A novel approach for ASD recognition based on graph attention networks.

Early detection and diagnosis of Autism Spectrum Disorder (ASD) can significantly improve the quality of life for affected individuals. Identifying ASD based on brain functional connectivity (FC) poses a challenge due to the high heterogeneity of subjects' fMRI data in different sites. Meanwhile, deep learning algorithms show efficacy in ASD identification but lack interpretability. In this paper, a novel approach for ASD recognition is proposed based on graph attention networks. Specifically, we treat the region of interest (ROI) of the subjects as node, conduct wavelet decomposition of the BOLD signal in each ROI, extract wavelet features, and utilize them along with the mean and variance of the BOLD signal as node features, and the optimized FC matrix as the adjacency matrix, respectively. We then employ the self-attention mechanism to capture long-range dependencies among features. To enhance interpretability, the node-selection pooling layers are designed to determine the importance of ROI for prediction. The proposed framework are applied to fMRI data of children (younger than 12 years old) from the Autism Brain Imaging Data Exchange datasets. Promising results demonstrate superior performance compared to recent similar studies. The obtained ROI detection results exhibit high correspondence with previous studies and offer good interpretability.

Supervised and Semisupervised Manifold Embedded Knowledge Transfer in Motor Imagery-Based BCI.

A long calibration procedure limits the use in practice for a motor imagery (MI)-based brain-computer interface (BCI) system. To tackle this problem, we consider supervised and semisupervised transfer learning. However, it is a challenge for them to cope with high intersession/subject variability in the MI electroencephalographic (EEG) signals. Based on the framework of unsupervised manifold embedded knowledge transfer (MEKT), we propose a supervised MEKT algorithm (sMEKT) and a semisupervised MEKT algorithm (ssMEKT), respectively. sMEKT only has limited labelled samples from a target subject and abundant labelled samples from multiple source subjects. Compared to sMEKT, ssMEKT adds comparably more unlabelled samples from the target subject. After performing Riemannian alignment (RA) and tangent space mapping (TSM), both sMEKT and ssMEKT execute domain adaptation to shorten the differences among subjects. During domain adaptation, to make use of the available samples, two algorithms preserve the source domain discriminability, and ssMEKT preserves the geometric structure embedded in the labelled and unlabelled target domains. Moreover, to obtain a subject-specific classifier, sMEKT minimizes the joint probability distribution shift between the labelled target and source domains, whereas ssMEKT performs the joint probability distribution shift minimization between the unlabelled target domain and all labelled domains. Experimental results on two publicly available MI datasets demonstrate that our algorithms outperform the six competing algorithms, where the sizes of labelled and unlabelled target domains are variable. Especially for the target subjects with 10 labelled samples and 270/190 unlabelled samples, ssMEKT shows 5.27% and 2.69% increase in average accuracy on the two abovementioned datasets compared to the previous best semisupervised transfer learning algorithm (RA-regularized common spatial patterns-weighted adaptation regularization, RA-RCSP-wAR), respectively. Therefore, our algorithms can effectively reduce the need of labelled samples for the target subject, which is of importance for the MI-based BCI application.

Two rice MYB transcription factors maintain male fertility in response to photoperiod by modulating sugar partitioning.

Photoperiod-dependent male fertility is a critical enabler of modern hybrid breeding. A MYB transcription factor, CSA, is a key regulator of sugar partitioning in rice anthers, disruption of which causes photoperiod-sensitive male sterility. However, little is known about the molecular mechanisms governing plant fertility in response to photoperiod. Here, we have obtained another rice photoperiod-sensitive male sterile mutant, csa2, which exhibits semi-sterility under long-day (LD) conditions, with normal fertility under short-day (SD) conditions. CSA2 specifically expressed in anthers, and here is shown to be indispensable for sugar partitioning to anthers under LD conditions. The CSA2 protein can restore the fertility of csa mutants under SD conditions when expressed in a CSA-specific pattern, indicating that the two proteins share common downstream regulatory targets. Transcriptomic analyses also reveal discrete regulatory targets in anthers. Furthermore, the regulatory role of CSA2 in sugar transport was influenced by the photoperiod conditions during floral initiation, not simply during anther development. Collectively, we propose that rice evolved at least two MYB proteins, CSA2 and CSA, that regulate sugar transport in anthers under LD and SD conditions, respectively. This finding provides insight into the molecular mechanisms that regulate male fertility in response to photoperiod.

Defective Pollen Wall 3 (DPW3), a novel alpha integrin-like protein, is required for pollen wall formation in rice.

In flowering plants, pollen wall is a specialized extracellular cell-wall matrix surrounding male gametophytes and acts as a natural protector of pollen grains against various environmental and biological stresses. The formation of pollen wall is a complex but well-regulated process, which involves the action of many different genes. However, the genetic and molecular mechanisms underlying this process remain largely unknown. In this study, we isolated and characterized a novel rice male sterile mutant, defective pollen wall3 (dpw3), which displays smaller and paler anthers with aborted pollen grains. DPW3 encodes a novel membrane-associated alpha integrin-like protein conserved in land plants. DPW3 is ubiquitously expressed in anther developmental stages and its protein is localized to the plasma membrane, endoplasmic reticulum (ER) and Golgi. Anthers of dpw3 plants exhibited unbalanced anther cuticular profile, abnormal Ubisch bodies, disrupted callose deposition, defective pollen wall formation such as abnormal microspore plasma membrane undulation and defective primexine formation, resulting in pollen abortion and complete male sterility. Our findings revealed a novel and vital role of alpha integrin-like proteins in plant male reproduction.

Comprehensive analysis of the molecular characterization of GM rice G6H1 using a paired-end sequencing approach.

Molecular characterization of exogenous DNA integrations in host genome is a key aspect in risk assessment of bioengineered crops. However, gaining a clear understanding of the molecular characters of a bioengineered crop using conventional techniques remains a challenging task. Herein, we report the full molecular characterization of one new transgenic rice event G6H1 via a paired-end sequencing approach and bioinformatics analysis pipelines. Also, the molecular characterization reported was validated using conventional PCR, Sanger sequencing, and digital PCR. The results showed there is only one copy of the exogenous DNA inserted, which is located within chromosome 7 of the G6H1 genome. There is no other unintended integration of sequences from the transformation plasmid. These results indicated that the paired-end sequencing approach, combined with bioinformatics pipeline developed, is well suited to elucidate the molecular characteristics of bioengineered crops, and is efficient, low cost, and comprehensive.

Functional connectivity-based classification of autism and control using SVM-RFECV on rs-fMRI data.

Considering the unsatisfactory classification accuracy of autism due to unsuitable features selected in current studies, a functional connectivity (FC)-based algorithm for classifying autism and control using support vector machine-recursive feature elimination (SVM-RFE) is proposed in this paper. The goal is to find the optimal features based on FC and improve the classification accuracy on a large sample of data. We chose 35 regions of interest based on the social motivation hypothesis to construct the FC matrix and searched for informative features in the complex high-dimensional FC dataset by the SVM-RFE with a stratified-4-fold cross-validation strategy. The selected features were then entered into an SVM with a Gaussian kernel for classification. A total of 255 subjects with autism and 276 subjects with typical development from 10 sites were involved in the study. For the data of global sites, the proposed classification algorithm could identify the two groups with an accuracy of 90.60% (sensitivity 90.62%, specificity 90.58%). For the leave-one-site-out test, the proposed algorithm achieved a classification accuracy of 75.00%-95.23% for data from different sites. These promising results demonstrate that the proposed classification algorithm performs better than those in recent similar studies in that the importance of features can be measured accurately and only the most discriminative feature subset is selected.

The Rice Actin-Binding Protein RMD Regulates Light-Dependent Shoot Gravitropism.

Light and gravity are two key determinants in orientating plant stems for proper growth and development. The organization and dynamics of the actin cytoskeleton are essential for cell biology and critically regulated by actin-binding proteins. However, the role of actin cytoskeleton in shoot negative gravitropism remains controversial. In this work, we report that the actin-binding protein Rice Morphology Determinant (RMD) promotes reorganization of the actin cytoskeleton in rice (Oryza sativa) shoots. The changes in actin organization are associated with the ability of the rice shoots to respond to negative gravitropism. Here, light-grown rmd mutant shoots exhibited agravitropic phenotypes. By contrast, etiolated rmd shoots displayed normal negative shoot gravitropism. Furthermore, we show that RMD maintains an actin configuration that promotes statolith mobility in gravisensing endodermal cells, and for proper auxin distribution in light-grown, but not dark-grown, shoots. RMD gene expression is diurnally controlled and directly repressed by the phytochrome-interacting factor-like protein OsPIL16. Consequently, overexpression of OsPIL16 led to gravisensing and actin patterning defects that phenocopied the rmd mutant. Our findings outline a mechanism that links light signaling and gravity perception for straight shoot growth in rice.

Computer-aided diagnosis of school-aged children with ASD using full frequency bands and enhanced SAE: A multi-institution study.

Autism spectrum disorder (ASD) is a neurodevelopmental and network-level disorder mainly diagnosed in children. The aim of the current study was to develop a computer-aided diagnosis method with high accuracy to distinguish school-aged children (5-12 years) with ASD from those typically developing (TD). The current study used multi-institutional functional magnetic resonance imaging (fMRI) datasets of 198 school-aged participants from the Autism Brain Imaging Data Exchange II database and employed enhanced stacked auto-encoders to distinguish between school-aged children with ASD from those TD. In the current study, the average diagnostic accuracy was 96.26% (average sensitivity=98.03%; average specificity=93.62%); these results of classification were higher than that observed in previous studies using single or two frequency bands. The current study demonstrated that the proposed computer-aided diagnosis method may be used to distinguish between school-aged children with ASD from those TD. Attempts to use full frequency bands, deep learning based algorithm and multi-institutional fMRI datasets to distinguish between school-aged children with ASD from TD may be a key step towards clinical auxiliary diagnosis independent of sex, handedness, intellectual level or scanning parameters of fMRI data.

Inter-laboratory validation of visual loop-mediated isothermal amplification assays for GM contents screening.

Loop-mediated isothermal amplification (LAMP) has been widely used in many fields of molecular diagnostics, including detection of genetically modified organisms (GMOs). Herein, we report a collaborative ring trial validation of three established visual LAMP assays targeting three common GM elements, namely CaMV35S promoter, FMV35S promoter and NOS terminator, respectively. The high specificity of each assay was confirmed in different GM events analyses, and the sensitivity of each was determined to be 10, 10, and 50 haploid genome equivalents (HGEs) for CaMV35S promoter, FMV35S promoter, and NOS terminator, respectively. The probability of detection was also determined based on specificity and sensitivity data from 10 participating laboratories that returned correct results for the practical sample tests. These results demonstrate that the three visual LAMP assays are sensitive and time-saving, with high application potential for on-spot testing and routine screening of GMOs.

One Novel Multiple-Target Plasmid Reference Molecule Targeting Eight Genetically Modified Canola Events for Genetically Modified Canola Detection.

Multiple-target plasmid DNA reference materials have been generated and utilized as good substitutes of matrix-based reference materials in the analysis of genetically modified organisms (GMOs). Herein, we report the construction of one multiple-target plasmid reference molecule, pCAN, which harbors eight GM canola event-specific sequences (RF1, RF2, MS1, MS8, Topas 19/2, Oxy235, RT73, and T45) and a partial sequence of the canola endogenous reference gene PEP. The applicability of this plasmid reference material in qualitative and quantitative PCR assays of the eight GM canola events was evaluated, including the analysis of specificity, limit of detection (LOD), limit of quantification (LOQ), and performance of pCAN in the analysis of various canola samples, etc. The LODs are 15 copies for RF2, MS1, and RT73 assays using pCAN as the calibrator and 10 genome copies for the other events. The LOQ in each event-specific real-time PCR assay is 20 copies. In quantitative real-time PCR analysis, the PCR efficiencies of all event-specific and PEP assays are between 91% and 97%, and the squared regression coefficients (R2) are all higher than 0.99. The quantification bias values varied from 0.47% to 20.68% with relative standard deviation (RSD) from 1.06% to 24.61% in the quantification of simulated samples. Furthermore, 10 practical canola samples sampled from imported shipments in the port of Shanghai, China, were analyzed employing pCAN as the calibrator, and the results were comparable with those assays using commercial certified materials as the calibrator. Concluding from these results, we believe that this newly developed pCAN plasmid is one good candidate for being a plasmid DNA reference material in the detection and quantification of the eight GM canola events in routine analysis.

Development and application of a multi-targeting reference plasmid as calibrator for analysis of five genetically modified soybean events.

Reference materials are important in accurate analysis of genetically modified organism (GMO) contents in food/feeds, and development of novel reference plasmid is a new trend in the research of GMO reference materials. Herein, we constructed a novel multi-targeting plasmid, pSOY, which contained seven event-specific sequences of five GM soybeans (MON89788-5', A2704-12-3', A5547-127-3', DP356043-5', DP305423-3', A2704-12-5', and A5547-127-5') and sequence of soybean endogenous reference gene Lectin. We evaluated the specificity, limit of detection and quantification, and applicability of pSOY in both qualitative and quantitative PCR analyses. The limit of detection (LOD) was as low as 20 copies in qualitative PCR, and the limit of quantification (LOQ) in quantitative PCR was 10 copies. In quantitative real-time PCR analysis, the PCR efficiencies of all event-specific and Lectin assays were higher than 90%, and the squared regression coefficients (R(2)) were more than 0.999. The quantification bias varied from 0.21% to 19.29%, and the relative standard deviations were from 1.08% to 9.84% in simulated samples analysis. All the results demonstrated that the developed multi-targeting plasmid, pSOY, was a credible substitute of matrix reference materials, and could be used as a reliable reference calibrator in the identification and quantification of multiple GM soybean events.

Human acyl-CoA:cholesterol acyltransferase 2 gene expression in intestinal Caco-2 cells and in hepatocellular carcinoma.

Humans express two ACAT (acyl-CoA:cholesterol acyltransferase) genes, ACAT1 and ACAT2. ACAT1 is ubiquitously expressed, whereas ACAT2 is primarily expressed in intestinal mucosa and plays an important role in intestinal cholesterol absorption. To investigate the molecular mechanism(s) responsible for the tissue-specific expression of ACAT2, we identified five cis-elements within the human ACAT2 promoter, four for the intestinal-specific transcription factor CDX2 (caudal type homeobox transcription factor 2), and one for the transcription factor HNF1alpha (hepatocyte nuclear factor 1alpha). Results of luciferase reporter and electrophoretic mobility shift assays show that CDX2 and HNF1alpha exert a synergistic effect, enhancing the ACAT2 promoter activity through binding to these cis-elements. In undifferentiated Caco-2 cells, the ACAT2 expression is increased when exogenous CDX2 and/or HNF1alpha are expressed by co-transfection. In differentiated Caco-2 cells, the ACAT2 expression significantly decreases when the endogenous CDX2 or HNF1alpha expression is suppressed by using RNAi (RNA interference) technology. The expression levels of CDX2, HNF1alpha, and ACAT2 are all greatly increased when the Caco-2 cells differentiate to become intestinal-like cells. These results provide a molecular mechanism for the tissue-specific expression of ACAT2 in intestine. In normal adult human liver, CDX2 expression is not detectable and the ACAT2 expression is very low. In the hepatoma cell line HepG2 the CDX2 expression is elevated, accounting for its elevated ACAT2 expression. A high percentage (seven of fourteen) of liver samples from patients affected with hepatocellular carcinoma exhibited elevated ACAT2 expression. Thus, the elevated ACAT2 expression may serve as a new biomarker for certain form(s) of hepatocellular carcinoma.

Two human ACAT2 mRNA variants produced by alternative splicing and coding for novel isoenzymes.

Acyl coenzyme A:cholesterol acyltransferase 2 (ACAT2) plays an important role in cholesterol absorption. Human ACAT2 is highly expressed in small intestine and fetal liver, but its expression is greatly diminished in adult liver. The full-length human ACAT2 mRNA encodes a protein, designated ACAT2a, with 522 amino acids. We have previously reported the organization of the human ACAT2 gene and the differentiation-dependent promoter activity in intestinal Caco-2 cells. In the current work, two human ACAT2 mRNA variants produced by alternative splicing are cloned and predicted to encode two novel ACAT2 isoforms, named ACAT2b and ACAT2c, with 502 and 379 amino acids, respectively. These mRNA variants differ from ACAT2a mRNA by lack of the exon 4 (ACAT2b mRNA) and exons 4-5 plus 8-9-10 (ACAT2c mRNA). Significantly, comparable amounts of the alternatively spliced ACAT2 mRNA variants were detected by RT-PCR, and Western blot analysis confirmed the presence of their corresponding proteins in human liver and intestinecells. Furthermore, phosphorylation and enzymatic activity analyses demonstrated that the novel isoenzymes ACAT2b and ACAT2c lacked the phosphorylatable site SLLD, and their enzymatic activities reduced to 25%-35% of that of ACAT2a. These evidences indicate that alternative splicing produces two human ACAT2 mRNA variants that encode the novel ACAT2 isoenzymes. Our findings might help to understand the regulation of the ACAT2 gene expression under certain physiological and pathological conditions.

Preparation of an anti-Cdx-2 antibody for analysis of different species Cdx-2 binding to acat2 promoter.

The homeodomain protein, Cdx-2, as transcription factor has been implicated in the transcriptional regulation of genes expressed in small intestine and the process of tumorgenesis. In current work, a conserved mouse Cdx-2 domain (mCdx-2D) coded by its cDNA fragment, which was amplified and cloned into the expression vector pGEX-4T1, was expressed as a fusion protein with GST (GST-mCd x-2D) and purified by one step of affinity chromatography. A polyclonal antibody against Cdx-2 was raised by using the recombinant fusion protein GST-mCdx-2D as antigen and was fractionated from the rabbit anti-serum. Western blot and EMSA (electrophoretic mobility shift assay) demonstrate that the natural and denatured Cdx-2s from different species (mouse and human) can be detected by the prepared anti-Cdx-2 antibody. Most notably, we found that the Cdx-2 in human intestine cell line Caco-2 is expressed in a differentiation-dependent manner and can efficiently bind to the mouse and human acat2 (acyl-coenzyme A: cholesterol acyltransferase 2) promoter regions, suggesting that the transcriptional factor Cdx-2 may play a role in regulating the acat2 expression in the intestinal cells.