augmentation.

Background: Emerging brain-computer interface (BCI) technology holds promising potential to enhance the quality of life for individuals with disabilities. Nevertheless, the constrained accuracy of electroencephalography (EEG) signal classification poses numerous hurdles in real-world applications. Methods: In response to this predicament, we introduce a novel EEG signal classification model termed EEGGAN-Net, leveraging a data augmentation framework. By incorporating Conditional Generative Adversarial Network (CGAN) data augmentation, a cropped training strategy and a Squeeze-and-Excitation (SE) attention mechanism, EEGGAN-Net adeptly assimilates crucial features from the data, consequently enhancing classification efficacy across diverse BCI tasks. Results: The EEGGAN-Net model exhibits notable performance metrics on the BCI Competition IV-2a and IV-2b datasets. Specifically, it achieves a classification accuracy of 81.3% with a kappa value of 0.751 on the IV-2a dataset, and a classification accuracy of 90.3% with a kappa value of 0.79 on the IV-2b dataset. Remarkably, these results surpass those of four other CNN-based decoding models. Conclusions: In conclusion, the amalgamation of data augmentation and attention mechanisms proves instrumental in acquiring generalized features from EEG signals, ultimately elevating the overall proficiency of EEG signal classification.

Enhancing crop production and carbon sequestration of wheat in arid areas by green manure with reduced nitrogen fertilizer.

Green manure with appropriate amount of chemical nitrogen fertilizer can increase crop yield, but also aggravate soil carbon emissions. However, it is unclear whether incorporation of green manure into the cropping pattern with reduced nitrogen amount can alleviate this situation and enhance carbon sequestration potential. So, a field experiment with split-plot design was set up in 2018 of northwest China, and studied the effects of nitrogen reduction on crop productivity, carbon emissions, and carbon sequestration potential in 2021-2023. The main plots were two cropping patterns, including multiple cropped green manure after wheat harvest (W-G) and fallow after wheat harvest (W). Three nitrogen application levels formed the split-plots, including local conventional nitrogen amount (N3, 180 kg ha-1), nitrogen amount reduced by 15% (N2, 153 kg ha-1) and 30% (N3, 126 kg ha-1). The results showed that W-G increased grain yield of wheat and energy yield of wheat multiple cropped green manure pattern. The multiple cropped green manure after wheat harvest with local conventional nitrogen amount reduced by 15% (W-GN2) had the significant increasing-effect, and increased grain yield of wheat by 9.6% and increased total energy yields by 39.3% compared to fallow after wheat harvest with local conventional nitrogen amount (W-N3). Relative to W-N3, W-GN2 did not significantly increase carbon emissions of wheat season, and increased total carbon emissions of cropping pattern by 11.1%. Compared to multiple cropped green manure after wheat harvest with local conventional nitrogen amount (W-GN3), W-GN2 decreased carbon emissions by 5.8% in wheat season and decreased by 3.9% in the whole cropping pattern. Therefore, W-GN2 gained high carbon emission efficiency based on grain yield, and were 9.9% and 11.2% higher than W-N3 and W-GN3, respectively. In addition, W-GN2 enhanced soil total nitrogen, carbon, and organic carbon contents, compared with W-N3, thus increasing soil carbon sequestration potential index (net primary productivity/carbon emissions). We conclude that multiple cropped leguminous green manure after wheat harvest with local conventional nitrogen amount reduced by 15% can enhance crop productivity and carbon sequestration potential of farmland in arid areas.

Reduced chemodiversity suppresses rhizosphere microbiome functioning in the mono-cropped agroecosystems.

BACKGROUND: Rhizodeposits regulate rhizosphere interactions, processes, nutrient and energy flow, and plant-microbe communication and thus play a vital role in maintaining soil and plant health. However, it remains unclear whether and how alteration in belowground carbon allocation and chemodiversity of rhizodeposits influences microbiome functioning in the rhizosphere ecosystems. To address this research gap, we investigated the relationship of rhizosphere carbon allocation and chemodiversity with microbiome biodiversity and functioning during peanut (Arachis hypogaea) continuous mono-cropping. After continuously labeling plants with 13CO2, we studied the chemodiversity and composition of rhizodeposits, along with the composition and diversity of active rhizosphere microbiome using metabolomic, amplicon, and shotgun metagenomic sequencing approaches based on DNA stable-isotope probing (DNA-SIP). RESULTS: Our results indicated that enrichment and depletion of rhizodeposits and active microbial taxa varied across plant growth stages and cropping durations. Specifically, a gradual decrease in the rhizosphere carbon allocation, chemodiversity, biodiversity and abundance of plant-beneficial taxa (such as Gemmatimonas, Streptomyces, Ramlibacter, and Lysobacter), and functional gene pathways (such as quorum sensing and biosynthesis of antibiotics) was observed with years of mono-cropping. We detected significant and strong correlations between rhizodeposits and rhizosphere microbiome biodiversity and functioning, though these were regulated by different ecological processes. For instance, rhizodeposits and active bacterial communities were mainly governed by deterministic and stochastic processes, respectively. Overall, the reduction in carbon deposition and chemodiversity during peanut continuous mono-cropping tended to suppress microbial biodiversity and its functions in the rhizosphere ecosystem. CONCLUSIONS: Our results, for the first time, provide the evidence underlying the mechanism of rhizosphere microbiome malfunctioning in mono-cropped systems. Our study opens new avenues to deeply disentangle the complex plant-microbe interactions from the perspective of rhizodeposits chemodiversity and composition and will serve to guide future microbiome research for improving the functioning and services of soil ecosystems. Video abstract.

Soil Respiration of Paddy Soils Were Stimulated by Semiconductor Minerals.

Large quantities of semiconductor minerals on soil surfaces have a sensitive photoelectric response. These semiconductor minerals generate photo-electrons and photo-hole pairs that can stimulate soil oxidation-reduction reactions when exposed to sunlight. We speculated that the photocatalysis of semiconductor minerals would affect soil carbon cycles. As the main component of the carbon cycle, soil respiration from paddy soil is often ignored. Five rice cropping areas in China were chosen for soil sampling. Semiconductor minerals were measured, and three main semiconductor minerals including hematile, rutile, and manganosite were identified in the paddy soils. The identified semiconductor minerals consisted of iron, manganese, and titanium oxides. Content of Fe2O3, TiO2, and MnO in the sampled soil was between 4.21-14%, 0.91-2.72%, and 0.02-0.22%, respectively. Most abundant semiconductor mineral was found in the DBDJ rice cropping area in Jilin province, with the highest content of Fe2O3 of 14%. Soils from the five main rice cropping areas were also identified as having strong photoelectric response characteristics. The highest photoelectric response was found in the DBDJ rice cropping area in Jilin province with a maximum photocurrent density of 0.48 µA/cm2. Soil respiration was monitored under both dark and light (3,000 lux light density) conditions. Soil respiration rates in the five regions were (from highest to lowest): DBDJ > XNDJ > XBDJ > HZSJ > HNSJ. Soil respiration was positively correlated with semiconductor mineral content, and soil respiration was higher under the light treatment than the dark treatment in every rice cropping area. This result suggested that soil respiration was stimulated by semiconductor mineral photocatalysis. This analysis provided indirect evidence of the effect semiconductor mineral photocatalysis has on the carbon cycle within paddy soils, while exploring carbon conversion mechanisms that could provide a new perspective on the soil carbon cycle.

Compost Addition Attenuates the Negative Impacts of High Soil Mineral Nitrogen Levels on Rhizosphere Microbial Characteristics and Enhances Cucumber Growth in Monoculture Systems.

Due to the increase in the human population, it is necessary to seek efficient methods of increasing crop productivity and, simultaneously, sustaining the soil. One way is to grow high demand crops continuously without rotating with other crops. This practice is often accompanied by increased rates of fertilizer application that can affect efficient nitrogen (N) cycling in the plant rhizosphere soil which, in turn, affects both plant growth and environmental pollution. In the present study, twelve various cucumber soils were selected from monoculture systems presenting different cropping years and divided into two groups including soils with relatively high mineral N (HMN) content (N > 100 mg kg−1 soil) and those with a lower mineral N (LMN) content (N < 100 mg kg−1 soil). All soils were amended with the addition of compost alone or in combination with bacterial inoculation to evaluate their effects on plant growth, microbial numbers, N mineralization, and N cycling genes. In general, the HMN soils increased (p < 0.05) net N mineralization (NNM) but did not statistically (p > 0.05) affect plant biomass compared to the LMN soils; however, compost addition increased both NNM and plant biomass in the HMN soils. In addition, the HMN soils had higher fungal pathogen numbers (FPNs) but lower total microbial biomass (TMB) and bacterial numbers (BNs) compared to the LMN soils; however, compost addition decreased FPNs but increased TMB and BNs in the HMN soils (all p < 0.05). Plant biomass was positively related to TMB, BN and NNM but was negatively related to FPN (all p < 0.05). In summary, compost addition reduced the high mineral N levels' adverse effects on the rhizosphere soil and plant growth.

Using Slit-Lamp Images for Deep Learning-Based Identification of Bacterial and Fungal Keratitis: Model Development and Validation with Different Convolutional Neural Networks.

In this study, we aimed to develop a deep learning model for identifying bacterial keratitis (BK) and fungal keratitis (FK) by using slit-lamp images. We retrospectively collected slit-lamp images of patients with culture-proven microbial keratitis between 1 January 2010 and 31 December 2019 from two medical centers in Taiwan. We constructed a deep learning algorithm consisting of a segmentation model for cropping cornea images and a classification model that applies different convolutional neural networks (CNNs) to differentiate between FK and BK. The CNNs included DenseNet121, DenseNet161, DenseNet169, DenseNet201, EfficientNetB3, InceptionV3, ResNet101, and ResNet50. The model performance was evaluated and presented as the area under the curve (AUC) of the receiver operating characteristic curves. A gradient-weighted class activation mapping technique was used to plot the heat map of the model. By using 1330 images from 580 patients, the deep learning algorithm achieved the highest average accuracy of 80.0%. Using different CNNs, the diagnostic accuracy for BK ranged from 79.6% to 95.9%, and that for FK ranged from 26.3% to 65.8%. The CNN of DenseNet161 showed the best model performance, with an AUC of 0.85 for both BK and FK. The heat maps revealed that the model was able to identify the corneal infiltrations. The model showed a better diagnostic accuracy than the previously reported diagnostic performance of both general ophthalmologists and corneal specialists.

[Effects of lime-ammonium bicarbonate fumigation and biofertilizer application on Fusarium wilt and biomass of continuous cropping cucumber and watermelon.] / çŸ³ç°ç¢³é“µç†è’¸ä¸Žæ–½ç”¨ç”Ÿç‰©æœ‰æœºè‚¥å¯¹è¿žä½œé»„ç“œå’Œè¥¿ç“œæž¯èŽç— åŠç”Ÿç‰©é‡çš„å½±å“.

In this study, the population size of soil microbes was determined using plate counting method after the application of lime-ammonium bicarbonate and ammonium bicarbonate fumigation. In addition, biofertilizer was applied after soil fumigation and population of Fusarium oxysporum, Fusarium wilt disease control efficiency and plant biomass were determined in the cucumber and watermelon continuous cropping soil. The results showed that the population of F. oxysporum in cucumber mono-cropped soil fumigated with lime-ammonium bicarbonate or ammonium bicarbonate was decreased by 95.4% and 71.4%, while that in watermelon mono-cropped soil was decreased by 87.3% and 61.2%, respectively compared with non-fumigated control (CK). Furthermore, the greenhouse experiment showed that biofertilizer application, soil fumigation and crop type showed significant effects on the number of soil F. oxysporum, Fusarium wilt disease incidence, disease control efficiency and plant biomass based on multivariate analysis of variance. In the lime-ammonium bicarbonate fumigated soil amended with biofertilizer (LFB), significant reductions in the numbers of F. oxysporum and Fusarium wilt disease incidence were observed in both cucumber and watermelon cropped soil compared to non-fumigated control soil applied with organic fertilizer. The disease control rate was 91.9% and 92.5% for cucumber and watermelon, respectively. Moreover, LFB also significantly increased the plant height, stem diameter, leaf SPAD, and dry biomass for cucumber and watermelon. It was indicated that biofertilizer application after lime-ammonium bicarbonate fumigation could effectively reduce the abundance of F. oxysporum in soil, control Fusarium wilt disease and improve plant biomass in cucumber and watermelon mono-cropping systems.

An Ectopic Network of Transcription Factors Regulated by Hippo Signaling Drives Growth and Invasion of a Malignant Tumor Model.

Cancer cells have abnormal gene expression profiles; however, to what degree these are chaotic or driven by structured gene regulatory networks is often not known. Here we studied a model of Ras-driven invasive tumorigenesis in Drosophila epithelial tissues and combined in vivo genetics with next-generation sequencing and computational modeling to decipher the regulatory logic of tumor cells. Surprisingly, we discovered that the bulk of the tumor-specific gene expression is controlled by an ectopic network of a few transcription factors that are overexpressed and/or hyperactivated in tumor cells. These factors are Stat, AP-1, the bHLH proteins Myc and AP-4, the nuclear hormone receptor Ftz-f1, the nuclear receptor coactivator Taiman/SRC3, and Mef2. Notably, many of these transcription factors also are hyperactivated in human tumors. Bioinformatic analysis predicted that these factors directly regulate the majority of the tumor-specific gene expression, that they are interconnected by extensive cross-regulation, and that they show a high degree of co-regulation of target genes. Indeed, the factors of this network were required in multiple epithelia for tumor growth and invasiveness, and knockdown of several factors caused a reversion of the tumor-specific expression profile but had no observable effect on normal tissues. We further found that the Hippo pathway effector Yorkie was strongly activated in tumor cells and initiated cellular reprogramming by activating several transcription factors of this network. Thus, modeling regulatory networks identified an ectopic and ordered network of master regulators that control a large part of tumor cell-specific gene expression.

Reducing environmental risk of excessively fertilized soils and improving cucumber growth by Caragana microphylla-straw compost application in long-term continuous cropping systems.

Continuous cropping is a common agricultural practice in the word. In China, farmers often apply excessive fertilizers to fields in an attempt to maintain yields in continuous cropping systems. However, this practice often results in high nutrient concentrations in soils, nutrient pollution in leaching water and more crop disease. Here, we investigated 8 different soils from continuously cropped cucumbers in Northern China that grouped into those with extremely high nutrient levels (EHNL) and those with lower nutrient levels (LNL). All soils were treated with Caragana microphylla-straw (CMS) compost addition, and then were used to measure soil physiochemical and microbial properties, leaching water quality, plant root growth and cucumber fruit yield. In general, the EHNL-soil showed higher nitrate, phosphorus and potassium concentrations in the leaching water compared to the LNL-soil. However, the CMS compost application increased soil nutrient and water holding capacities, total microbial biomass (bacteria and fungi), root length, plant biomass and fruit yields, but decreased nutrient concentrations in the leaching water from the EHNL-soil. In addition, the CMS compost decreased the number of Fusarium oxysporum f. sp. cucumerinum in soils with very high concentration of mineral nitrogen. Our results infer that CMS compost application was an effective method for reducing environmental risk of excessively fertilized soils.

Suppression of Pratylenchus penetrans populations in potato and tomato using african marigolds.

Current strategies for management of Pratylenchus penetrans in both white potato and tomato consist of the use of fumigant or non-fumigant nematicides or crop rotation. The objective of this study was to determine if double-cropping African marigolds (Tagetes erecta) with potatoes or tomatoes could reduce P. penetrans populations. Plots were 10 m x 3 m arranged in a randomized complete block design with four replications. Treatments included marigolds, potatoes or tomatoes, and natural weedy fallow followed by either potatoes or tomatoes. Nematode populations were sampled before spring planting, between crops in August and after harvest in November. During the 3 years of the study, P. penetrans soil population density declined by an average of 93% from the pre-plant level when marigold was grown in rotation with potato and by 98% when marigold was grown.in rotation with tomato. Weedy fallow preceding potato resulted in an average decline in P. penetrans soil population density of 38%, and a similar decrease (37%) was seen when fallow preceded tomato. There was a significant reduction in the number of P. penetrans found in both potato and tomato roots when the crops followed marigolds. These results suggest that P. penetrans population density may be significantly reduced when marigolds are double-cropped with potatoes or tomatoes.

Nematode control related to fusarium wilt in soybean and root rot and zinc deficiency in corn.

Nematode and disease problems of irrigated, double-cropped soybean and corn, and zinc deficiency of corn were investigated. Ethylene dibromide, phenamiphos, and aldicarb were equally effective for controlling nematodes and increasing yields of corn planted minimum-till and soybean planted in a moldboard plow prepared seedbed. The residual effects on yields of nematicides applied to the preceeding crop occurred during 3 years for soybean and 1 year for corn. Fusarium wilt symptoms of soybean that developed during 2 years of the study were less severe in all nematicide-treated plots than in control plots. Typical zinc deficiency symptoms on 30-day-old corn plants were observed during 1 year of the study in certain plots. Symptoms were not evident on plants grown on plots treated with ethylene dibromide, and only occasional plants had symptoms on plots treated with phenamiphos and aldicarb. The amount of yield response directly related to nematode control could not be determined because of the apparent interaction of nematodes on the expression of Fusarium wilt of soybean. Our study strongly indicates that the expression of Fusarium wilt of soybean and zinc deficiency in corn are influenced by nematodes and that nematicides will reduce their severity.