Find research datasets worth reusing

opencc-by-4.0Sep 2021View details →

N2 and N2O concentrations and fluxes for the six large river networks in China

This dataset is a collation of 4-year direct measurement values of N2 and  N2O concentrations and fluxes for the six large river networks in China, along with information on location, hydrological, physical, and chemical conditions of the 56 sampling sites. These data will be very valuable for balancing the regional and global N budget. In addition, the Monte Carlo simulation code for upscaling and analyzing uncertainties are presented here.

OMEN data set: Tokheim pancancer data set with genome-wide interaction network

OMEN is a Network-based Driver Gene Identification method that exploits Mutual Exclusivity. This repository stores the data used in a pancancer experiment showcasing this method. It consists of - [tokheim_pancancer_somatic_CADD.pl] A file containing CADD probabilities for gene-patient pairs (derived from the pancancer data set in Tokheim, Collin J., et al. "Evaluating the evaluation of cancer driver genes." Proceedings of the National Academy of Sciences 113.50 (2016): 14330-14335.) formatted to be used by OMEN. - [tokheim_pancancer_somatic_coverage_ranks.pl] A file containing gene coverage data (derived from the CADD data) formatted to be used by OMEN. - [network.pl] A file containing a genome-wide interaction network consisting of high quality metabolic interactions from Recon X and literature curated interactions from Intact. Recon X and Intact data was acquired from Pathway Commons version 8. The resulting network covers 7901 samples, and contains 15.694 nodes and 178.051 edges.

Understanding the Influence of Receptive Field and Network Complexity in Neural-Network-Guided TEM Image Analysis

TEM images of Au nanoparticles of various sizes on ultra-thin carbon substrates and their corresponding labels for semantic segmentation. The images have a dataset label of "images" and the labels have a dataset label of "labels".

opencc-zeroApr 2022View details →

Positive associations of soil organic matter and crop yields across a regional network of working farms

The amount of soil organic matter (SOM) is considered a key indicator of soil properties associated with higher fertility. Despite the ubiquity of assumptions surrounding SOM's contributions to soil functioning, we lack quantitative relationships between SOM and yield outcomes on working farms. We quantified the relationship between SOM and yields of corn (Zea mays L.) and silage for a dataset of 170 fields arrayed across 49 farms in a network of growers based in Wisconsin and Minnesota, USA. As SOM concentrations increase, so do yields, though gains start to level off around 4% SOM. When examining the relationship between yield and soil health indicators representative of biologically active carbon pools, we found that mineralizable carbon (min-C) has a stronger relationship with yield than permanganate oxidizable C (POXC). Mineral fertilizer, manure, and SOM had relationships of similar magnitude with yield, highlighting that SOM in combination with exogenous inputs likely plays an important role in driving agricultural productivity in this region. An SOM by crop rotation interaction indicated that the impact of SOM on crop yields varied depending on rotation (continuous corn versus corn in rotation). That is, continuous corn had lower yields than corn in rotation despite higher SOM concentrations. Our findings provide insight into the relationship between indicators of soil health, farm management, and crop yields for a set of working farms and lend support to the goals of soil health initiatives that rest on building SOM in agricultural soils to improve agricultural outcomes.

opencc-zeroApr 2022View details →

Echolocation clicks and anthropogenic detections with neural network labels in Hawaiian Island HARP data from Kona, Kaua`i, and Pearl and Hermes Reef

This dataset consists of echolocation clicks and detections of anthropogenic signals at three sites in the Hawaiian Islands Archipelago. These sites are Hawaii/Hawaii_K, Kauai/KA, and Pearl and Hermes Reef/PHR. Echolocation clicks were grouped into 5 minute bins, for which summary data is provided. Files are in .mat format that can be read using any desired coding language using a netcdf reading script. Files are separated by site, deployment, and neural network class (i.e. sitedeployment_cbins_class or site_deployment_cbins_class). Manual labels are provided.

A Tungsten Deep Neural-Network Potential for Simulating Mechanical Property Degradation Under Fusion Service Environment

The DP-HYB and DP-SE2potential and the W training database.

opencc-by-4.0Nov 2021View details →

Expanding the Concept of Comprehensive Area Ratio Parameter to the South-Central States: Towards Simplifying the Structural Evaluation of Flexible Pavements at the Network Level

The surface deflection bowl data collected through falling weight deflectometer (FWD) test is utilized by highway agencies in assessing the performance of the flexible pavement. However, a robust method to evaluate pavement sections utilizing FWD data from all the sensors is seldom developed. There is always a need for DOTs and highway agencies to have a simplified procedure, which can be directly implemented in agencies' databases. This study focuses on expanding and validating the concept of previously developed area ratio parameters towards the pavement section of South-Central States (Arkansas, Louisiana, New Mexico, Oklahoma, and Texas) in effectively analyzing the pavement performances. Simulation-based deflections are utilized to develop enhanced deflection-based parameters and to reduce the need for extensive FWD testing in the field. Ninety-seven pavement sections in these states are considered to implement and validate simplified procedures that will be readily available to various transportation agencies to evaluate their pavement conditions at the network level. Due to this purpose, a pavement ranking chart is proposed for the five South-Central states, which categorizes the pavement section into very good, good, fair, and poor pavement sections. Eventually, load-induced effects concerning developed parameters are effectively analyzed to predict the remaining service life of the flexible pavement structures. The developed methodologies will be helpful for DOTs and highway agencies to carry out the rehabilitation and maintenance work in time and estimate the budget required in these procedures.

opencc-zeroJun 2021View details →

Forest cover and connectivity have pervasive effects on the maintenance of evolutionary distinct interactions in seed dispersal networks

Seed dispersal by animals is one of the most important ecological processes in tropical forests, entailing millions of years of evolutionary adaptations of plants and frugivorous animals forming networks of interactions that, ultimately, contribute to the resilience of such forests. We analyze 29 seed dispersal networks in the threatened Atlantic Forest biodiversity hotspot, with data on the frequency of feeding visits by birds to fruiting plants to answer: (1) which are the effects of forest cover and landscape connectivity on the maintenance of phylogenetic diversity (PD) of interacting birds and plants and the evolutionary distinctiveness of the interactions (EDi) between them; and (2) how EDi and plant/bird PD affects the robustness of the interaction networks? We found that forest cover positively influences both plant and bird PD and EDi. Landscape connectivity is an important predictor of bird PD, but not plant PD, suggesting that the spatial arrangement of forest remnants is essential for guaranteeing bird movement among forest fragments. Furthermore, interaction networks of areas with higher PD and EDi had great robustness to the simulated extinction of species, which underscore the importance of larger forest blocks for conserving evolutionary information and, consequently, the health and natural resistance of seed dispersal networks against environmental change.

Training and validation datasets for "Three-Dimensional Implicit Structural Modeling Using Convolutional Neural Network"

This is training and validation datasets used in manuscript "Three-Dimensional Implicit Structural Modeling Using Convolutional Neural Network". In this manuscript, we propose an efficient deep learning method using a Convolutional Neural Network (CNN)  to predict a scalar field from sparse structural data associated with multiple distinct stratigraphic layers and faults. The CNN architecture is beneficial for the flexible incorporation of empirical geological knowledge when trained with numerous and realistic structural models that are automatically generated from a data simulation workflow. It also presents an expressive characteristic of integrating various types of structural constraints by optimally minimizing a hybrid loss function to compare predicted and reference structural models, opening new opportunities for further improving geological modeling.

opencc-zeroApr 2022View details →

Network analysis reveals that acute stress exacerbates gene regulatory responses of the gill to seawater in Atlantic salmon

The transition from freshwater to seawater represents a physiological challenge for Atlantic salmon smolts preparing for downstream migration. Stressors occurring during downstream migration to the ocean impair the ability of smolts to maintain osmotic/ionic homeostasis in seawater. The molecular mechanisms underlying this interaction are not fully understood, especially at the organ level. We combined RNA-Seq with measures of whole-animal homeostasis to examine gene expression dynamics in the gills of smolts associated with impaired seawater tolerance after an aquaculture-related stressor. Smolts were given a 24 h seawater tolerance test before and after exposure to an acute handling/confinement stress. RNA-Seq followed by differential expression and weighted gene correlation network analysis (WGCNA) was used to quantify the transcriptional response of the gill to handling/confinement stress, seawater and their interaction. Exposure to acute stress was associated with a general stress response and impaired osmotic/ionic homeostasis in seawater. We identified gene networks in the gill exhibiting response to acute stress alone, seawater alone, and others exhibiting combined effects of both stress and seawater. Our findings indicate that acute handling/ confinement stress increases the intensity of seawater-related gene expression and suggest that increased investment in mechanisms related to ion transport may be part of a compensatory response to impaired seawater tolerance in smolts.

Data from: Flowering overlap and floral trait similarity help explain the structure of pollination network

Co-flowering communities are usually characterized by high plant generalization but knowledge of the underlying factors leading to high levels of generalization and pollinator sharing, and how these may contribute to network structure is still limited. Flowering phenology and floral trait similarity are considered among the most important factors determining plant generalization and pollinator sharing. However, these have been evaluated independently even though they can act in concert with each other. Moreover, the importance of flowering phenology and floral similarity, via their effects on plant generalization, in the structure of plant–pollinator networks have been scarcely studied. Here, we aim to evaluate the effect of flowering phenology and floral similarity in mediating the degree of pollinator sharing and plant generalization in two coastal communities and uncover their importance as drivers of plant–pollinator network structure. We recorded flower production per species, as well as the identity and frequency of floral visitors along the entire flowering season. We estimated the degree of flowering overlap, the degree of floral similarity (using floral traits associated with size and color), and the degree of pollinator sharing among plant species within both communities. Structural equation models (SEM) showed a positive effect of flowering overlap on pollinator sharing and plant generalization. Pollinator sharing and plant generalization positively affected network nestedness. Furthermore, SEM showed a direct positive effect of flowering overlap on network modularity. The SEM analyses also revealed a significant interaction effect of floral similarity and flowering overlap on pollinator sharing, with consequences for network nestedness in one community. Our results highlight the importance of integrating multiple axes of differentiation such as flowering phenology and floral similarity into our understanding of the drivers of plant–pollinator network structure.

Plant-root pathogenic fungal and plant-mycorrhizal fungal association networks in a subtropical forest

Although rhizosphere fungi are essential for plant survival and ecosystem functioning, little is known about the processes that structure plant–fungal association networks. In this study, we constructed association networks between 43 plant species and two groups of root-associated fungi (mycorrhizal and pathogenic fungi; MF and PF, respectively) in a diverse subtropical forest. We then evaluated the modularity of plant–MF and plant–PF networks and linked them to the functional traits and phylogenies of both plants and fungi. We observed strong modularity in both plant–MF and plant–PF networks. Phylogenetically related fungi tended to emerge in the same modules. MF from distinct modules associated with plants with different specific root length and specific root area in plant–MF networks. PF from distinct modules associated with plants with different dark respiration rate and light compensation point in plant–PF networks. Plant affiliation to modules was explained by both plant traits and phylogeny (22% for plant–MF and 37% for plant–PF networks). In contrast, fungal affiliation to modules was explained by fungal phylogeny (16% for plant–MF and 29% for plant–PF networks). Our results elucidate the link between modularity in plant–root fungal networks and the functional traits and phylogeny of the plants and fungi. Our study highlights the importance of traits and phylogeny in governing root fungal community assembly from network perspective.

Range-wide habitat use of the Harpy Eagle indicates four major tropical forest gaps in the Key Biodiversity Area network

Quantifying habitat use is important for understanding how animals meet their requirements for survival and provides information for conservation planning. Currently, assessments of range-wide habitat use that delimit species distributions are incomplete for many taxa. The Harpy Eagle (Harpia harpyja) is a raptor of conservation concern, widely distributed across Neotropical lowland forests, that currently faces threats from habitat loss and fragmentation. Here, we use penalized logistic regression to identify species-habitat associations and predict habitat suitability based on a new International Union for the Conservation of Nature range metric, termed Area of Habitat. From the species-habitat model, we performed a gap analysis to identify areas of high habitat suitability in regions with limited coverage in the Key Biodiversity Area (KBA) network. Range-wide habitat use indicated that Harpy Eagles prefer areas of 70-75% evergreen forest cover, low elevation, and high vegetation species richness. Conversely, Harpy Eagles avoid areas of >10% cultivated landcover and mosaic forest, and topographically complex areas. Our species-habitat model identified a large continuous area of potential habitat across the pan-Amazonia region, and a habitat corridor from the Chocó-Darién ecoregion of Colombia running north along the Caribbean coast of Central America. Little habitat was predicted across the Atlantic Forest biome, which is now severely degraded. The current KBA network covered 18% of medium to high Harpy Eagle habitat exceeding a target biodiversity area representation of 10%, based on species range size. Four major areas of high suitability habitat lacking coverage in the KBA network were identified in north and west Colombia, western Guyana, and north-west Brazil. We recommend these multiple gaps of habitat as new KBAs for strengthening the current KBA network. Modelled area of habitat estimates as described here are a useful tool for large-scale conservation planning and can be readily applied to many taxa.

Neuronal Transfer Networks (Trainings)

Training versions of the AlexNet Convolutional Neural Network (CNN), related to the evaluation of eye diseases.    -The following datasets were used in this work: <ul> <li>LAG. Contains fundus images with positive and negative glaucoma samples obtained from Beijing Tongren Hospital. Each fundus image is diagnosed by qualified glaucoma specialists, taking into consideration of both morphologic and functional analysis. https://arxiv.org/abs/1903.10831</li> <li>APTOS. Contains images of diabetic retinopathy that were used in the APTOS 2019 blindness screening competitions. Each image has been resized and cropped to have a maximum size of 1024px. A certified clinician rated each image according to the severity of diabetic retinopathy on a scale of 0 to 4. https://www.kaggle.com/c7934597/resized-2015-2019-diabetic-retinopathy-detection/metadata/</li> <li>HRF. Contains 15 images of healthy patients, 15 images of patients with diabetic retinopathy and 15 images of glaucomatous patients. They were captured by a Canon CR-1 fundus camera with a field of view of 45 degrees with a resolution of 3504×2336 px. https://www5.cs.fau.de/research/data/fundus-images/</li> <li>ODIR. Contains images of 5000 patients with various eye diseases collected by Shanggong Medical Technology Co., Ltd. from different hospitals/medical centers in China. The fundus images are captured with various cameras on the market, resulting in varied image resolutions. They classify patients into eight labels based on the images of both eyes. https://odir2019.grand-challenge.org/dataset/</li> <li>sjchoi86-HRF. This dataset contains 601 fundus images divided into 4 groups: normal (300 images), glaucoma (101 images), cataract (100 images) and retina disease (100 images). https://github.com/cvblab/retina_dataset</li> </ul> -Specifications of the data used for training in each version of the Convolutional Neural Network (CNN): <ul> <li>netTransfer: Based on glaucoma and non-glaucoma imag-es cases existing in the LAG-database.</li> <li>netTransfer2: Based on glaucoma and non-glaucoma imag-es cases existing in the LAG-database and the sjchoi86-HRF database (this data-base also contains images on other pathologies, but only the glaucoma samples were used for this training)</li> <li>netTransfer3: Based on glaucoma, diabetic retinopathy and non-disease images cases existing in the LAG-database, sjchoi86-HRF database and the HRF database.</li> <li>netTransfer4: Based on glaucoma, diabetic retinopathy and non-disease images cases existing in the LAG-database, sjchoi86-HRF database, HRF database and the APTOS database. However, data from the fourth iteration was corrupted and then there for lost.</li> <li>netTransfer5: Based on glaucoma, diabetic retinopathy and non-disease images cases existing in the LAG-database, sjchoi86-HRF database, HRF database and the APTOS database (Images were cropped slightly before training).</li> <li>netTransfer6: Based on glaucoma, diabetic retinopa-thy and non-disease images cases existing in the LAG-database, sjchoi86-HRF database, HRF database, APTOS database and ODIR database (this database also contains images on several other diseases, but only the glaucoma and diabetic retinopathy samples were used for this training; furthermore, cropping was used for the images in order the eliminate black borders).</li> </ul> PLEASE CITATE AS: Arias-Serrano I, Velásquez-López PA, Avila-Briones LN et al. Artificial intelligence based glaucoma and diabetic retinopathy detection using MATLAB — retrained AlexNet convolutional neural network [version 1; peer review: 1 approved with reservations]. F1000Research 2023, 12:14 (<a href="https://doi.org/10.12688/f1000research.122288.1">https://doi.org/10.12688/f1000research.122288.1</a>)

Reproducibility package for: AllSynth: Transiently Correct Network Update Synthesis Accounting for Operator Preferences

This is a reproducibility package for the paper "AllSynth: Transiently Correct Network Update Synthesis Accounting for Operator Preferences" to appear in TASE'22.

Data from: Building communities of teaching practice and data-driven open education resources with NEON faculty mentoring networks

With the growing availability and accessibility of big data in ecology, we face an urgent need to train the next generation of scientists in data science practices and tools. One of the biggest barriers for implementing a data-driven curriculum in undergraduate classrooms is the lack of training and support for educators to develop their own skills and time to incorporate these principles into existing courses or develop new ones. Alongside the research goals of the National Ecological Observatory Network (NEON), providing education and training are key components for building a community of scientists and users equipped to utilize large-scale ecological and environmental data. To address this need, the NEON Data Education Fellows program formed as a collaborative Faculty Mentoring Network (FMN) between scientists from NEON and university faculty interested in using NEON data and resources in their ecology classrooms. Like other FMNs, this group has two main goals: 1) to provide tools, resources, and support for faculty interested in developing data-driven curriculum, and (2) to make teaching materials that have been implemented and tested in the classroom available as open educational resources for other educators. We hosted this program using an open education and collaboration platform from the Quantitative Undergraduate Biology Education and Synthesis (QUBES) project. Here, we share lessons learned from facilitating five FMN cohorts and emphasize the successes, pitfalls, and opportunities for developing open education resources through community-driven collaborations.

Transcriptional networks underlying a primary ovarian insufficiency disorder in alligators naturally exposed to EDCs: Transformed read counts and supplementary materials

Interactions between the endocrine system and environmental contaminants are responsible for impairing reproductive development and function. Despite the taxonomic diversity of affected species and attendant complexity inherent to natural systems, the underlying signaling pathways and cellular consequences are mostly studied in lab models. To resolve the genetic and endocrine pathways that mediate affected ovarian function in organisms exposed to endocrine disrupting contaminants in their natural environments, we assessed broad-scale transcriptional and steroidogenic responses to exogenous gonadotropin stimulation in juvenile alligators (Alligator missippiensis) originating from a lake with well-documented pollution (Lake Apopka, FL) and a nearby reference site (Lake Woodruff, FL). We found that individuals from Lake Apopka are charachterized by hyperandrogenism and display hyper-sensitive transcriptional responses to gonadotropin stimulation when compared to individuals from Lake Woodruff. Site-specific transcriptomic divergence appears to be driven by wholly distinct subsets of transcriptional regulators, indicating alterations to fundamental genetic pathways governing ovarian function. Consistent with broad-scale transcriptional differences, ovaries of Lake Apopka alligators displayed impediments to folliculogenesis, with larger germinal beds and decreased numbers of late-stage follicles. After resolving the ovarian transcriptome into clusters of co-expressed genes, most site-associated modules were correlated to ovarian follicule phenotypes across individuals. However, expression of two site-specific clusters were independent of ovarian cellular architecture and are hypothesized to represent alterations to cell-autonomous transcriptional programs. Collectively, our findings provide high resolution mapping of transcriptional patterns to specific reproductive function and advance our mechanistic understanding regarding impaired reproductive health in an established model of environmental endocrine disruption.

Data from: Network analysis of sea turtle movements and connectivity: a tool for conservation prioritization

Aim: Understanding the spatial ecology of animal movements is a critical element in conserving long-lived, highly mobile marine species. Analysing networks developed from movements of six sea turtle species reveals marine connectivity and can help prioritize conservation efforts. Location: Global. Methods: We collated telemetry data from 1,235 individuals and reviewed the literature to determine our dataset's representativeness. We used the telemetry data to develop spatial networks at different scales to examine areas, connections, and their geographic arrangement. We used graph theory metrics to compare networks across regions and species and to identify the role of important areas and connections. Results: Relevant literature and citations for data used in this study had very little overlap. Network analysis showed that sampling effort influenced network structure and the arrangement of areas and connections for most networks was complex. However, important areas and connections identified by graph theory metrics can be different than areas of high data density. For the global network, marine regions in the Mediterranean had high closeness while links with high betweenness among marine regions in the South Atlantic were critical for maintaining connectivity. Comparisons among species-specific networks showed that functional connectivity was related to movement ecology, resulting in networks composed of different areas and links. Main conclusions: Network analysis identified the structure and functional connectivity of the sea turtles in our sample at multiple scales. These network characteristics could help guide the coordination of management strategies for wide-ranging animals throughout their geographic extent. Most networks had complex structures that can contribute to greater robustness, but may be more difficult to manage changes when compared to simpler forms. Area-based conservation measures would benefit sea turtle populations when directed towards areas with high closeness dominating network function. Promoting seascape connectivity of links with high betweenness would decrease network vulnerability.

REIP: a Reconfigurable Environmental Intelligence Platform and Software Framework for Fast Sensor Network Prototyping - use case dataset

Sensor networks have dynamically expanded our ability to monitor and study the world. Their presence and need keep increasing, and new hardware configurations expand the range of physical stimuli that can be accurately recorded. Sensors are also no longer simply recording the data, they process it and transform into something useful before uploading to the cloud. However, building sensor networks is costly and very time consuming. It is difficult to build upon other people’s work and there are only a few open-source solutions for integrating different devices and sensing modalities. We introduce REIP, a Reconfigurable Environmental Intelligence Platform for fast sensor network prototyping. REIP’s first and most central tool, implemented in this work, is an open-source software framework, an SDK, with a flexible modular API for data collection and analysis using multiple sensing modalities. REIP is developed with the aim of being user-friendly, device-agnostic, and easily extensible, allowing for fast prototyping of heterogeneous sensor networks. Furthermore, our software framework is implemented in Python to reduce the entrance barrier for future contributions. We show the potential and versatility of REIP in real world applications, along with performance studies and benchmark REIP SDK against similar systems. This dataset was created for the case study in Section 5 of the paper.

Research data: "Effects of Network Structures on the Production Planning in Closed-loop Supply Chains – A Case Study based Analysis for Lithium-ion Batteries in Europe"

This data set belongs to the paper Effects of Network Structures on the Production Planning in Closed-loop Supply Chains – A Case Study based Analysis for Lithium-ion Batteries in Europe in the International Journal of Production Economics (DOI). The BatPac model, as well as, the model for the economic assessment of the recycling route are not included. The needed data can be found in the file (name). Further, the BatPaC model can be gather from the website of Argonne National Laboratory and the assessment tool for the recycling routes via this DOI: 10.5281/zenodo.6500946.   This work is part of the research project Recycling 4.0 (EFRE | ZW 6-85018080), which is funded by the European Regional Development Fund and managed by the development bank for the German federal state of Lower Saxony (NBank).