RESUMEN
The success of graph neural networks (GNNs) in graph-based web mining highly relies on abundant human-annotated data, which is laborious to obtain in practice. When only a few labeled nodes are available, how to improve their robustness is key to achieving replicable and sustainable graph semi-supervised learning. Though self-training is powerful for semi-supervised learning, its application on graph-structured data may fail because 1) larger receptive fields are not leveraged to capture long-range node interactions, which exacerbates the difficulty of propagating feature-label patterns from labeled nodes to unlabeled nodes and 2) limited labeled data makes it challenging to learn well-separated decision boundaries for different node classes without explicitly capturing the underlying semantic structure. To address the challenges of capturing informative structural and semantic knowledge, we propose a new graph data augmentation framework, augmented graph self-training (AGST), which is built with two new (i.e., structural and semantic) augmentation modules on top of a decoupled GST backbone. In this work, we investigate whether this novel framework can learn a robust graph predictive model under the low-data context. We conduct comprehensive evaluations on semi-supervised node classification under different scenarios of limited labeled-node data. The experimental results demonstrate the unique contributions of the novel data augmentation framework for node classification with few labeled data.
RESUMEN
Anomaly detection on attributed graphs has received increasing research attention lately due to the broad applications in various high-impact domains, such as cybersecurity, finance, and healthcare. Heretofore, most of the existing efforts are predominately performed in an unsupervised manner due to the expensive cost of acquiring anomaly labels, especially for newly formed domains. How to leverage the invaluable auxiliary information from a labeled attributed graph to facilitate the anomaly detection in the unlabeled attributed graph is seldom investigated. In this study, we aim to tackle the problem of cross-domain graph anomaly detection with domain adaptation. However, this task remains nontrivial mainly due to: 1) the data heterogeneity including both the topological structure and nodal attributes in an attributed graph and 2) the complexity of capturing both invariant and specific anomalies on the target domain graph. To tackle these challenges, we propose a novel framework COMMANDER for cross-domain anomaly detection on attributed graphs. Specifically, COMMANDER first compresses the two attributed graphs from different domains to low-dimensional space via a graph attentive encoder. In addition, we utilize a domain discriminator and an anomaly classifier to detect anomalies that appear across networks from different domains. In order to further detect the anomalies that merely appear in the target network, we develop an attribute decoder to provide additional signals for assessing node abnormality. Extensive experiments on various real-world cross-domain graph datasets demonstrate the efficacy of our approach.
RESUMEN
The aim of the present study was to investigate the effect of the histone H3K9 demethylase inhibitor, IOX1, on the mechanism of hepatic fibrosis in TGF-ß-induced human hepatic stellate LX-2 cells. Cellular proliferation, apoptosis, histone H3K9 dimethylation (H3K9me2), protein expression of extracellular matrix (ECM)-related proteins α-smooth muscle actin (SMA), type I collagen (Col I), MMP-1 and TIMP-1 were measured. H3K9me2 levels in the promoter region of ECM-related genes were detected by real-time cell analysis (RTCA), flow cytometry, western blotting and chromatin immunoprecipitation (ChIP) in LX-2 cells. IOX1 significantly inhibited cell proliferation and the IC50 of IOX1 was 100 µM in cells treated with IOX1 for 48 h. IOX1 significantly induced apoptosis in LX-2 cells in a concentration-dependent manner. In addition, different concentration of IOX1 increased the level of H3K9me2 and downregulated the expression of α-SMA, Col I, MMP-1 and TIMP-1 in TGF-ß-induced LX-2 cells. ChIP measurements indicated that H3K9me2 levels in the promotor region of the corresponding genes were increased in TGF-ß-induced LX-2 cells. IOX1 may elevate H3K9me2 in the promotor region of Col I, MMP-1, and TIMP-1 genes to regulate α-SMA, Col I, MMP-1 and TIMP-1 protein expression to induce cell apoptosis, inhibit LX-2 cell proliferation and oppose hepatic fibrotic activity.
RESUMEN
OBJECTIVE: To investigate the morphological characters of the central nervous system in mature larvae, mature pupae and newly emerged adults of Simulium (Wilhelmia) xingyiense. METHODS: From August to November 2009, the blackfly larvae were collected from the rivulets nearby Niujiao Island in Huaxi of Guiyang City. The mature larvae of S. (Wi.) xingyiense were confirmed based on the diagnostic characteristics of gill spots, postgenal cleft, and rectal gill. The mature pupae were obtained from the rivulets of Da'ao Town and Qingyan Town in Guiyang in March 2011, which were selected according to the characteristics of cocoon and respiratory filaments. Nervous system of the larvae, pupae and newly emerged adults was observed under the light microscope with HE-stained paraffin sections. RESULTS: The central nervous system was composed of brain, subesophageal ganglion and ventral nerve cord. The brain of the larva was divided into two narrowly interconnected egg-shaped lobes. Ventral nerve-cord of the larva consisted of three pairs of thoracic ganglia and eight pairs of abdominal ganglia. The brain of the pupa and adult was composed of protocerebuim, deutocerebrum, and tritocerebrum. The ventral nerve cord of the pupa and adult was similar to that of larva. From outside to inside, the structure characters of the brain and ganglia were similar with nerve sheath, neurocyte and neuropile. The neuropile of protocerebrum contained a pair of mushroom bodies, a central complex and a pair of accessory lobes. The optic lobe was composed of medulla interna, medulla externa and lamina ganglionaris. The deutocerebrum consisted of the antennal lobe and the dorsal lobe. The tritocerebrum connected to the subesophageal ganglion by perioperative esophageal nerve. CONCLUSION: The central nervous system of S. (Wi.) xingyiense is similar to other simuliid blackilies. There is a difference in the number of abdominal ganglion of the mature larva.