Find research datasets worth reusing

NPIP: A Comprehensive Analysis Pipeline for Rapid Pathogen Detection in Clinical Samples Based on Nanopore Sequencing

Background: Rapid and accurate pathogen detection is important for effective control of infectious diseases. However, traditional pathogen culture methods have a very long detection time, as well as high rates of false-positive and false-negative results. Third generation sequencing (TGS) technology brings the new possibility of being used as a pathogen detection method. However, the practicability of a pathogen detection report based on TGS is still lacking. There is also a lack of professional and accurate report interpretation. Results: Here, we report on the development of a pathogen detection and analysis tool (NPIP) based on third generation nanopore sequencing technology. We also prove the practicability of nanopore sequencing and NPIP analysis tools in emergency and clinical pathogen detection by demonstrating its use in a practical case. Conclusions: This platform provides an effective, convenient, and fast analysis tool for clinicians and public health personnel to more successfully apply TGS in pathogen detection.

High-dimensional Anomaly Detection with Radiative Return in e+e- Collisions

Numpy files of Pythia + Delphes simulated e+e- collisions used as DNN/PFN training inputs.

Data for the research article: "Detecting Seismo-ionospheric Anomalies Possibly Associated with the 2019 Ridgecrest (California) Earthquakes by GNSS, CSES and Swarm Observations"

The zip archives contain 10 .mat files (MATLAB readable). Each .mat file can be loaded into the MATLAB workspace using the load command.

opencc-zeroAug 2021View details →

DNA sequences of transgenes detected via environmental DNA (raw ABI files, processed FASTA files, and reference alignments)

We demonstrate that simple, non-invasive environmental DNA (eDNA) methods can detect transgenes of genetically modified (GM) animals from terrestrial and aquatic sources in invertebrate and vertebrate systems. We detected transgenic fragments between 82-234 bp through targeted PCR amplification of environmental DNA extracted from food media of GM fruit flies (Drosophila melanogaster), feces, urine, and saliva of GM laboratory mice (Mus musculus), and aquarium water of GM tetra fish (Gymnocorymbus ternetzi). With rapidly growing accessibility of genome-editing technologies such as CRISPR, the prevalence and diversity of GM animals will increase dramatically. GM animals have already been released into the wild with more releases planned in the future. eDNA methods have the potential to address the critical need for sensitive, accurate, and cost-effective detection and monitoring of GM animals and their transgenes in nature.

opencc-zeroAug 2021View details →

A comprehensive approach to detect hybridization sheds light on the evolution of Earth's largest lizards

Hybridization between species occurs more frequently in vertebrates than traditionally thought, but distinguishing ancient hybridization from other phenomena that generate similar evolutionary patterns remains challenging. Here, we used a comprehensive workflow to discover evidence of ancient hybridization between the Komodo dragon (Varanus komodoensis) from Indonesia and a common ancestor of an Australian group of monitor lizards known colloquially as sand monitors. Our data comprise >300 nuclear loci, mitochondrial genomes, phenotypic data, fossil and contemporary records, and past/present climatic data. We show that the four sand monitor species share more nuclear alleles with V. komodoensis than expected given a bifurcating phylogeny, likely as a result of hybridization between the latter species and a common ancestor of sand monitors. Sand monitors display phenotypes that are intermediate between their closest relatives and V. komodoensis. Biogeographic analyses suggest that V. komodoensis and ancestral sand monitors co-occurred in northern Australia. In agreement with the fossil record, this provides further evidence that the Komodo dragon once inhabited the Australian continent. Our study shows how different sources of evidence can be used to thoroughly characterize evolutionary histories that deviate from a treelike pattern, that hybridization can have long-lasting effects on phenotypes, and that detecting hybridization can improve our understanding of evolutionary and biogeographic patterns.

Assessing reproduciblity of Inherited Variants Detected with Short-read Whole Genome Sequencing

This dataset is part of the following study: <a href="https://www.fda.gov/science-research/bioinformatics-tools/microarraysequencing-quality-control-maqcseqc">https://www.fda.gov/science-research/bioinformatics-tools/microarraysequencing-quality-control-maqcseqc</a> The raw sequencing data can be downloaded from SRA: <a href="https://www.ncbi.nlm.nih.gov/bioproject/PRJNA723125">https://www.ncbi.nlm.nih.gov/bioproject/PRJNA723125</a> and NODE: <a href="https://www.biosino.org/node/project/detail/OEP0018966">https://www.biosino.org/node/project/detail/OEP0018966</a>

opencc-by-nc-nd-4.0Aug 2021View details →

Data for Finnish Dialect Detection

The data used in the paper "Finnish Dialect Identification: The Effect of Audio and Text". If you use the data, please cite: Hämäläinen, Mika; Alnajjar, Khalid; Partanen, Niko & Rueter, Jack (2021). <a href="https://aclanthology.org/2021.emnlp-main.692/">Finnish Dialect Identification: The Effect of Audio and Text</a>. In Proceedings of the 2021 Conference on Empirical Methods in Natural Language Processing (EMNLP). The metadata.json contains the dialectal and normalized transcriptions, the length of the wav files in milliseconds, the path to the wav file, the role of the speaker and the dialect. The data.zip* files contain the wav files. They are partial zip files to make Zenodo upload easier. Code_models.zip contains the code for training the bimodal model and the final trained model presented in the paper. There is also an OpenNMT model that is the text only model described in the paper. You can use it by running: python3 -m onmt.bin.translate -model nmt-model_step_100000.pt -src source_test.txt -output pred.txt -replace_unk Our dataset is licensed under CC BY NC ND 4.0. Academic use only.   The corpus is based on <a href="http://urn.fi/urn:nbn:fi:lb-2017020201">Suomenkielen näytteitä</a> (CC BY  <a href="http://www.kotus.fi/">Kotimaisten kielten keskus</a>).

opencc-zeroAug 2021View details →

Deep learning object detection to estimate the nectar sugar mass of flowering vegetation

Floral resources are a key driver of pollinator abundance and diversity, yet their quantification in the field and laboratory is laborious and requires specialist skills. Using a dataset of 25000 labelled tags of fieldwork-realistic quality, a Convolutional Neural Network (Faster R-CNN) was trained to detect the nectar-producing floral units of 25 taxa in surveyors' quadrat images of native, weed-rich grassland in the UK. Floral unit detection on a test set of 50 model-unseen images of comparable vegetation returned a precision of 90%, recall of 86% and F1 score (the harmonic mean of precision and recall) of 88%. Model performance was consistent across the range of floral abundance in this habitat. Comparison of the nectar sugar mass estimates made by the CNN and three human surveyors returned similar means and standard deviations. Over half of the nectar sugar mass estimates made by the model fell within the absolute range of those of the human surveyors. The optimal number of quadrat image samples was determined to be the same for the CNN as for the average human surveyor. For a standard quadrat sampling protocol of 10–15 replicates, this application of deep learning could cut pollinator-plant survey time per stand of vegetation from hours to minutes. The CNN is restricted to a single view of a quadrat, with no scope for manual examination or specimen collection, though in contrast to human surveyors its object detection is deterministic and floral unit definition is standardised. As agri-environment schemes move from prescriptive to results-based, this approach provides an independent barometer for grassland management which is usable by both landowner and scheme administrator. The model can be adapted to visual estimations of other ecological resources such as winter bird food, floral pollen volume, insect infestation and tree flowering/fruiting, and by adjustment of classification threshold may show acceptable taxonomic differentiation for presence-absence surveys.

Variation in plant leaf traits affects transmission and detectability of herbivore vibrational cues

Many insects use plant-borne vibrations to obtain important information about their environment, such as where to find a mate or a prey, or when to avoid a predator. Plant species can differ in the way they vibrate, possibly affecting the reliability of information, and ultimately the decisions that are made by animals based on this information. We examined whether the production, transmission and possible perception of plant-borne vibrational cues is affected by variation in leaf traits. We recorded vibrations of 69 Spodoptera exigua caterpillars foraging on four plant species that differed widely in their leaf-traits (cabbage, beetroot, sunflower and corn). We carried out a transmission and an airborne noise absorption experiment to assess whether leaf traits influence amplitude and frequency characteristics, and background noise levels of vibrational-chewing cues. Our results reveal that species-specific leaf traits can influence transmission and potentially perception of herbivore-induced chewing vibrations. Experimentally-induced vibrations attenuated stronger on plants with thicker leaves. Amplitude and frequency characteristics of chewing vibrations measured near a chewing caterpillar were, however, not affected by leaf traits. Furthermore, we found a significant effect of leaf area, water content and leaf thickness - important plant traits against herbivory, on the vibrations induced by airborne noise. On larger leaves higher amplitude vibrations were induced, whereas on thicker leaves containing more water airborne noise induced higher peak frequencies. Our findings indicate that variation in leaf traits can be important for the transmission and possibly detection of vibrational cues.

Data for the detection of the boreal chorus frog (Pseudacris maculata) using environmental DNA and call surveys at 180 ponds sampled in 2017-2018 in southeastern Québec, Canada

The boreal chorus frog (Pseudacris maculata) is at risk of extinction in parts of its range in Canada. Our objectives were to quantify the influence of local and landscape characteristics on the occurrence of the species in wetlands in southern Québec. We hypothesized that site occupancy depends on local characteristics and landscape characteristics contributing to site connectivity. We developed an environmental DNA (eDNA) method to detect the species and compared the detection probability of this method to traditional call surveys. We collected water samples at a total of 180 sites (90 in 2017, 110 in 2018), whereas we surveyed a subset of 63 sites using both eDNA and call surveys in 2018. Site occupancy varied across years, but was higher in sites where the species had been previously detected during the last 12 years by other studies. Site occupancy did not vary with other local and landscape characteristics, in part due to an apparent decrease in the number of sites occupied by the species since the last 12 years. Detection probability via eDNA (0.81; 95% CI: [0.31; 0.98]) did not differ from that of call surveys (0.62; 95% CI: [0.25; 0.89]). To identify the optimal sampling period for the boreal chorus frog, future studies should estimate the detection probability of eDNA during the breeding season and the larval development period of the species.

Detection histories of mesocarnivores in agricultural areas of Southern Chile

We obtained mesocarnivore detection/non-detection data from 180 sampling units (4 km2 each) located in the agricultural landscapes of southern Chile from January-April of 2019. We used single-species occupancy models to investigate the associations of forest fragmentation, forest loss, and private land ownership subdivision (as a measure of human use intensification) with the occurrence of four mesocarnivores (güiña, grey fox, culpeo fox, and Molina's hog-nosed skunk), and extended this framework to two-species occupancy models to assess patterns of mesocarnivore co-occurrence with domestic carnivores. We also assessed whether co-occurrence of native and domestic carnivores led to shifts in species' temporal activity. We hypothesized that the occurrence of native mesocarnivores was largely mediated by human use intensification variables only, the occurrence of domestic mesocarnivores only, or a combination of both. Our results largely supported the human use intensification hypothesis, with some influence of domestic mesocarnivores. Mesocarnivore occurrence shifted from a native to a domestic species composition as private land ownership subdivision increased, and native mesocarnivores shifted their behaviour temporally when co-occurring with domestics. In addition, the presence of domestic dogs was associated with an absence of native mesocarnivores, possibly driving a defaunation process in agricultural areas.

opencc-by-4.0Sep 2021View details →

Annotated datasets of Scots pine cross-sectional images for root rot detection and resin detection

These are the annotated datasets used in the publication available at <a href="https://doi.org/10.1080/14942119.2024.2327247">https://doi.org/10.1080/14942119.2024.2327247</a> The first dataset is for root rot detection, the other dataset is for resin detection. Please see the publication for details.

A multi-modal sensor dataset for continuous stress detection of nurses in a hospital

Advances in wearable technologies provide the opportunity to monitor many physiological variables continuously. Stress detection has gained increased attention in recent years, especially because early stress detection can help individuals better manage health to minimize the negative impacts of long-term stress exposure. This paper provides a unique stress detection dataset created in a natural working environment in a hospital. This dataset is a collection of biometric data of nurses during the COVID-19 outbreak. Studying stress in a work environment is complex due to the influence of many social, cultural, and individuals experience in dealing with stressful conditions. In order to address these concerns, we captured both the physiological data and associated context pertaining to the stress events. We monitored specific physiological variables, including electrodermal activity, heart rate, skin temperature, and accelerometer data of the nurse subjects. A periodic smartphone-administered survey also captured the contributing factors for the detected stress events. A database containing the signals, stress events, and survey responses is available upon request.

opencc-by-4.0Sep 2021View details →

WHUS2-CD+ dataset for sentinel-2 cloud detection validation

WHUS2-CD+ is a cloud validation detection dataset for Sentinel-2A images. WHUS2-CD+ contains 36 manually labeled cloud masks at 10m resolution and corresponding Sentinel-2A images evenly distributed over China mainland. If you use this dataset for your research, please cite us accordingly: #Reference:  [1] J. Li, Z. Wu, Z. Hu, C. Jian, S. Luo, L. Mou, X. Zhu, and M. Molinier, "A lightweight deep learning based cloud detection method for Sentinel-2A imagery fusing multi-scale spectral and spatial features," in IEEE Transactions on Geoscience and Remote Sensing, 2021. <a href="https://doi.org/10.1109/TGRS.2021.3069641">https://doi.org/10.1109/TGRS.2021.3069641</a>. [2] Z. Wu, J. Li, Y. Wang, Z. Hu and M. Molinier, "Self-Attentive Generative Adversarial Network for Cloud Detection in High Resolution Remote Sensing Images," in IEEE Geoscience and Remote Sensing Letters, vol. 17, no. 10, pp. 1792-1796, Oct. 2020. <a href="https://doi.org/10.1109/LGRS.2019.2955071">https://doi.org/10.1109/LGRS.2019.2955071</a>. The training and testing list is (The challenging senes are marked in bold): <table> <tbody> <tr> <td>Training set</td> </tr> <tr> <td>S2A_MSIL1C_20190714T043711_N0208_R033_T46TFN_20190714T073938</td> </tr> <tr> <td>S2A_MSIL1C_20191219T040151_N0208_R004_T47SQU_20191219T055033</td> </tr> <tr> <td>S2A_MSIL1C_20190630T045701_N0207_R119_T45SWC_20190630T080543</td> </tr> <tr> <td>S2A_MSIL1C_20191215T042151_N0208_R090_T46RGV_20191215T065406</td> </tr> <tr> <td>S2A_MSIL1C_20180930T044701_N0206_R076_T45SXR_20180930T074413</td> </tr> <tr> <td>S2A_MSIL1C_20200317T024541_N0209_R132_T51TWM_20200317T053350</td> </tr> <tr> <td>S2A_MSIL1C_20180816T053641_N0206_R005_T44TKK_20180816T093424</td> </tr> <tr> <td>S2A_MSIL1C_20191023T040821_N0208_R047_T47TQF_20191023T074550</td> </tr> <tr> <td>S2A_MSIL1C_20180824T031541_N0206_R118_T50TKL_20180824T061636</td> </tr> <tr> <td>S2A_MSIL1C_20191118T025011_N0208_R132_T50RMN_20191118T071843</td> </tr> <tr> <td>S2A_MSIL1C_20190916T023551_N0208_R089_T50RQS_20190916T042547</td> </tr> <tr> <td>S2A_MSIL1C_20190819T031541_N0208_R118_T49SFU_20190819T065332</td> </tr> <tr> <td>S2A_MSIL1C_20190815T051651_N0208_R062_T44TPN_20190815T090034</td> </tr> <tr> <td>S2A_MSIL1C_20200410T022551_N0209_R046_T51TXG_20200410T042047</td> </tr> <tr> <td>S2A_MSIL1C_20191002T025551_N0208_R032_T50TQQ_20191002T054113</td> </tr> <tr> <td>S2A_MSIL1C_20180429T032541_N0206_R018_T49SCV_20180429T062304</td> </tr> <tr> <td>S2A_MSIL1C_20200506T024551_N0209_R132_T51UWS_20200506T043639</td> </tr> <tr> <td>S2A_MSIL1C_20200325T034531_N0209_R104_T47RQL_20200325T065315</td> </tr> <tr> <td>S2A_MSIL1C_20190928T031541_N0208_R118_T49RBJ_20190928T061248</td> </tr> <tr> <td>S2A_MSIL1C_20180827T032541_N0206_R018_T48RYV_20180827T062627</td> </tr> <tr> <td>S2A_MSIL1C_20200222T030731_N0209_R075_T49QEE_20200222T060244</td> </tr> <tr> <td>S2A_MSIL1C_20180722T030541_N0206_R075_T49RFP_20180722T060550</td> </tr> <tr> <td>S2A_MSIL1C_20180729T025551_N0206_R032_T49RGL_20180729T055945</td> </tr> <tr> <td>S2A_MSIL1C_20200506T024551_N0209_R132_T50SPE_20200506T052918</td> </tr> <tr> <td>Testing set</td> </tr> <tr> <td>S2A_MSIL1C_20180930T030541_N0206_R075_T49QDD_20180930T060706</td> </tr> <tr> <td>S2A_MSIL1C_20191105T023901_N0208_R089_T51STR_20191105T054744</td> </tr> <tr> <td>S2A_MSIL1C_20190812T032541_N0208_R018_T48RXU_20190812T070322</td> </tr> <tr> <td>S2A_MSIL1C_20190602T021611_N0207_R003_T52TES_20190602T042019</td> </tr> <tr> <td>S2A_MSIL1C_20190328T033701_N0207_R061_T49TCF_20190328T071457</td> </tr> <tr> <td>S2A_MSIL1C_20191001T050701_N0208_R019_T45TXN_20191002T142939</td> </tr> <tr> <td>S2A_MSIL1C_20200416T042701_N0209_R133_T46SFE_20200416T074050</td> </tr> <tr> <td>S2A_MSIL1C_20200528T050701_N0209_R019_T44SPC_20200528T082127</td> </tr> <tr> <td>S2A_MSIL1C_20210207T023851_N0209_R089_T52UCU_20210207T040210</td> </tr> <tr> <td>S2A_MSIL1C_20210126T052111_N0209_R062_T44SNE_20210126T063836</td> </tr> <tr> <td>S2A_MSIL1C_20210102T054231_N0209_R005_T43SFB_20210102T065941</td> </tr> <tr> <td>S2A_MSIL1C_20201206T041141_N0209_R047_T47SMV_20201206T053320</td> </tr> </tbody> </table>

opencc-by-4.0Sep 2021View details →

Anomaly Detection and Machine Learning

The datasets were preprocessed. Correlated features were removed. Related papers: [1] Iman Sharafaldin, Arash Habibi Lashkari, and Ali A. Ghorbani, “Toward Generating a New Intrusion Detection Dataset and Intrusion Traffic Characterization”, 4th International Conference on Information Systems Security and Privacy (ICISSP), Portugal, January 2018 [2] Nour Moustafa, October 16, 2019, "UNSW_NB15 dataset", IEEE Dataport, doi: https://dx.doi.org/10.21227/8vf7-s525. [3] “Sebastian Garcia, Agustin Parmisano, & Maria Jose Erquiaga. (2020). IoT-23: A labeled dataset with malicious and benign IoT network traffic (Version 1.0.0) [Data set]. Zenodo. http://doi.org/10.5281/zenodo.4743746” [4]  A. D. Kent, “Comprehensive, Multi-Source Cybersecurity Events,” Los Alamos National Laboratory, http://dx.doi.org/10.17021/1179829, 2015.

openetalab-2.0Jun 2021View details →

Detection of bow echoes in French kilometer-scale models (AROME-EPS & AROME models of Météo-France)

To detect bow echoes (BE) directly in simulated reflectivity fields from French kilometer-scale models, three datasets are available. These datasets allowed for fitting and testing a univariate U-Net (convolutional neural network) using simulated reflectivity fields. Grid area : 717x1121 grid points, Western Europe (12°W-16°E and 37.5°N-55.4°N, resolution 0.025°) In each tar.gz file, files named 'input_BE' are reflectivity fields from operational models (in mm per hour), 'groundTruth' files correspond to manually labeled contours of BE (field with 1 for each grid point in BE and 0 for outside). The file format is HDF5. HDF5 files contain an array of N fields (array shape : (N,1,717,1121)) - dataset_train.tar.gz (N=6206): input and target fields to train the U-Net (only from the AROME-EPS model) - dataset_validation.tar.gz (N=2620): input and target fields to validate the U-Net (only from the AROME-EPS model). Training and validation databases contain independent weather case studies. - dataset_det_AROME.tar.gz  (N=348): input and target fields to apply U-Net to the French deterministic AROME model (same grid than one of AROME-EPS).  Weather case studies are the same than those in the validation database. - optimal_UNet_architecture.json : architecture of the optimal U-Net configuration - optimal_UNet_weights.h5 : weights of the trained optimal U-Net configuration

opencc-by-sa-4.0Jan 2018View details →

A comprehensive evaluation of module detection methods for gene expression data

A critical step in the analysis of large genome-wide gene expression datasets is the use of module detection methods to group genes into co-expression modules. Because of limitations of classical clustering methods, numerous alternative module detection methods have been proposed, which improve upon clustering by handling co-expression in only a subset of samples, modeling the regulatory network, or allowing overlap between modules. In this study we use known regulatory networks to do a comprehensive and robust evaluation of these different methods. Overall, decomposition methods outperform all other strategies, while we could not find any clear advantages of biclustering and network-inference based approaches.  Using our evaluation workflow, we also investigate several practical aspects of module detection, such as parameter estimation and the use of alternative similarity measures, and conclude with recommendations for the further development of these methods.

opencc-by-4.0Oct 2021View details →

Trained network for cell-cycle slowdown detection - DetecDiv (id03)

<pre>Trained network for cell-cycle slowdown detection. Related to the dataset: <a href="https://doi.org/10.5281/zenodo.5553796">doi.org/10.5281/zenodo.5553796</a></pre> ------------------------------------------ Author(s): Théo, ASPERT Contact email: theo.aspert@gmail.com Affiliation: IGBMC, Université de Strasbourg Funding bodies: This work was supported by the Agence Nationale pour la Recherche, the grant ANR-10-LABX-0030-INRT, a French State fund managed by the Agence Nationale de la Recherche under the frame program Investissements d'Avenir ANR-10-IDEX-0002-02.

opencc-by-4.0Apr 2021View details →

DeepPlastic: An Open Source Image Dataset for Epipelagic Marine Plastic Detection

Deep Plastic <ul> <li>Enhanced Object Detection for Epipelagic Plastic.</li> <li>This repository contains source code for the method developed in <a href="https://arxiv.org/pdf/2105.01882.pdf">DeepPlastic: Identifying Marine Plastic In The Epipelagic Zone using Computer Vision and Deep Learning</a></li> <li> Information: </li> <li>Paper: [Coming Soon]</li> <li>YouTube video of Results: <a href="https://youtu.be/8zBdFxaK4Os">https://youtu.be/8zBdFxaK4Os</a></li> <li> Object Detection Model </li> <li>Four models: YOLOv4, YOLOv5, MobileSSD, Faster RCNN Inception V2</li> <li>Small efficient and high precision models can be used for real-time object detection.</li> <li>Model architecture and implementation details: <a href="https://arxiv.org/">https://arxiv.org/</a></li> <li>Weights for YOLOv4 and YOLOv5 are provided in the model/ <ul> <li>YOLOv4: best. weights; use <a href="https://drive.google.com/file/d/1YOTtZ2cHbqgxHukzLp01OVsUoa2CwwXs/view?usp=sharing">best.weights</a></li> <li>YOLOv5: best.pt; use <a href="https://drive.google.com/file/d/14mBOhtLrE2d3hudqjwBZmawKAvTF4zxS/view?usp=sharing">best.pt</a></li> </ul> </li> <li> Google Colab Links Note: Click on File and Save Copy in Drive. If you try to edit my file it'll ask you for permission and send me an email. Please make your own copy. </li> <li>YOLOv5: <a href="https://colab.research.google.com/drive/1_qzbpBWkNfxQ0ny-DvsKicCeM0aFU4eW?usp=sharing">https://colab.research.google.com/drive/1_qzbpBWkNfxQ0ny-DvsKicCeM0aFU4eW?usp=sharing</a></li> <li> DeepTrash DataSet </li> <li>1900 training images, 637 test images, 637 validation images (60, 20, 20 split)</li> <li>Field images taken from Lake Tahoe, San Francisco Bay and Bodega Bay in CA.</li> <li>Deep Sea images are from JAMSTEK JEDI dataset: <a href="http://www.godac.jamstec.go.jp/">http://www.godac.jamstec.go.jp/</a></li> </ul>