Due to both natural and anthropogenic forces, the south west part of the Ganges-Brahmaputra coastal area is facing diverse problems such as waterlogging, salinity, and loss of biodiversity. In order to address these challenges, local people have identified 'tidal river management (TRM)' as a comprehensive approach for sustainably managing this part of the GangesBrahmaputra Basin. However, due to institutional limitations, mismanagement and social conflicts, application of the TRM approach is not straightforward. In order to identify existing implementation barriers and to effectively apply the TRM approach, a transdisciplinary approach is examined for its potential to inform the re-shaping of TRM governing values and actions. It is argued that a thorough application of a transdisciplinary framework is essential, supported by the active involvement of key agencies and local stakeholders. The proposed transdisciplinary framework can potentially be applied to TRM projects for solving waterlogging and associated problems in order to achieve greater sustainability of the area.

Deep learning with Convolutional Neural Networks has shown great promise in image-based classification and enhancement but is often unsuitable for predictive modeling using features without spatial correlations. We present a feature representation approach termed REFINED (REpresentation of Features as Images with NEighborhood Dependencies) to arrange highdimensional vectors in a compact image form conducible for CNN-based deep learning. We consider the similarities between features to generate a concise feature map in the form of a two-dimensional image by minimizing the pairwise distance values following a Bayesian Metric Multidimensional Scaling Approach. We hypothesize that this approach enables embedded feature extraction and, integrated with CNN-based deep learning, can boost the predictive accuracy. We illustrate the superior predictive capabilities of the proposed framework as compared to state-of-the-art methodologies in drug sensitivity prediction scenarios using synthetic datasets, drug chemical descriptors as predictors from NCI60, and both transcriptomic information and drug descriptors as predictors from GDSC.

No abstract

No abstract

Integrated Water Resources Management (IWRM) is considered as a practical approach in solving water-related problems, which are socio-ecologically complex in nature. Bangladesh has also embraced the IWRM approach against its earlier attempt to flood control. In this paper, we evaluate the current status of IWRM in Bangladesh through the lens of policy shifts, institutional transitions and project transformations using seven key dimensions of IWRM. Looking at IWRM from such perspectives is lacking in current literature. A thorough review of policy shifts suggests that all the key dimensions of IWRM are "highly reflected" in the current policy documents. The dimension of "integrated management" is "highly reflected" in both institutional transition and project-level transformation. Most other dimensions are also recognised at both institutional and project levels. However, such reflections gradually weaken as we move from policies to institutions to projects. Despite catchment being considered as a spatial unit of water management at both institutional and project levels, transboundary basin planning is yet to be accomplished. The participation of local people is highly promoted in various recent projects. However, equity and social issues have received less attention at project level, although it has significant potential for supporting some of the key determinants of adaptive capacity. Thus, the IWRM dimensions are in general reflected in recent policies, institutional reforms and project formulation in Bangladesh. However, to solve the complex water-problems, basin scale management through transboundary cooperation and equity and social issues need to be implemented at institutional and project levels.