Search

Multi-label feature selection attracts considerable attention from multi-label learning. Information-theory based multi-label feature selection methods intend to select the most informative features and reduce the uncertain amount of information of labels. Previous methods regard the uncertain amount of information of labels as constant. In fact, as the classification information of the label set is captured by features, the remaining uncertainty of each label is changing dynamically. In this paper, we categorize labels into two groups: one contains the labels with few remaining uncertainty, which means that most of classification information with respect to the labels has been obtained by the already-selected features; another group contains the labels with extensive remaining uncertainty, which means that the classification information of these labels is neglected by already-selected features. Feature selection aims to select the new features with highly relevant to the labels in the second group. Existing methods do not distinguish the difference between two label groups and ignore the dynamic change amount of information of labels. To this end, a Relevancy Ratio is designed to clarify the dynamic change amount of information of each label under the condition of the already-selected features. Afterwards, a Weighted Feature Relevancy is defined to evaluate the candidate features. Finally, a new multi-label Feature Selection method based on Weighted Feature Relevancy (WFRFS) is proposed. The experiments obtain encouraging results of WFRFS in comparison to six multi-label feature selection methods on thirteen real-world data sets.

In daily life, the most direct and important way of human communication is voice. With the rise of the Internet and the development of communication technology, the proportion of non voice signals such as image and data in communication system is increasing. However, in most communication systems, voice transmission function is generally required, so it is still one of the necessary functions for many communication systems to transmit voice information effectively. With the rapid development of the Internet, especially the recent 5G technology commercial and future civil, human–computer interaction will become more and more intelligent in the future, which also poses a greater challenge to speech recognition as a human–computer interface. Noise interference is one of the biggest obstacles to the practical application of speech system. Although a large number of noisy data based on deep learning can solve part of the noise robustness problem, non-stationary noise interference is still a great challenge for speech recognition system in very low SNR complex scenes. In addition, in the multi information fusion communication system, there are many kinds of data to be transmitted, and there are high requirements for bandwidth and storage space. Hence, this paper studies the construction of voice signal communication system based on the artificial intelligence and 5G technology. The model is designed and implemented considering the complex scenarios. The performance is validated through the simulations compared with the state-of-the-art methods.

To compare the outcomes of macular buckling (MB) surgery between myopic foveal detachment (FD) eyes with and without ellipsoid zone (EZ) disruption. A retrospective, case-control study. Forty-four consecutive eyes from 44 patients received MB surgery for myopic FD between November 2017 and January 2019 were included. The eyes were divided into two groups according to the integrity of EZ on spectral-domain optical coherence tomography (SD-OCT): 28 eyes with disrupted EZ band and 16 eyes with intact EZ band. Main outcome measures were visual acuity and the duration of subfoveal fluid (SFF) after MB. The mean follow-up time was 17.64 ± 6.61 and 16.06 ± 5.78 months in the disrupted EZ and intact EZ group, respectively (P = 0.430). The logMAR best-corrected visual acuity (BCVA) improved significantly, from 1.13 ± 0.46 and 1.12 ± 0.39 at baseline to 0.85 ± 0.65 (P = 0.002) and 0.53 ± 0.33 (P = 0.000) for the disrupted EZ group and intact EZ group, respectively. The mean visual improvement was 15.00 ± 14.14 Early Treatment Diabetic Retinopathy Study (ETDRS) letters for the disrupted EZ group and 26.88 ± 19.48 ETDRS letters for the intact EZ group. Significant difference was found on both final postoperative BCVA (P = 0.035) and visual improvement (P = 0.025). At 6 months, SFF remained in 53.57% (15/28) of the eyes in the disrupted EZ group and in only 12.50% (2/16) of the eyes in the intact EZ group (P = 0.018). The intact EZ group showed better functional and anatomical outcomes than the disrupted EZ group after MB surgery.

The advancement of cost-effective and selective electrocatalyst towards CO2 to CO conversion is crucial for renewable energy conversion and storage, thus to achieve carbon-neutral cycle in a sustainable manner. In this communication, we report that Cu2Sb decorated Cu nanowire arrays on Cu foil act as a highly active and selective electrocatalyst for CO2 to CO conversion. In CO2-saturated 0.1 M KHCO3, it achieves a high Faraday efficiency (FE) of 86.5% for CO, at −0.90 V vs. reversible hydrogen electrode (RHE). The H2/CO ratio is tunable from 0.08:1 to 5.9:1 by adjusting the potential. It is worth noting that HCOO− product was totally suppressed on such catalyst, compared with Sb counterpart. The improving selectivity for CO could be attributed to the bimetallic effect and nanowire arrays structure.

Tobacco (Nicotiana tabacum L.) is the most important non-food cash crop worldwide and has recently been considered a Cd hyperaccumulator. Metallothionein, a small–molecular weight protein with cysteine-rich domains, plays a key role in plant growth, development, and maintenance of metal homeostasis. The understanding of the tobacco metallothionein (MT) gene family remains unclear. Herein, we systematically characterized twelve NtMT genes in the tobacco genome and classified them into three phylogenetic subfamilies. A number of cis-elements related to plant responses to hormones and abiotic stresses were detected in the promoters of NtMT genes. Tissue expression pattern analysis indicated that NtMT4A/4B were expressed only in seeds, and NtMT2C/2F/2G were mainly expressed in roots. Moreover, most of the NtMT genes were highly induced by heavy metal stress and ion deficiency, suggesting their critical role in relieving metal toxicity and maintaining ion homeostasis in tobacco. This was the first study to describe the genome-wide analysis of the NtMT gene family in tobacco, and the results lay a foundation for understanding the functions of NtMT genes for further enhancing plant tolerance to heavy metal toxicity.

A data layout method suitable for workflow in a cloud computing environment with speech applications is designed and implemented in this research. With the continuous advancement of cloud computing, Internet of Things, wireless communications, mobile Internet and other technologies, data is growing and accumulating at an unprecedented rate. This paper provides the novel efficient speech data processing layout considering prior information and the data structure. On the basis of in-depth research on existing data layout strategies, this paper summarizes several aspects such as incompatibility with energy-saving modes based on storage area division, incompatibility with the heterogeneous HDFS clusters, inflexible data block segmentation algorithms, and randomness of general storage node selection aspects. Especially, considering that speech data related lower-case read-ahead and basic read-ahead are quite different, reflected in the purpose of read-ahead, read-ahead objects, read-ahead execution level, read-ahead trigger conditions, read-ahead feasibility, we propose the efficient robust data mining model to capture the data structure and feature. Through comparing with the other state-of-the-art models, the experiment section gives the clear analysis on the performance.

The mixture of glucose and β-disaccharide (MGD) synthesized by transglycosylation of glucose as a low-cost soluble carbon source can efficiently induce cellulase production in Trichoderma reesei, which holds potential for the biorefining of lignocellulosic biomass. However, it is not yet fully understood how MGD induces T. reesei cellulase. In this study, transcriptomic analyses were conducted to investigate the molecular basis of MGD for lignocellulose-degrading enzyme production of T. reesei Rut C30 compared with that on lactose. Particular attention was paid to CAZymes, transcription factors, transporters and other protein processing pathways related to lignocellulose degradation. As a result, MGD can elicit transcription of GH5-, GH6- and GH7-encoding cellulases that is up to 1.4-fold higher than that induced by lactose, but GH11- and GH74-encoding xylanases are downregulated by 1.7- and 4.4-fold, respectively. Gene expression profiles suggest that the transcription activators xyr1 and vib1 are significantly upregulated and that the mitogen-activated protein kinase pathway is strengthened compared to the case of lactose induction. In addition, hac1-encoding UPR-specific transcription factors are significantly upregulated by MGD, which may be enhanced due to proper folding and processing of nascent proteins. These findings provide a theoretical basis for further understanding the characterization of efficient cellulase production using MGD as an inducer in T. reesei and offer potential strategies for strain improvement.

Genomic prediction benefits hybrid rice breeding by increasing selection intensity and accelerating breeding cycles. With the rapid advancement of technology, other omic data, such as metabolomic data and transcriptomic data, are readily available for predicting breeding values for agronomically important traits. In this study, the best prediction strategies were determined for yield, 1000 grain weight, number of grains per panicle, and number of tillers per plant of hybrid rice (derived from recombinant inbred lines) by comprehensively evaluating all possible combinations of omic datasets with different prediction methods. It was demonstrated that, in rice, the predictions using a combination of genomic and metabolomic data generally produce better results than single-omics predictions or predictions based on other combined omic data. Best linear unbiased prediction (BLUP) appears to be the most efficient prediction method compared to the other commonly used approaches, including least absolute shrinkage and selection operator (LASSO), stochastic search variable selection (SSVS), support vector machines with radial basis function and epsilon regression (SVM-R(EPS)), support vector machines with radial basis function and nu regression (SVM-R(NU)), support vector machines with polynomial kernel and epsilon regression (SVM-P(EPS)), support vector machines with polynomial kernel and nu regression (SVM-P(NU)) and partial least squares regression (PLS). This study has provided guidelines for selection of hybrid rice in terms of which types of omic datasets and which method should be used to achieve higher trait predictability. The answer to these questions will benefit academic research and will also greatly reduce the operative cost for the industry which specializes in breeding and selection.

Retinal detachment can lead to severe visual loss if not treated timely. The early diagnosis of retinal detachment can improve the rate of successful reattachment and the visual results, especially before macular involvement. Manual retinal detachment screening is time-consuming and labour-intensive, which is difficult for large-scale clinical applications. In this study, we developed a cascaded deep learning system based on the ultra-widefield fundus images for automated retinal detachment detection and macula-on/off retinal detachment discerning. The performance of this system is reliable and comparable to an experienced ophthalmologist. In addition, this system can automatically provide guidance to patients regarding appropriate preoperative posturing to reduce retinal detachment progression and the urgency of retinal detachment repair. The implementation of this system on a global scale may drastically reduce the extent of vision impairment resulting from retinal detachment by providing timely identification and referral., Li et al. develop a cascaded deep learning system for automated retinal detachment and macular status detection based on ultra-widefield fundus (UWF) images. With reliable and comparable performance to an experienced opthamologist, this system can also provide guidance to patients regarding appropriate preoperative posturing to reduce RD progression.

In this study, we present novel molecular mechanisms by which FOXO1 functions as a tumor suppressor to prevent the pathogenesis of nasopharyngeal carcinoma (NPC). First, we observed that FOXO1 not only controlled tumor stemness and metastasis, but also sensitized NPC cells to cisplatin (DDP) in vitro and in vivo. Mechanistic studies demonstrated that FOXO1-induced miR-200b expression through the GSK3β/β-catenin/TCF4 network-mediated stimulation of ZEB1, which reduced tumor stemness and the epithelial–mesenchymal transition (EMT) signal. Furthermore, we observed FOXO1 interaction with MYH9 and suppression of MYH9 expression by modulating the PI3K/AKT/c-Myc/P53/miR-133a-3p pathway. Decreased MYH9 expression not only reduced its interactions with GSK3β, but also attenuated TRAF6 expression, which then decreased the ubiquitin-mediated degradation of GSK3β protein. Increased GSK3β expression stimulated the β-catenin/TCF4/ZEB1/miR-200b network, which increased the downstream tumor stemness and EMT signals. Subsequently, we observed that chemically synthesized cinobufotalin (CB) strongly increased FOXO1-induced DDP chemosensitivity by reducing MYH9 expression, and the reduction in MYH9 modulated GSK3β/β-catenin and its downstream tumor stemness and EMT signal in NPC. In clinical samples, the combination of low FOXO1 expression and high MYH9 expression indicated the worst overall survival rates. Our studies demonstrated that CB potently induced FOXO1-mediated DDP sensitivity by antagonizing its binding partner MYH9 to modulate tumor stemness in NPC.

A total of 395 single-zoospore isolates of Phytophthora sojae that were obtained from 467 soybean fields in Heilongjiang, China from 2011 to 2015 were identified for pathotypes using differential soybean cultivars with Rps1a, Rps1b, Rps1c, Rps1d, Rps1k, Rps2, Rps3a, Rps3b, Rps3c, Rps4, Rps5, Rps6, Rps7, and Rps8 resistance genes. The results showed that P. sojae was widespread in Heilongjiang, but not evenly distributed. A heavy infestation of P. sojae in eastern Heilongjiang coincided with the serious diseases caused by this pathogen in that region. Among 135 pathotypes were identified, 20 isolates matched races of P. sojae based on published race definitions. To our knowledge, this is the first report of races 2, 8, 23, 25, 26, 29, 33, 39, 42, 43, and 48 in Heilongjiang. Race 1, previously considered as the dominant race in Heilongjiang, only comprised 1.52 % of the isolates in the present study. Virulence frequencies of 395 isolates to 14 Rps genes ranged from 17.85 to 82.41 %. Less than 30 % of the isolates were virulent on cultivars containing Rps genes 1k, 1c, or 3a, which indicated that these Rps genes were more effective than other genes in Heilongjiang. All isolates could defeat more than two Rps genes. Ninety-six percent of the isolates were virulent against more than four Rps genes, which indicated that multi-virulence isolates existed in Heilongjiang. All 223 tested isolates were sensitive to metalaxyl, but EC50 for this fungicide against the pathogen increased 18-fold in the last 20 years. The results of the present study suggest that incorporating Rps genes 1c, 1k, and 3a into soybean cultivars and the use of metalaxyl-coated seeds should provide integrated approach to manage Phytophthora root and stem rot of soybean in Heilongjiang, China.

Background Subfoveal choroidal thickness (SFCT) in highly myopic eyes was found to be correlated with increasing age, refractive error (spherical equivalent), and axial length. Which factor is the most significant predictor of SFCT remains controversial. Methods A hospital-based cohort of highly myopic eyes (with spherical equivalent equal to or over 6.00 diopter) were retrospectively screened. Data from only right eye in those bilateral high myopia, and unilateral high myopia in any eye, were used for analysis. Correlations among the four biometric factors were analyzed. Linear correlation was performed to analyze the predictors of SFCT. Results A cohort of 312 eyes from 312 adults (98 men) was enrolled. Statistical analysis showed that axial length (R = − 0.592), spherical equivalent (R = − 0.471), and age (R = − 0.296) were significantly correlated with SFCT (P

The changes in retinal thickness and visual function in type 2 diabetic patients without clinical evidence of diabetic retinopathy were evaluated. A total of 141 diabetic subjects without retinopathy and 158 healthy subjects were enrolled in this study. Superior macular ganglion cell complex thicknesses were significantly decreased in diabetic cases, and no significant peripapillary retinal nerve fiber layer thickness changes were observed. The contrast sensitivities at all space frequencies were significantly different between diabetic patients and controls. The mean P50 amplitude from pattern electroretinogram results was reduced significantly in the diabetic group. In the diabetic group, average superior ganglion cell complex thicknesses positively correlated with both contrast sensitivities at high spatial frequencies and P50 amplitudes. The results indicated that ganglion cell complex thickness and visual function changes could be observed in diabetic subjects before the onset of any significant diabetic retinopathy. Macular ganglion cell complex reduction occurred much earlier than peripapillary retinal nerve fiber layer thinning in diabetic patients without retinopathy.

Isoprene is an aviation fuel of high quality and an important polymer building block in the synthetic chemistry industry. In light of high oil prices, sustained availability, and environmental concerns, isoprene from renewable materials is contemplated as a substitute for petroleum-based product. Escherichia coli with advantages over other wild microorganisms, is considered as a powerful host for biofuels and chemicals. Here, we constructed a synthetic pathway of isoprene in E. coli by introducing an isoprene synthase (ispS) gene from Populus nigra, which catalyzes the conversion of dimethylallyl diphosphate (DMAPP) to isoprene. To improve the isoprene production, we overexpressed the native 1-deoxy-D: -xylulose-5-phosphate (DXP) synthase gene (dxs) and DXP reductoisomerase gene (dxr) in E. coli, which catalyzed the first step and the second step of MEP pathway, respectively. The fed-batch fermentation results showed that overexpression of DXS is helpful for the improvement of isoprene production. Surprisingly, heterologous expression of dxs and dxr from Bacillus subtilis in the E. coli expressing ispS resulted in a 2.3-fold enhancement of isoprene production (from 94 to 314 mg/L). The promising results showed that dxs and dxr from B. subtilis functioned more efficiently on the enhancement of isoprene production than native ones. This could be caused by the consequence of great difference in protein structures of the two original DXSs. It could be practical to produce isoprene in E. coli via MEP pathway through metabolic engineering. This work provides an alternative way for production of isoprene by engineered E. coli via MEP pathway through metabolic engineering.

Ten polymorphic microsatellite loci from the fungus Puccinia emaculata, responsible for rust disease of switchgrass (Panicum virgatum) were developed. Loci were characterized using 20 single-pustule derived isolates of P. emaculata collected from switchgrass plants growing in the southeastern US. The number of alleles per locus ranged from 2 to 5. Observed heterozygosity ranged from 0 to 0.89 and expected heterozygosity ranged from 0.21 to 0.77. These microsatellite loci provide a sufficient set of markers to perform genetic diversity and epidemiological studies of P. emaculata.