DeepPN: a deep parallel neural network based on convolutional neural network and graph convolutional network for predicting RNA-protein binding sites.

BACKGROUND: Addressing the laborious nature of traditional biological experiments by using an efficient computational approach to analyze RNA-binding proteins (RBPs) binding sites has always been a challenging task. RBPs play a vital role in post-transcriptional control. Identification of RBPs binding sites is a key step for the anatomy of the essential mechanism of gene regulation by controlling splicing, stability, localization and translation. Traditional methods for detecting RBPs binding sites are time-consuming and computationally-intensive. Recently, the computational method has been incorporated in researches of RBPs. Nevertheless, lots of them not only rely on the sequence data of RNA but also need additional data, for example the secondary structural data of RNA, to improve the performance of prediction, which needs the pre-work to prepare the learnable representation of structural data. RESULTS: To reduce the dependency of those pre-work, in this paper, we introduce DeepPN, a deep parallel neural network that is constructed with a convolutional neural network (CNN) and graph convolutional network (GCN) for detecting RBPs binding sites. It includes a two-layer CNN and GCN in parallel to extract the hidden features, followed by a fully connected layer to make the prediction. DeepPN discriminates the RBP binding sites on learnable representation of RNA sequences, which only uses the sequence data without using other data, for example the secondary or tertiary structure data of RNA. DeepPN is evaluated on 24 datasets of RBPs binding sites with other state-of-the-art methods. The results show that the performance of DeepPN is comparable to the published methods. CONCLUSION: The experimental results show that DeepPN can effectively capture potential hidden features in RBPs and use these features for effective prediction of binding sites.

Coastal Image Classification and Pattern Recognition: Tairua Beach, New Zealand.

The study of coastal processes is critical for the protection and development of beach amenities, infrastructure, and properties. Many studies of beach evolution rely on data collected using remote sensing and show that beach evolution can be characterized by a finite number of "beach states". However, due to practical constraints, long-term data displaying all beach states are rare. Additionally, when the dataset is available, the accuracy of the classification is not entirely objective since it depends on the operator. To address this problem, we collected hourly coastal images and corresponding tidal data for more than 20 years (November 1998-August 2019). We classified the images into eight categories according to the classic beach state classification, defined as (1) reflective, (2) incident scaled bar, (3) non-rhythmic, attached bar, (4) attached rhythmic bar, (5) offshore rhythmic bar, (6) non-rhythmic, 3-D bar, (7) infragravity scaled 2-D bar, (8) dissipative. We developed a classification model based on convolutional neural networks (CNN). After image pre-processing with data enhancement, we compared different CNN models. The improved ResNext obtained the best and most stable classification with F1-score of 90.41% and good generalization ability. The classification results of the whole dataset were transformed into time series data. MDLats algorithms were used to find frequent temporal patterns in morphology changes. Combining the pattern of coastal morphology change and the corresponding tidal data, we also analyzed the characteristics of beach morphology and the changes in morphodynamic states.

Spatially explicit models for exploring COVID-19 lockdown strategies.

This article describes two spatially explicit models created to allow experimentation with different societal responses to the COVID-19 pandemic. We outline the work to date on modeling spatially explicit infective diseases and show that there are gaps that remain important to fill. We demonstrate how geographical regions, rather than a single, national approach, are likely to lead to better outcomes for the population. We provide a full account of how our models function, and how they can be used to explore many different aspects of contagion, including: experimenting with different lockdown measures, with connectivity between places, with the tracing of disease clusters, and the use of improved contact tracing and isolation. We provide comprehensive results showing the use of these models in given scenarios, and conclude that explicitly regionalized models for mitigation provide significant advantages over a "one-size-fits-all" approach. We have made our models, and their data, publicly available for others to use in their own locales, with the hope of providing the tools needed for geographers to have a voice during this difficult time.

Modelling the cost differential between healthy and current diets: the New Zealand case study.

BACKGROUND: Evidence on whether healthy diets are more expensive than current diets is mixed due to lack of robust methodology. The aim of this study was to develop a novel methodology to model the cost differential between healthy and current diets and apply it in New Zealand. METHODS: Prices of common foods were collected from 15 supermarkets, 15 fruit/vegetable stores and from the Food Price Index. The distribution of the cost of two-weekly healthy and current household diets was modelled using a list of commonly consumed foods, a set of min and max quantity/serves constraints for each, and food group and nutrient intakes based on dietary guidelines (healthy diets) or nutrition survey data (current diets). The cost differential between healthy and current diets was modelled for several diet, prices and policy scenarios. Acceptability of resulting meal plans was validated. RESULTS: The average cost of healthy household diets was $27 more expensive than the average cost of current diets, but 25.8% of healthy diets were cheaper than the average cost of current diets. This cost differential could be reduced if fruits and vegetables became exempt from Goods and Services Tax. Healthy diets were cheaper with an allowance for discretionary foods and more expensive when including takeaway meals. For Maori and Pacific households, healthy diets were on average $40 and $60 cheaper than current diets due to large energy intakes. Discretionary foods and takeaway meals contributed 30-40% to the average cost of current diets. CONCLUSION: Healthy New Zealand diets were on average more expensive than current diets, but one-quarter of healthy diets were cheaper than the average cost of current diets. The impact of diet composition, types of prices and policies on the cost differential was substantial. The methodology can be used in other countries to monitor the cost differential between healthy and current household diets.

Biodiversity data should be published, cited, and peer reviewed.

Concerns over data quality impede the use of public biodiversity databases and subsequent benefits to society. Data publication could follow the well-established publication process: with automated quality checks, peer review, and editorial decisions. This would improve data accuracy, reduce the need for users to 'clean' the data, and might increase data use. Authors and editors would get due credit for a peer-reviewed (data) publication through use and citation metrics. Adopting standards related to data citation, accessibility, metadata, and quality control would facilitate integration of data across data sets. Here, we propose a staged publication process involving editorial and technical quality controls, of which the final (and optional) stage includes peer review, the most meritorious publication standard in science.

Plant pathogen culture collections: it takes a village to preserve these resources vital to the advancement of agricultural security and plant pathology.

ABSTRACT Plant pathogen culture collections are essential resources in our fight against plant disease and for connecting discoveries of the present with established knowledge of the past. However, available infrastructure in support of culture collections is in serious need of improvement, and we continually face the risk of losing many of these collections. As novel and reemerging plant pathogens threaten agriculture, their timely identification and monitoring depends on rapid access to cultures representing the known diversity of plant pathogens along with genotypic, phenotypic, and epidemiological data associated with them. Archiving such data in a format that can be easily accessed and searched is essential for rapid assessment of potential risk and can help track the change and movement of pathogens. The underexplored pathogen diversity in nature further underscores the importance of cataloguing pathogen cultures. Realizing the potential of pathogen genomics as a foundation for developing effective disease control also hinges on how effectively we use the sequenced isolate as a reference to understand the genetic and phenotypic diversity within a pathogen species. In this letter, we propose a number of measures for improving pathogen culture collections.

Multivariate Analysis and Geovisualization with an Integrated Geographic Knowledge Discovery Approach.

The discovery, interpretation, and presentation of multivariate spatial patterns are important for scientific understanding of complex geographic problems. This research integrates computational, visual, and cartographic methods together to detect and visualize multivariate spatial patterns. The integrated approach is able to: (1) perform multivariate analysis, dimensional reduction, and data reduction (summarizing a large number of input data items in a moderate number of clusters) with the Self-Organizing Map (SOM); (2) encode the SOM result with a systematically designed color scheme; (3) visualize the multivariate patterns with a modified Parallel Coordinate Plot (PCP) display and a geographic map (GeoMap); and (4) support human interactions to explore and examine patterns. The research shows that such "mixed initiative" methods (computational and visual) can mitigate each other's weakness and collaboratively discover complex patterns in large geographic datasets, in an effective and efficient way.

Visualization for constructing and sharing geo-scientific concepts.

Representations of scientific knowledge must reflect the dynamic nature of knowledge construction and the evolving networks of relations between scientific concepts. In this article, we describe initial work toward dynamic, visual methods and tools that support the construction, communication, revision, and application of scientific knowledge. Specifically, we focus on tools to capture and explore the concepts that underlie collaborative science activities, with examples drawn from the domain of human-environment interaction. These tools help individual researchers describe the process of knowledge construction while enabling teams of collaborators to synthesize common concepts. Our visualization approach links geographic visualization techniques with concept-mapping tools and allows the knowledge structures that result to be shared through a Web portal that helps scientists work collectively to advance their understanding. Our integration of geovisualization and knowledge representation methods emphasizes the process through which abstract concepts can be contextualized by the data, methods, people, and perspectives that produced them. This contextualization is a critical component of a knowledge structure, without which much of the meaning that guides the sharing of concepts is lost. By using the tools we describe here, human-environment scientists are given a visual means to build concepts from data (individually and collectively) and to connect these concepts to each other at appropriate levels of abstraction.