RESUMEN
Olfactory integration is important for survival in a natural habitat. However, how the nervous system processes signals of two odorants present simultaneously to generate a coherent behavioral response is poorly understood. Here, we characterize circuit basis for a form of olfactory integration in Caenorhabditis elegans. We find that the presence of a repulsive odorant, 2-nonanone, that signals threat strongly blocks the attraction of other odorants, such as isoamyl alcohol (IAA) or benzaldehyde, that signal food. Using a forward genetic screen, we found that genes known to regulate the structure and function of sensory neurons, osm-5 and osm-1, played a critical role in the integration process. Loss of these genes mildly reduces the response to the repellent 2-nonanone and disrupts the integration effect. Restoring the function of OSM-5 in either AWB or ASH, two sensory neurons known to mediate 2-nonanone-evoked avoidance, is sufficient to rescue. Sensory neurons AWB and downstream interneurons AVA, AIB, RIM that play critical roles in olfactory sensorimotor response are able to process signals generated by 2-nonanone or IAA or the mixture of the two odorants and contribute to the integration. Thus, our results identify redundant neural circuits that regulate the robust effect of a repulsive odorant to block responses to attractive odorants and uncover the neuronal and cellular basis for this complex olfactory task.
Asunto(s)
Caenorhabditis elegans/fisiología , Células Receptoras Sensoriales/fisiología , Olfato/fisiología , Animales , Caenorhabditis elegans/genética , Cetonas/farmacología , Mutación , Odorantes , Pentanoles/farmacología , Olfato/efectos de los fármacosRESUMEN
Perfluoroalkyl and polyfluoroalkyl substance (PFAS) is a large group of fluorinated synthetic chemicals, e.g., perfluorooctanoic acid (PFOA), perfluorooctanesulfonic acid (PFOS), perfluorohexanesulfonic acid (PFHxS), perfluorodecanoic acid (PFDA), and perfluorononanoic acid (PFNA). Many epidemiological studies have found that PFAS exposure is associated with hypertension risk, but others possess a different opinion. Overall, the relationship between PFASs and hypertension risk remains controversial. We sought to conduct a systematic review and meta-analysis to clarify the association between PFAS exposure and human risk of hypertension.We conducted a meta-analysis based on population-involving studies published from 1975 to 2023, which we collected from Web of Science, PubMed, and Embase databases. The odds ratio (OR) and standardized mean difference (SMD), with their 95% confidence interval (CI), were used to assess the risk of hypertension with PFAS exposure. The statistical heterogeneity among studies was assessed with the Q-test and I2 statistics. Research publications related to our meta-analysis topic were systematically reviewed.Fourteen studies involving 71,663 participants, in which 26,281 suffered hypertension, met the inclusion criteria. Our analyses suggest that exposure to general PFAS (OR = 1.09, 95% CI = 1.04-1.14) or PFOS (OR = 1.17, 95% CI = 1.05-1.30) is associated with hypertension risk. Specifically, elevated levels of general PFAS (SMD = 0.25, 95% CI = 0.08-0.42), PFHxS (SMD = 0.17, 95% CI = 0.07-0.27), and PFDA (SMD = 0.08, 95% CI = 0.02-0.13) are associated with a high risk of hypertension.Our meta-analysis indicates that PFAS exposure is a risk factor for hypertension, and increased hypertension risk is associated with higher PFAS levels. Further study may eventually provide a better and more comprehensive elucidation of the potential mechanism of this association.
Asunto(s)
Ácidos Alcanesulfónicos , Contaminantes Ambientales , Fluorocarburos , Humanos , Ácidos Alcanesulfónicos/efectos adversos , Fluorocarburos/efectos adversosRESUMEN
In modern industries, high precision dimensional measurement plays a pivotal role in product inspection and sub-pixel edge detection is the core algorithm. Traditional interpolation and moment methods have achieved some success. However, those methods still have shortcomings. For example, the accuracy is still insufficient with the resolution limitation of the image sensor. Moreover, prediction results can be affected by image noise. With the recent success of deep learning technology, we propose a sub-pixel edge detection method based on convolution neural network (CNN) and bi-directional long short-term memory (LSTM). First, one-dimensional visual geometry group-16 (VGG-16) is employed to extract edge features. Then, a transformation operation is developed to generate sequence information. Lastly, bi-directional LSTM with fully-connected layers is introduced to output edge positions. Experimental results on our steel plate dataset demonstrate that our method achieves superior accuracy and anti-noise ability than traditional methods.
RESUMEN
The stockline, which describes the measured depth of the blast furnace (BF) burden surface with time, is significant to the operator executing an optimized charging operation. For the harsh BF environment, noise interferences and aberrant measurements are the main challenges of stockline detection. In this paper, a novel encoder-decoder architecture that consists of a convolution neural network (CNN) and a long short-term memory (LSTM) network is proposed, which suppresses the noise interferences, classifies the distorted signals, and regresses the stockline in a learning way. By leveraging the LSTM, we are able to model the longer historical measurements for robust stockline tracking. Compared to traditional hand-crafted denoising processing, the time and efforts could be greatly saved. Experiments are conducted on an actual eight-radar array system in a blast furnace, and the effectiveness of the proposed method is demonstrated on the real recorded data.
RESUMEN
Efficient and accurate semantic segmentation is the key technique for automatic remote sensing image analysis. While there have been many segmentation methods based on traditional hand-craft feature extractors, it is still challenging to process high-resolution and large-scale remote sensing images. In this work, a novel patch-wise semantic segmentation method with a new training strategy based on fully convolutional networks is presented to segment common land resources. First, to handle the high-resolution image, the images are split as local patches and then a patch-wise network is built. Second, training data is preprocessed in several ways to meet the specific characteristics of remote sensing images, i.e., color imbalance, object rotation variations and lens distortion. Third, a multi-scale training strategy is developed to solve the severe scale variation problem. In addition, the impact of conditional random field (CRF) is studied to improve the precision. The proposed method was evaluated on a dataset collected from a capital city in West China with the Gaofen-2 satellite. The dataset contains ten common land resources (Grassland, Road, etc.). The experimental results show that the proposed algorithm achieves 54.96% in terms of mean intersection over union (MIoU) and outperforms other state-of-the-art methods in remote sensing image segmentation.
RESUMEN
The extraction of spatiotemporal neuron activity from calcium imaging videos plays a crucial role in unraveling the coding properties of neurons. While existing neuron extraction approaches have shown promising results, disturbing and scattering background and unused depth still impede their performance. To address these limitations, we develop an automatic and accurate neuron extraction paradigm, dubbed as decomposition-estimation-reconstruction (DER), consisting of D-procedure, E-procedure, and R-procedure. Specifically, the D-procedure first decomposes the raw data into a low-rank background and a sparse neuron signal, and regularizes L0 -norm priors of intensity and gradient of the neuron signal to suppress blurring and artifact effects. Then, the E-procedure estimates the depth-dependent transmission of the neuron signal based on its bright and dark channel priors. The R-procedure finally integrates the depth estimation of the neuron signal as a content-importance weight into a constrained non-negative matrix decomposition framework, which facilitates accurate neuron locations to boost the quality of extracted neurons. These three procedures are coupled in a cascade manner, where the former copes with calcium imaging data to facilitate the subsequent one. Comprehensive experiments on neuron extraction from calcium imaging videos demonstrate the superiority of our DER paradigm in both qualitative results and quantitative assessments over state-of-the-art methods.
RESUMEN
Sodium azide (NaN 3 ) is widely used as an anesthetic in the C. elegans community for studying animal behavior. It is not known whether other azides can function as anesthetics. This is quite important for the C. elegans labs in which NaN 3 is not a convenient choice, such as all the labs located in China, where NaN 3 is under tight regulation, and alternative anesthetics need to be characterized. In the present study, we focused on another three azides, potassium azide (KN 3 ), trimethylsilyl azide (TMSA), and diphenyl phosphoryl azide (DPPA), which are not regulated in China. We characterized their performance in chemotactic behavioral assays and buffer-based assays. Our results suggest that KN 3 can immobilize worms as effectively as NaN 3 in the above-mentioned assays. Therefore, we recommend KN 3 as a routine anesthetic for C. elegans labs.
RESUMEN
The miniMOS technique has been widely used in the C. elegans community to generate single copy insertions. A worm is considered as a potential insertion candidate if it is resistant to G418 antibiotics and does not express a co-injected fluorescence marker. If the expression of the extrachromosomal array is very low, it is possible for a worm to be mistakenly identified as a miniMOS candidate, as this low expression level can still confer resistance to G418 without producing a detectable fluorescence signal from the co-injection marker. This may increase the workload for identifying the insertion locus in the subsequent steps. In the present study, we modified the plasmid platform for miniMOS insertion by incorporating a myo-2 promoter-driven TagRFP or a ubiquitous H2B::GFP expression cassette into the targeting vector and introducing two loxP sites flanking the selection cassettes. Based on this new miniMOS tool kit, the removable fluorescence reporters can be used to visualize the single copy insertions, greatly reducing insertion locus identification efforts. In our experience, this new platform greatly facilitates the isolation of the miniMOS mutants.
Asunto(s)
Caenorhabditis elegans , Ingeniería Genética , Animales , Animales Modificados Genéticamente/genética , Caenorhabditis elegans/genética , Ingeniería Genética/métodosRESUMEN
Objective. Magnetic resonance imaging (MRI) has been widely used for the analysis and diagnosis of brain diseases. Accurate and automatic brain tumor segmentation is of paramount importance for radiation treatment. However, low tissue contrast in tumor regions makes it a challenging task.Approach. We propose a novel segmentation network named Category Guided Attention U-Net (CGA U-Net). In this model, we design a Supervised Attention Module (SAM) based on the attention mechanism, which can capture more accurate and stable long-range dependency in feature maps without introducing much computational cost. Moreover, we propose an intra-class update approach to reconstruct feature maps by aggregating pixels of the same category.Main results. Experimental results on the BraTS 2019 datasets show that the proposed method outperformers the state-of-the-art algorithms in both segmentation performance and computational complexity.Significance. The CGA U-Net can effectively capture the global semantic information in the MRI image by using the SAM module, while significantly reducing the computational cost. Code is available athttps://github.com/delugewalker/CGA-U-Net.
Asunto(s)
Neoplasias Encefálicas , Procesamiento de Imagen Asistido por Computador , Algoritmos , Neoplasias Encefálicas/diagnóstico por imagen , Humanos , Procesamiento de Imagen Asistido por Computador/métodos , Imagen por Resonancia Magnética/métodos , Redes Neurales de la ComputaciónRESUMEN
The encoder-decoder architecture is widely used as a lightweight semantic segmentation network. However, it struggles with a limited performance compared to a well-designed Dilated-FCN model for two major problems. First, commonly used upsampling methods in the decoder such as interpolation and deconvolution suffer from a local receptive field, unable to encode global contexts. Second, low-level features may bring noises to the network decoder through skip connections for the inadequacy of semantic concepts in early encoder layers. To tackle these challenges, a Global Enhancement Method is proposed to aggregate global information from high-level feature maps and adaptively distribute them to different decoder layers, alleviating the shortage of global contexts in the upsampling process. Besides, aLocal Refinement Module is developed by utilizing the decoder features as the semantic guidance to refine the noisy encoder features before the fusion of these two (the decoder features and the encoder features). Then, the two methods are integrated into a Context Fusion Block, and based on that, a novel Attention guided Global enhancement and Local refinement Network (AGLN) is elaborately designed. Extensive experiments on PASCAL Context, ADE20K, and PASCAL VOC 2012 datasets have demonstrated the effectiveness of the proposed approach. In particular, with a vanilla ResNet-101 backbone, AGLN achieves the state-of-the-art result (56.23% mean IOU) on the PASCAL Context dataset. The code is available at https://github.com/zhasen1996/AGLN.
RESUMEN
The widely used SiO2 and TiO2 nanoparticles (NPs) can accumulate in industrial wastewaters, thereby posing challenge to biological wastewater treatment processes. In this work, the performance of aerobic granular sludge (AGS) reactors fed with wastewater containing 50 mg L-1 SiO2 and TiO2 NPs were investigated. The results showed that the granules could resist the NPs in wastewater (no disintegration of granules was observed). SiO2 NPs had a negative effect on the settleability of granules, with the SVI30 increased by 64.5% and protein secretion decreased by 29.9%. To the contrary, the settleability of granules was improved in the presence of TiO2 NPs due to the increase of the protein secretion. Possibly because of the compact and layered structure of granules, in the presence of both types of NPs, no obvious reduction of the overall removal efficiency of organics was found, and nanoparticle-resistant strains were enriched. The overall nitrification and denitrification efficiencies were hardly affected by SiO2 NPs while significantly inhibited by TiO2 NPs. Some functional genera, such as Hyphomicrobium and Acidovorax, showed growth inhibition with TiO2 NPs, which might be responsible for the reduction of nitrification and denitrification efficiencies.
Asunto(s)
Microbiota , Nanopartículas , Aerobiosis , Reactores Biológicos , Nanopartículas/toxicidad , Nitrógeno/metabolismo , Aguas del Alcantarillado/química , Dióxido de Silicio , Titanio , Eliminación de Residuos Líquidos/métodos , Aguas ResidualesRESUMEN
Gravity-driven membrane (GDM) filtration system is a promising process for decentralized drinking water treatment. During the operation, membrane relaxation and shear stress could be simply achieved by intermittent filtration and water disturbance (created by occasionally shaking membrane model or stirring water in membrane tank), respectively. To better understand the impact of membrane relaxation and shear stress on the biofouling layer and stable flux in GDM system, action of daily 60-min intermission, daily flushing (cross-flow velocity = 10 cm s-1, 1 min), and the combination of the two (flushed right after the 60-min intermission) were compared. The results showed that membrane relaxation and shear stress lonely was ineffective in improving the stable flux, while their combination enhanced the stable flux by 70%. A more open and spatially heterogeneous biofouling layer with a low extracellular polymeric substance (EPS) content and a high microbial activity was formed under the combination of membrane relaxation and shear stress. In-situ optical coherence tomography (OCT) observation revealed that, during intermission, the absence of pushing force by water flow induced a reversible expansion of biofouling layer, and the biofouling layer restored to its initial state soon after resuming filtration. Shear stress caused abrasion and erosion on the biofouling surface, but it exerted little effect on the interior of biofouling layer. Under the combination, however, both the surface and interior of biofouling layer were disturbed because of 1) the water vortexes caused by rough biofouling layer surface, and 2) the porous structure after 60-min intermission. This disturbance, in turn, helped the biofouling layer maintain its roughness and porosity, thereby improving the stable flux of GDM system.
Asunto(s)
Incrustaciones Biológicas , Purificación del Agua , Matriz Extracelular de Sustancias Poliméricas , Filtración , Membranas ArtificialesRESUMEN
Cu(II) is one of the most widely-existed heavy metal ions in industrial effluents. A high concentration of Cu(II) leads to strong toxic effects on microorganisms and sludge for treating industrial wastewater which often contains aromatic pollutants. Granular sludge has different characteristics compared with floc sludge, and it may exhibit unique responses to the high concentration of Cu(II). Therefore, in this study, the variations of sludge properties and pollutant removal were investigated in the aerobic granular sludge (AGS) system with 0, 5, and 10 mg L-1 of Cu(II). The results suggested that both levels of Cu(II) promoted protein secretion and bounded with extracellular polymeric substances; thus, led to more compact granules with better settleability. Cu(II) had limited impacts on the overall organic degradation and denitrification efficiency, while it exerted significant negative effect on nitrification. The average NH4+-N concentration reached 1.4 ± 0.5, 6.7 ± 3.1, and 8.4 ± 1.5 mg L-1 in the effluent when the influent contained 0, 5, and 10 mg L-1 of Cu(II), respectively. The microbial community succession showed that no reduction was observed for the total relative abundance of main groups involved in organic removal such as Pseudoxanthomonas, Acidovorax, Acinetobacter, and Thauera. However, the growth of some functional groups such as Saccharibacteria for nitrification was inhibited by the toxic effect of Cu(II). These findings suggested that AGS could resist to the long-term toxic effects of Cu(II) by multiple rationales.
Asunto(s)
Cobre/toxicidad , Eliminación de Residuos Líquidos/métodos , Contaminantes Químicos del Agua/toxicidad , Aerobiosis , Reactores Biológicos/microbiología , Desnitrificación , Microbiota , Nitrificación , Nitrógeno , Aguas del Alcantarillado/microbiología , Aguas ResidualesRESUMEN
HYPOTHESIS: A superhydrophilic membrane with rough and hierarchical structures is possibly fouled by surfactant-stabilized oil and organic foulants, because these foulants could not be hindered by the water layer formed on superhydrophilic membrane surface. A dynamic membrane was possibly an effective method to address this fouling problem. EXPERIMENTS: A microfiltration membrane, a nanofiber membrane, and a dynamic membrane were used for the separation of surfactant-free emulsions, surfactant-stabilized emulsions, and the surfactant-stabilized emulsions containing typical organic foulants. The oil rejection and membrane fouling were compared. FINDINGS: The microfiltration membrane, nanofiber membrane, and dynamic membrane had high resistances to the fouling by surfactant-free emulsions because these membranes were underwater superoleophobic. However, these membranes showed low resistances to the fouling by surfactant-stabilized oil droplets and organic foulants. For the dynamic membrane, the oil droplets and organic foulants trapped in the separation layer could be readily removed in the detachment-washing-recoating steps; therefore, almost no physically irreversible fouling was observed in the multi-cycle filtration. With the size distributions of oil droplets in the emulsions and the particle of the dynamic membrane, the rejection of oil by the dynamic membrane could be calculated by simply assuming that the particle was spherical, uniform, and tightly packed.
RESUMEN
Due to the lack of robust ways to quantify aggregations, fouling of two-foulant aggregations is poorly understood. This work systematically reports the ultrafiltration membrane fouling by aggregations formed from two oppositely charged organic foulants (i.e., humic acid (HA) and lysozyme (LYS)) with the aid of resonance light scattering (RLS) technique. RLS provides an effective approach to detecting the aggregation concentration and reveals that the HA-LYS aggregations were formed at a mass ratio of m(LYS)/m(HA)â¯=â¯2.77. During the filtration of the mixture of HA and LYS, aggregations over individual foulants were identified to be the main substances deposited on the membrane surface, where the mass of deposition had a good linear relationship with the feed concentration of the aggregations. The HA-LYS aggregations might decrease the total fouling due to their large size, but reduce the fouling reversibility. In the pH range of 5.5-9.2, the pH value had limiting effects on the concentration of HA-LYS aggregations, as well as the consequent fouling. At low ionic strength, the membrane fouling by HA-LYS aggregations decreased as the ionic strength increased due to the reduction of the aggregation concentration. Oppositely, at high ionic strength, this tendency was reversed due to the electrical double layer compression effect. These results suggest that RLS is a simple and effective way to quantify the aggregations of foulants, and the aggregations of foulants have distinct fouling behaviors compared with the individual foulants.
Asunto(s)
Membranas Artificiales , Purificación del Agua , Filtración , Sustancias Húmicas , Concentración Osmolar , UltrafiltraciónRESUMEN
Semantic segmentation and single-view depth estimation are two fundamental problems in computer vision. They exploit the semantic and geometric properties of images, respectively, and are thus complementary in scene understanding. In this paper, we propose a collaborative deconvolutional neural network (C-DCNN) to jointly model these two problems for mutual promotion. The C-DCNN consists of two DCNNs, of which each is for one task. The DCNNs provide a finer resolution reconstruction method and are pretrained with hierarchical supervision. The feature maps from these two DCNNs are integrated via a pointwise bilinear layer, which fuses the semantic and depth information and produces higher order features. Then, the integrated features are fed into two sibling classification layers to simultaneously learn for semantic segmentation and depth estimation. In this way, we combine the semantic and depth features in a unified deep network and jointly train them to benefit each other. Specifically, during network training, we process depth estimation as a classification problem where a soft mapping strategy is proposed to map the continuous depth values into discrete probability distributions and the cross entropy loss is used. Besides, a fully connected conditional random field is also used as postprocessing to further improve the performance of semantic segmentation, where the proximity relations of pixels on position, intensity, and depth are jointly considered. We evaluate our approach on two challenging benchmarks: NYU Depth V2 and SUN RGB-D. It is demonstrated that our approach effectively utilizes these two kinds of information and achieves state-of-the-art results on both the semantic segmentation and depth estimation tasks.
RESUMEN
We investigated biofouling in ultrafiltration (UF) for drinking water treatment and its control by backwashing with chlorinated-water or pure water. By using sodium azide to suppress biological growth, the relative contribution of biofouling to total fouling was estimated, and its value (5.3-56.0%) varied with the feed water, and increased with the increases of filtration time and membrane flux. The biofouling layer could partially remove biodegradable organic matter and ammonia (32.9-74.2%). Backwashing using chlorinated-water partly inactivated the microorganisms (23.8%) but increased the content of extracellular polymeric substances (7.7%) in the biofouling layer. In contrast, backwashing using pure water led to a looser and more porous fouling layer according to optical coherence tomography observation. Consequently, the latter was more effective in reducing fouling resistance (33.41% reduction) compared to backwashing by chlorinated-water (8.6%). These findings reveal the critical roles of biofouling in pollutants removal in addition to membrane permeability, which has important implications for addressing seasonal ammonia pollution.
Asunto(s)
Incrustaciones Biológicas , Purificación del Agua/métodos , Membranas Artificiales , Ultrafiltración , AguaRESUMEN
Gravity-driven membrane (GDM) filtration is a promising decentralized drinking water treatment process. To improve the performance of GDM system, a thin layer of adsorbent was pre-deposited on the membrane surface prior to filtration (adsorbent-laden GDM system). The tested adsorbents include powdered activated carbon (PAC) and anion exchange resin (AER), and an unmodified GDM system and a SiO2-laden GDM system were used as controls. In the adsorbent-laden GDM systems, the adsorption of the PAC and AER increased the removal efficiency of natural organic matter by 7.2-43.5% and microcystin-LR, atrazine, and bisphenol A by 7.9-81.2%. The presence of adsorbent particles increased the amount of microorganisms in the cake layer and therefore increased the removal efficiency of assimilable organic matter (AOC) by 20.1-34.4%. In the adsorbent-laden GDM systems, the physically irrecoverable fouling decreased because of the reduction in membrane foulants by the adsorbent layer. However, the presence of adsorbent particles in the cake layer counteracted this effect and increased the physically recoverable fouling. Consequently, the pre-deposited adsorbent layers had only a limited effect on the stabilized flux (2.26-2.65 L/m2 h). A bilayer structure was found in the cake layer of the adsorbent-laden GDM systems via scanning electron microscopy (SEM), and the cake layer was looser in the presence of adsorbent particles. These results demonstrate that pre-depositing a thin layer of adsorbents on the membrane surface of the GDM system can significantly improve the quality of the permeate without decreasing the stabilized flux.