Autoencoder frameworks have received attention for graph embedding, particularly those utilizing generative adversarial networks (GAN). However, GAN-based frameworks do not fully utilize the original graph information and lack stable updates in the GAN component. To bridge this gap, we propose a dual-adversarial framework for graph embedding that expands mutual information (MI) in positive and negative samples for adversarial training using GAN. We further improve model performance by incorporating reinforcement learning ideas. Our framework includes two variants: a pessimistic adversarial graph autoencoder (PAGAE), and a pessimistic adversarial graph autoencoder with PO loss (PAGAEPO). Essentially, we present a pessimistic module to negative sample generator to boost original discriminator, thereby reinforcing the generator’s ability. Additionally, we designed a PO loss function on discriminator to stabilize the learning process and it will further improve the ability of model. Experimental results show that our models are competitive with state-of-the-art GAEs on benchmark datasets.

Consumers and breeders value strawberry fruit with high sugar content and sweetness. However, many factors negatively affect this trait. bZIP-S1 subfamily was closely correlated with the carbohydrate metabolism. Moreover, it can regulate plant development and growth, and respond to various stress. In the present study, a total of four bZIP-S1 members were identified in strawberry. Quantitative RT-PCR analysis showed a member of this subfamily, designated as bZIP11, most highly expressed in different tissues of strawberry. Especially, it had a significant transcript accumulation during the fruit development and ripening. Constitutive expression of FvbZIP11 in tomato caused severe plant growth retardation which was accompanied by wrinkled and curly leaves. The leaf yellowish and chlorotic patches were reflected in decrease of chlorophyll. Meanwhile, epidermal cells were smaller in leaves of transgenic plant. In addition, the total soluble solid and soluble sugar content of the fruit was significantly higher than WT during the fruit ripening. Taken together, the heterologous overexpression of strawberry bZIP11 induced sugar accumulation and hindered plant growth in tomato, which provided more information for improving strawberry fruit quality.

Neurofibromatosis type 2 is an autosomal dominant multiple neoplasia syndrome and is usually caused by mutations in the neurofibromin 2 (NF2) gene, which encodes a tumor suppressor and initiates the Hippo pathway. However, the mechanism by which NF2 functions in the Hippo pathway isn't fully understood. Here we identified a NF2 c.770-784del mutation from a neurofibromatosis type 2 family. MD simulations showed that this mutation significantly changed the structure of the F3 module of the NF2-FERM domain. Functional assays indicated that the NF2 c.770-784del variant formed LLPS in the cytoplasm with LATS to restrain LATS plasma membrane localization and inactivated the Hippo pathway. Besides, this deletion partly caused a skipping of exon 8 and reduced the protein level of NF2, collectively promoting proliferation and tumorigenesis of meningeal cells. We identified an unrecognized mechanism of LLPS and splicing skipping for the NF2-induced Hippo pathway, which provided new insight into the pathogenesis of neurofibromatosis type 2.