Find research datasets worth reusing
Search datasets from major research repositories and use ShareScore to quickly assess how well each record supports discovery, access, and reuse.
125
datasets available to search
ShareScore release 0.9.0
Dataset results
125 results for “Structure learning”
Repository for "Direct simulation and machine learning structure identification unravel soft martensitic transformation and twinning dynamics"
<p>These files contain data and generation codes for figures, a numerical code for the analysis of structure factors, and a numerical code for the simulation, in an article "Direct simulation and machine learning structure identification unravel soft martensitic transformation and twinning dynamics".</p>
Dataset for deep-learning density functional theory Hamiltonian for efficient ab initio electronic-structure calculation
<p>Dataset files of atomic structures and Hamiltonian matrices of graphene, MoS<sub>2</sub>, bilayer graphene and bilayer bismuthene.</p> <p>Please note that the DFT results in this dataset were calculated using OpenMX. This means that if you want to use a DeepH model trained on this dataset to calculate properties, you need to use the <a href="https://github.com/mzjb/overlap-only-OpenMX">overlap calculated using OpenMX</a>. The orbital information required for overlap calculations can be found in the <a href="https://www.nature.com/articles/s43588-022-00265-6">paper</a>.</p>
EMRNA: Accurate RNA structure determination from cryo-EM maps by deep learning and integrated modeling
<p>EMRNA: Accurate RNA structure determination from cryo-EM maps by deep learning and integrated modeling.</p><p>Here stores the input files and output structures of EMRNA and the reproduction result of auto-DRRAFTER.</p>
360 Structured Light with Learned Metasurfaces: dataset
<p>Dataset repository for "360 structured light with learned metasurfaces"</p>
Linking satellites to genes with machine learning to estimate phytoplankton community structure from space
<p><strong>General description</strong></p> <p>The datasets presented in this repository have served in the development of a new ocean color algorithm to derive the relative cell abundance of seven phytoplankton groups (output of algorithm #1, called SOMRCA), as well as their contribution to total chlorophyll a (ChlaPG, output of algorithm #2, called SOMChlF) at the global scale using an omic-based marker: psbO. The outputs of the algorithm SOMChlF were compared to the HPLC-based definition of phytoplankton groups.</p> <p>All the details about this study are found in El Hourany, R., Pierella Karlusich, J., Zinger, L., Loisel, H., Levy, M., and Bowler, C.: Linking satellites to genes with machine learning to estimate phytoplankton community structure from space, Ocean Sci., 20, 217–239, https://doi.org/10.5194/os-20-217-2024, 2024.</p> <p>In "Tara_Oceans_psbO_dataset_Final.xlsx", it can be found the Tara Oceans' psbO metagenomic counts converted into relative cell abundance and Chlorophyll-a contribution for seven phytoplankton groups alongside satellite matchups. This dataset was used for algorithm development. In "Assets HPLC_SOMChlF.xlsx", the HPLC database was used as a comparison with Satellite-derived ChlaPG. Each data document presents a description sheet.</p> <p>In the following, the datasets used in this study are described.</p> <p><strong>Tara Oceans psbO metagenomic abundances</strong><br>The psbO gene is a single-copy gene in most eukaryotes and prokaryotes. We used psbO reads from the metagenomes generated by the Tara Oceans expedition as a proxy for phytoplankton relative cell abundance (see more details in Pierella Karlusich et al., 2023 Mol Ecol Res; https://doi.org/10.1111/1755-0998.13592).</p> <p>Among the 210 Tara Oceans stations, 145 stations sampled metagenomes in different ocean regimes from oligotrophic to eutrophic waters (Chl a from 0.01 to 10 mg m−3, median at 0.3 mg m−3) from 2009 to 2013. Seawater samples were filtered to differentiate five planktonic size fractions (0.22–3, 0.8–5, 5–20, 20–180, 180–2000 µm). <br>We retrieved the psbO read abundances from each Tara Oceans size-fractionated seawater sample from the Supplementary Material from Pierella Karlusich et al., 2023 Mol Ecol Res (https://www.ebi.ac.uk/biostudies/files/S-BSST761/psbO_mapping_against_Tara_Oceans_metagenomes.tsv).</p> <p>We used the psbO data to taxonomically differentiate seven phytoplankton groups: diatoms, dinoflagellates, green algae, haptophytes, pelagophytes, cryptophytes, and prokaryotes (cyanobacteria). The psbO read abundances of these seven groups are expressed as relative phytoplankton cell abundance (%) and their contribution to the Chlorophyll-a (Chla PG, in mg m-3). Phytoplankton that were not assigned to any of these seven groups (unclassified) represented less than 5 % of the total relative cell abundance among all size classes. To obtain a single value of relative cell abundance per station, we pooled the five size fractions into a single aggregated sample. For Chl a content estimation, we used a conversion via size-dependent weights (see formula 1 in El Hourany et al., 2024).</p> <p>There are two levels of information derived from the molecular dataset: relative abundance of psbO reads as a proxy for relative cell abundance and the fraction of Chl a that each group represents. Both types of information have different implications. Chl a is often used as a proxy for biomass, which is a relevant parameter for energy and matter fluxes (e.g., food webs, biogeochemical cycles). At the same time, cell abundance corresponds to species abundance for unicellular organisms, which is an important measure for inferring community assembly processes.<br> <br><strong>Satellite Matchups</strong><br>We used ocean color products from the GlobColour project (R2019, full archive reprocessed, 2020) to retrieve satellite matchups for the psbO-derived abundances. These products were constructed by merging data from various satellite sensors: Sea-viewing Wide Field-of-view Sensor (SeaWiFS), Moderate Resolution Imaging Spectroradiometer (MODIS), Visible Infrared Imaging Radiometer Suite (VIIRS), Medium Resolution Imaging Spectrometer (MERIS), and Ocean and Land Colour Instrument (OLCI).</p> <p>We used 16 GlobColour products as inputs to retrieve the phytoplankton community structure: chlorophyll a concentration (Chl a, product name: CHL1-AVW), remote sensing reflectances (Rrs) at 11 wavelengths (412, 443, 469, 490, 510, 531, 547, 555, 620, 645, and 670 nm), light attenuation coefficient at 490 nm (Kd490), photosynthetically available radiation (PAR), normalized fluorescence light height (NFLH), and particulate backscattering at 443 nm (bbp). These products have daily and 4 km spatiotemporal resolution. In addition, we used the Climate Change Initiative (CCI) sea surface temperature (SST) product at 4 km resolution and daily frequency distributed by the Copernicus Marine Services (CMEMS) portal.</p> <p><strong>HPLC datasets</strong><br>To compare satellite-derived phytoplankton group Chla fractions' distribution (outputs of the algorithm named SOMChlF) with more conventional DPA-based products, we compiled a global HPLC dataset regrouping 12 000 HPLC observations from several HPLC datasets between 1997 and 2014. This HPLC dataset was collocated with the SOMChlF-based ChlaPG. This dataset depicts the abundance of the pigments most widely used to identify major phytoplankton groups: fucoxanthin (Fuco), peridinin (Perid), alloxanthin (Allo), zeaxanthin (Zea), chlorophyll b (Chl b), 19-hexanoyloxyfucoxanthin (19HF), and 19-butanoyloxyfucoxanthin (19BF).</p> <p>Diagnostic pigments were used to estimate the Chl a fraction for each phytoplankton group, namely diatoms, dinoflagellates, haptophytes, green algae, cryptophytes, pelgophytes, and prokaryotes. The Chl a fraction per group is expressed by</p> <p>HPLC-based ChlaPG = Chla in-situ · DP · α / Sum (DP · α) where "α" is a coefficient associated with a diagnostic pigment (DP) for a specific PG.</p> <p>All the details are found in El Hourany, R., Pierella Karlusich, J., Zinger, L., Loisel, H., Levy, M., and Bowler, C.: Linking satellites to genes with machine learning to estimate phytoplankton community structure from space, Ocean Sci., 20, 217–239, https://doi.org/10.5194/os-20-217-2024, 2024.</p> <p> </p>
Dataset for "Generalizing deep-learning electronic structure calculation to plane-wave basis"
<div> <div>This is the dataset for the paper "Generalizing deep-learning electronic structure calculation to plane-wave basis". For details, please see README.md.</div> </div>
Reconstructing the 2D Material Structure from Diffraction Pattern with Physical-Sensitive Deep Learning
<p>Dataset, saved Models' parameters, and test set</p>
Deep Learning for Structure-Preserving Universal Stable Koopman-Inspired Embeddings for Nonlinear Canonical Hamiltonian Dynamics
<p>This dataset provides python implementation of the paper: Deep Learning for Structure-Preserving Universal Stable Koopman-Inspired Embeddings for Nonlinear Canonical Hamiltonian Dynamics, which is available on arXiv https://arxiv.org/abs/2308.13835.</p>
Data related to the publication "structure and transport properties of LiTFSI-based deep eutectic electrolytes from machine-learned interatomic potential simulations"
<p>Reference training and test datasets, trained ML potential models, and input scripts for the training (Allegro) and MD simulations (LAMMPS).</p>
Data for "Supervised machine learning methods for crystal structure prediction of the binary Cs-Te system"
<p>Crystal structures, high-throughput calculations and trained machine learning models presented in the paper "Supervised machine learning methods for crystal structure prediction of the binary Cs-Te system".</p> <ul> <li><em>crystal_datasets </em>contains the input/output data sets of crystal structures for high-throughput calculations and ML models.</li> <li><em>aiida_ht_calculations </em>contains the data regarding the high-throughput DFT calculations.</li> <li><em>ml_models</em> contains the trained ML models.</li> </ul> <p>Eeach zip-archive contains a jupyter-notebook examplifying how the data can be accessed and reused.</p>
A Deep Learning Approach to the Forward Prediction and Inverse Design of Plasmonic Metasurface Structural Color - Raw Data
<p>Reflection spectra of PDMS - Al nanorod metamaterials were collected using LUMERICAL FDTD simulations. PDMS material properties were defined using a refractive index of 1.41 and Al material properties were defined using frequency selective permittivities from the handbook of Palik. A total of 4620 structures were simulated, sweeping the following dimension parameters:</p> <ul> <li>Aluminium thickness (t)</li> <li>Pillar height (h)</li> <li>Pillar diameter (d)</li> </ul> <p>Reflectance spectra were converted into CIE 1931 chromaticity values (x,y). This dataset is comprehensive and allows for the development of deep learning models for the forward and inverse design of the given metamaterial structure as detailed in the associated manuscript. The associated manuscript and supporting documentation provide extensive details of data collection and processing methods.</p> <p> </p>
FIGURE 6 in Learning from Linnaeus: towards developing the foundation for a general structure concept for morphology*
FIGURE 6. The part of the definition of the genus Mandragora that refers to its stamen, taken from Linnaeus's Genera plantarum (1737) and transformed into a RDF graph.
FIGURE 5 in Learning from Linnaeus: towards developing the foundation for a general structure concept for morphology*
FIGURE 5. Implementation of the structure concept in RDF ontology: The trait to be described is represented by the 'Subject' of a RDF triple. The 'property' represents one perceptual category of the structure concept and functions as an empirical question that can only be answered by studying the trait. The answer to the question is represented by the 'Object' of the RDF triple and corresponds with one of the values that are allowed for this category according to the structure concept. One such describing RDF triple represents a morphological datum – the smallest piece of morphological information possible.
FIGURE 4 in Learning from Linnaeus: towards developing the foundation for a general structure concept for morphology*
FIGURE 4. Example of two definitions of concepts expressed in RDF: a) the definition of 'polarized junctioned cell' (for more information see text); b) the definition of 'protonephridium'—a protonoephridium always consists of a nephropore cell, a duct cell, a terminal cell, and extracellular matrix. However, some structures cannot be defined satisfactorily without reference to dispositions of being able to actively participate in specific biological processes. This is also the case for protonephridium, which participates in the process of excretion. This process can be partitioned into different phases, which at their turn have different parts of the protonephridium as their participants.
Data for "Machine-learning-aided atomic structure identification of interfacial ionic hydrates from AFM images"
<p>Dataset for Neural Network training and testing of paper entitled "Machine-learning-aided atomic structure identification of interfacial ionic hydrates from AFM images" (<a href="https://doi.org/10.1093/nsr/nwac282">https://doi.org/10.1093/nsr/nwac282</a>).</p> <p>Each file contains named-dependent simulated AFM Images at different tip height and corresponding atomic structure file in POSCAR format. (See detailed description in the manuscript <a href="https://doi.org/10.1093/nsr/nwac282">https://doi.org/10.1093/nsr/nwac282</a>)</p> <p> </p>
Dataset for deep-learning electronic-structure calculation of magnetic superstructures
<p>Dataset files of atomic structures, magnetic configurations and Hamiltonian matrices of monolayer NiBr<sub>2</sub>, monolayer CrI<sub>3</sub> and bilayer CrI<sub>3</sub>. The dataset file of bilayer CrI<sub>3</sub> exceeds the file size limit of Zenodo and can be downloaded in <a href="https://cloud.tsinghua.edu.cn/f/53b2c86b785442ae8916/?dl=1">this link</a> (115 GB).</p>
Supplementary materials for "Machine Learning Electronic Structure Methods Based On The One-Electron Reduced Density Matrix"
<p><strong>Supplementary materials for "Machine Learning Electronic Structure Methods Based On The One-Electron Reduced Density Matrix"</strong></p> <p>by X. Shao, L. Paetow, M. E. Tuckerman and M. Pavanello</p> <p><strong>The QMLearn software</strong></p> <p>A current snapshot of the QMLearn software is available on GitLab at https://gitlab.com/pavanello-research-group/qmlearn. Video tutorials and other examples (including Jupyter Notebooks) are available at http://qmlearn.rutgers.edu</p> <p><strong>Data availability</strong></p> <p>We share all training/test sets and notebooks needed to reproduce Table II and Figures 1-7. For all figures, we provide Jupyter notebooks and the needed data to exactly reproduce the figures. Trajectory files for all IR spectra are shared. The collection of all materials is available in this dataset.</p> <p>The QMLearn version used for this work is 0.0.1.</p> <p>Specific QMLearn dependencies and their version used for all calculations are listed as follows:<br> - ase (3.22.1)<br> - h5py (3.7.0)<br> - numpy (1.23.1)<br> - pyscf (2.0.1)<br> - scikit-learn (1.1.1)<br> - scipy (1.8.1)</p>
A deep learning framework combining molecular image and protein structural representation identifies candidate drugs for chronic pain
<p>Official dataset for <i>A deep learning framework combining molecular image and protein structural representation identifies candidate drugs for chronic pain</i>. </p><p>The related code can be found at <a href="https://github.com/yuxin212/GPCR-public">here</a> and <a href="https://github.com/ChengF-Lab/LISA-CPI">here</a>. </p><p>The dataset is now public. </p>
Structure-based self-supervised learning enables ultrafast prediction of stability changes upon mutation at the protein universe scale
<p>Pythia computed all single mutations of <em>E.coli</em> proteome, high quality high quality of Swiss-Prot structures and thermophilic proteins used in analysis.</p>
Human Learning of New Structured Information Across Time and Sleep
ClinicalTrials.gov study NCT05910762. IPD Sharing: YES. Countries: 1. Publications: 54.
ScienceDex guides
Understand access before you commit
These curated guides explain access requirements, typical timelines, costs, and reuse considerations for widely used research datasets.
Allen Brain Atlas
Allen Brain Atlas is an Allen Institute collection of brain map atlases, datasets, APIs, and analysis tools covering mouse, human, and non-human primate brain resources.
Annotated Behaviour and Observability Dataset (ABODe)
ABODe is a University of Edinburgh DataShare dataset for behavior classification in group-housed mice using home-cage video, identities, bounding boxes, ground-plate positions, and annotator labels.
DANDI Archive for NWB datasets
DANDI is a BRAIN Initiative archive for publishing and sharing neurophysiology data, including electrophysiology, optophysiology, and behavioral data packaged as NWB and related standards.
International Brain Laboratory public data
The International Brain Laboratory public data releases expose standardized mouse decision-making experiments, including Neuropixels recordings, widefield calcium imaging, behavior, and session metadata accessed through the ONE API.
OpenNeuro
OpenNeuro is a free, open platform for sharing neuroimaging datasets, with public search, dataset pages, and download paths for web, S3, DataLad, and the OpenNeuro CLI.