Search Results (4)

Analysis of data with a censored survival response and high-dimensional omics measurements is now common. Most of the existing analyses are based on specific (semi)parametric models, in particular the Cox model. Such analyses may be limited by not having sufficient flexibility, for example, in accommodating nonlinearity. For categorical and continuous responses, neural networks (NNs) have provided a highly competitive alternative. Comparatively, NNs for censored survival data remain limited. Omics measurements are usually high-dimensional, and only a small subset is expected to be survival-associated. As such, regularized estimation and selection are needed. In the existing NN studies, this is usually achieved via penalization. In this article, we propose adopting the threshold gradient descent regularization (TGDR) technique, which has competitive performance (for example, when compared to penalization) and unique advantages in regression analysis, but has not been adopted with NNs. The TGDR-based NN has a highly sensible formulation and an architecture different from the unregularized and penalization-based ones. Simulations show its satisfactory performance. Its practical effectiveness is further established via the analysis of two cancer omics datasets. Overall, this study can provide a practical and useful new way in the NN paradigm for survival analysis with high-dimensional omics measurements. Full article

The possible negative impacts of flow regulation on riparian zone conditions can be observed due to the disruption of the natural flow regime in reservoirs. In spite of considerable literature on the qualitative effects of external disturbances on riparian health indicators (RHIs), quantitative evaluations of such changes induced by pressure are rare in the literature. Our study evaluated the effects of pressure indicators on the RHIs, and the responses of RHIs relevant to the riparian zones of the Three Gorges Dam Reservoir (TGDR), China, by using the field-based approach. This paper is a component of a large project—rapid appraisal of riparian condition for the TGDR, China. The analysis has compared pressures (13 indicators) and RHIs (27 indicators) determined from the transects (259) identified throughout the TGDR (within 15 counties) by categorizing into upstream, midstream, and downstream. By using basic statistical techniques (Kruskal-Wallis tests and Pearson’s correlation), pressure indicators were found to significantly differently influence RHIs for the categorized three sections of the riparian zones of the TGDR. The correlation analysis confirmed that the pressure indicators correlated (range of r = −0.496–0.971) with the RHIs (enlisted as habitat, plant cover, regeneration, erosion, and exotic parameters). Moreover, pressure indicators were found to have a highly significant influence on erosion and habitat parameters, but moderate effects on plant cover, exotic and regeneration parameters. In addition, the highest relative effect of the pressure indicators was detected in the upstream transects, whereas the lowest was in the downstream transects. Agglomerative Hierarchical Cluster analysis also confirmed the substantial dissimilarity in the upstream transects, whereas significant similarities were identified between midstream and downstream transects. These results may be particularly important in the planning stages, to help administrators and planners form better priorities and treatments for reach-scale conservation and restoration of wide-ranging riparian zones. Full article

Soil enzymes and microbes are central to the decomposition of plant and microbial detritus, and play important roles in carbon, nitrogen, and phosphorus biogeochemistry cycling at the ecosystem level. In the present study, we characterized the soil enzyme activity and microbial biomass in revegetated (with Taxodium distichum (L.) Rich. and Cynodon dactylon (L.) Pers.) versus unplanted soil in the riparian zone of the Three Gorges Dam Reservoir (TGDR), in order to quantify the effect of revegetation on the edaphic microenvironment after water flooding in situ. After revegetation, the soil physical and chemical properties in revegetated soil showed significant differences to those in unplanted soil. The microbial biomass carbon and phosphorus in soils of T. distichum were significantly higher than those in C. dactylon and unplanted soils, respectively. The microbial biomass nitrogen in revegetated T. distichum and C. dactylon soils was significantly increased by 273% and 203%, respectively. The enzyme activities of T. distichum and C. dactylon soils displayed no significant difference between each other, but exhibited a great increase compared to those of the unplanted soil. Elements ratio (except C/N (S)) did not vary significantly between T. distichum and C. dactylon soils; meanwhile, a strong community-level elemental homeostasis in the revegetated soils was found. The correlation analyses demonstrated that only microbial biomass carbon and phosphorus had a significantly positive relationship with soil enzyme activities. After revegetation, both soil enzyme activities and microbial biomasses were relatively stable in the T. distichum and C. dactylon soils, with the wooded soil being more superior. The higher enzyme activities and microbial biomasses demonstrate the C, N, and P cycling and the maintenance of soil quality in the riparian zone of the TGDR. Full article

Submergence-tolerant trees are essential for vegetation restoration of the hydro-fluctuation zone of the Three Gorges Dam Reservoir (TGDR) area. Thus, it is of great significance to select the submergence-tolerant plant species by conducting in situ studies. To restore degraded riparian vegetation under the circumstances of dynamic impoundment of the TGDR, Salix matsudana Koidz., a flooding-tolerant native tree species, was introduced to conduct an in situ practical study to test its performance in re-vegetating and restoring the hydro-fluctuation zone of the TGDR. Effects of periodic moderate submergence (MS) and deep submergence (DS) on photosynthesis and growth of Salix matsudana Koidz. were investigated after three water cycles compared to a control (i.e., shallow submergence, abbreviated as SS) in order to specifically assess its application prospects in vegetation restoration under such extreme environment. Results showed that net photosynthetic rate (P_n), intrinsic water use efficiency (WUEi) and limiting value of stomata (Ls) of S. matsudana were significantly reduced in DS. However, pigment content had no significant change in all submergence treatments. Diameter at breast height (DBH) and tree height of S. matsudana were significantly decreased in both MS and DS when compared to that of SS, respectively. In contrast, the primary branch number of S. matsudana was significantly increased as submergence increased. In addition, relative diameter and height growth rates of S. matsudana were also reduced under submergence. Considering the sustained growth of this species, S. matsudana saplings are tolerant to long-term periodic submergence and can be applied to the vegetative restoration of the hydro-fluctuation zone of the TGDR region. Full article