Spatiotemporal variations and influential factors affecting reproductive dynamics of Pacific saury in the Northwest Pacific during fishing season.

This study examined the spatial and monthly variations in reproductive dynamics of Pacific saury collected in the high sea (40°N-49°N, 149°E-168°E) of the northwestern Pacific during the fishing season of the Taiwanese stick-held dip-net fisheries (July-November) from 2018 to 2019. Generalized additive mixed-effects models (GAMMs) were applied to explore the relationship between the probability of maturing and spawning occurrence (PMOS ) and the explanatory variables (month, body length, sea-surface temperature [SST] as fixed effects, and fishing set as a random effect) for the age 0 and age 1 fish, respectively. In addition, the impact of the geographical difference in growth rates of age 0 fish, quantified as the radius of the otolith annual ring, on the PMOS of age 1 fish was explored in the GAMMs. Results showed that the mean values of the condition factor (CF) varied by months, and higher mean CFs were observed north of 45°N. However, the mean values of the gonado-somatic index did not show a clear spatiotemporal pattern. The spatial distribution of the ovarian maturation states revealed that the spawning ground of Pacific saury during July-November was broader than that previously known and had partly overlapped with the high-sea fishing ground. The best GAMM showed that the predicted PMOS of the age 0 fish were generally low (mean = 19%, standard deviation [SD] = 10%) and tended to increase with increasing body length. In contrast, the predicted PMOS were relatively higher (mean = 37%, SD = 10%) for the age 1 fish and increased with increasing SST and decreasing latitude. Two candidate GAMMs, which are equally supported as the best model (ΔAIC < 2), provide evidence that a higher growth rate of age 0 fish may result in higher reproductive activity in age 1 fish. This is one of few studies focusing on the reproductive dynamics of Pacific saury during the fishing season; current limitations, future directions, and conservation implications were discussed.

Characteristic-preserving Latent Space for Unpaired Cross-domain Translation of 3D Point Clouds.

This paper aims at unpaired shape-to-shape transformation for 3D point clouds, for instance, turning a chair to its table counterpart. Recent work for 3D shape transfer or deformation highly relies on paired inputs or specific correspondences. However, it is usually not feasible to assign precise correspondences or prepare paired data from two domains. A few methods start to study unpaired learning, but the characteristics of a source model may not be preserved after transformation. To overcome the difficulty of unpaired learning for transformation, we propose alternately training the autoencoder and translators to construct shape-aware latent space. This latent space based on novel loss functions enables our translators to transform 3D point clouds across domains and maintain the consistency of shape characteristics. We also crafted a test dataset to objectively evaluate the performance of point-cloud translation. The experiments demonstrate that our framework can construct high-quality models and retain more shape characteristics during cross-domain translation compared to the state-of-the-art methods. Moreover, we also present shape editing applications with our proposed latent space, including shape-style mixing and shape-type shifting, which do not require retraining a model.

Targeting cancer stemness mediated by BMI1 and MCL1 for non-small cell lung cancer treatment.

Lung cancer is the leading cause of cancer-associated death, with a global 5-year survival rate <20%. Early metastasis and recurrence remain major challenges for lung cancer treatment. The stemness property of cancer cells has been suggested to play a key role in cancer plasticity, metastasis and drug-resistance, and is a potential target for drug development. In this study, we found that in non-small cell lung cancer (NSCLC), BMI1 and MCL1 play crucial roles of cancer stemness including invasion, chemo-resistance and tumour initiation. JNK signalling serves as a link between oncogenic pathway or genotoxicity to cancer stemness. The activation of JNK, either by mutant EGFR or chemotherapy agent, stabilized BMI1 and MCL1 proteins through suppressing the expression of E3-ubiquitin ligase HUWE1. In lung cancer patient samples, high level of BMI1 is correlated with poor survival, and the expression of BMI1 is positively correlated with MCL1. A novel small-molecule, BI-44, was developed, which effectively suppressed BMI1/MCL1 expressions and inhibited tumour formation and progression in preclinical models. Targeting cancer stemness mediated by BMI1/MCL1 with BI-44 provides the basis for a new therapeutic approach in NSCLC treatment.