GRA: Graph Representation Alignment for Semi-Supervised Action Recognition.

Graph convolutional networks (GCNs) have emerged as a powerful tool for action recognition, leveraging skeletal graphs to encapsulate human motion. Despite their efficacy, a significant challenge remains the dependency on huge labeled datasets. Acquiring such datasets is often prohibitive, and the frequent occurrence of incomplete skeleton data, typified by absent joints and frames, complicates the testing phase. To tackle these issues, we present graph representation alignment (GRA), a novel approach with two main contributions: 1) a self-training (ST) paradigm that substantially reduces the need for labeled data by generating high-quality pseudo-labels, ensuring model stability even with minimal labeled inputs and 2) a representation alignment (RA) technique that utilizes consistency regularization to effectively reduce the impact of missing data components. Our extensive evaluations on the NTU RGB+D and Northwestern-UCLA (N-UCLA) benchmarks demonstrate that GRA not only improves GCN performance in data-constrained environments but also retains impressive performance in the face of data incompleteness.

Proteomics approach reveals mechanism underlying susceptibility of loquat fruit to sunburn during color changing period.

The objective of this work was to investigate why loquat fruit peels are more sensitive to high temperature and strong sunlight, making them highly susceptible to sunburn, during the color changing period (CCP). Two dimensional gel electrophoresis (2-DE) of the fruit peel proteins was performed over three developmental periods, namely green fruit period (GFP), color changing period and yellow ripening period (YRP). Fifty-five protein spots with at least 2-fold differences in abundance were successfully identified by MALDI-TOF-TOF/MS. The identified proteins were divided into categories related to heat-shock response, stress response and defense, energy metabolism, photosynthesis and protein biosynthesis. The results showed that expression of proteins related to anaerobic respiration and photorespiration were increased while the proteins related to ROS scavenging, polyamine biosynthesis, defense pathogens and photosynthesis were decreased during CCP under heat stress. Our findings provide new insights into the molecular mechanism of loquat fruit susceptible to sunburn during CCP.

Genome-wide identification, classification and analysis of heat shock transcription factor family in maize.

BACKGROUND: Heat shock response in eukaryotes is transcriptionally regulated by conserved heat shock transcription factors (Hsfs). Hsf genes are represented by a large multigene family in plants and investigation of the Hsf gene family will serve to elucidate the mechanisms by which plants respond to stress. In recent years, reports of genome-wide structural and evolutionary analysis of the entire Hsf gene family have been generated in two model plant systems, Arabidopsis and rice. Maize, an important cereal crop, has represented a model plant for genetics and evolutionary research. Although some Hsf genes have been characterized in maize, analysis of the entire Hsf gene family were not completed following Maize (B73) Genome Sequencing Project. RESULTS: A genome-wide analysis was carried out in the present study to identify all Hsfs maize genes. Due to the availability of complete maize genome sequences, 25 nonredundant Hsf genes, named ZmHsfs were identified. Chromosomal location, protein domain and motif organization of ZmHsfs were analyzed in maize genome. The phylogenetic relationships, gene duplications and expression profiles of ZmHsf genes were also presented in this study. Twenty-five ZmHsfs were classified into three major classes (class A, B, and C) according to their structural characteristics and phylogenetic comparisons, and class A was further subdivided into 10 subclasses. Moreover, phylogenetic analysis indicated that the orthologs from the three species (maize, Arabidopsis and rice) were distributed in all three classes, it also revealed diverse Hsf gene family expression patterns in classes and subclasses. Chromosomal/segmental duplications played a key role in Hsf gene family expansion in maize by investigation of gene duplication events. Furthermore, the transcripts of 25 ZmHsf genes were detected in the leaves by heat shock using quantitative real-time PCR. The result demonstrated that ZmHsf genes exhibit different expression levels in heat stress treatment. CONCLUSIONS: Overall, data obtained from our investigation contributes to a better understanding of the complexity of the maize Hsf gene family and provides the first step towards directing future experimentation designed to perform systematic analysis of the functions of the Hsf gene family.