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.
39
datasets available to search
ShareScore release 0.7.1
Dataset results
39 results for “Link Prediction”
Link-prediction on Biomedical Knowledge Graphs
<p>Release of code and experimental data from the paper <em>Towards Linking Graph Topology to Model Performance for Biomedical Knowledge Graph Completion </em>(<em>Machine Learning for Life and Material Sciences</em> workshop @ ICML2024) and <a href="https://arxiv.org/abs/2409.04103" rel="nofollow">The Role of Graph Topology in the Performance of Biomedical Knowledge Graph Completion Models</a>.</p> <div> <div>Knowledge Graph Completion has been increasingly adopted as a useful method for several tasks in biomedical research, like drug repurposing or drug-target identification. To that end, a variety of datasets and Knowledge Graph Embedding models has been proposed over the years. However, little is known about the properties that render a dataset useful for a given task and, even though theoretical properties of Knowledge Graph Embedding models are well understood, their practical utility in this field remains controversial. We conduct a comprehensive investigation into the topological properties of publicly available biomedical Knowledge Graphs and establish links to the accuracy observed in real-world applications. By releasing all model predictions we invite the community to build upon our work and continue improving the understanding of these crucial applications.</div> <div> </div> <div>Experiments were conducted on six datasets: five from the biomedical domain (<a href="../records/268568">Hetionet</a>, <a href="https://dataverse.harvard.edu/dataset.xhtml?persistentId=doi:10.7910/DVN/IXA7BM">PrimeKG</a>, <a href="../records/4077338">PharmKG</a>, <a href="../records/5361324">OpenBioLink2020 HQ</a>, <a href="../records/7011027">PharMeBINet</a>) and one trivia KG (<a href="https://aclanthology.org/W15-4007.pdf">FB15k-237</a>). All datasets were randomly split into training, validation and test set (80% / 10% / 10%; in the case of PharMeBINet, 99.3% / 0.35% / 0.35% to mitigate the increased inference cost on the larger dataset).</div> <div>On each dataset, five different KGE models were compared: <a href="https://dl.acm.org/doi/10.5555/2999792.2999923">TransE</a>, <a href="https://arxiv.org/abs/1412.6575">DistMult</a>, <a href="https://arxiv.org/abs/1902.10197">RotatE</a>, <a href="https://arxiv.org/abs/2209.08271">TripleRE</a>, <a href="https://dl.acm.org/doi/10.5555/3504035.3504256">ConvE</a>. Hyperparameters were tuned on the validation split (see final train configurations in <code>train/scripts</code>). We release results for tail predictions on the test split. In particular, each test query <code>(h,r,?)</code> is scored against all entities in the KG and we compute the rank of the score of the correct completion <code>(h,r,t)</code> , after masking out scores of other <code>(h,r,t')</code> triples contained in the graph.</div> <div>Note: the ranks provided are computed as the average between the optimistic and pessimistic ranks of triple scores.</div> <div> </div> <div>Inside <code>experimental_data.zip</code>, the following files are provided.</div> <div> <ul> <li><code>datasets/{dataset}</code>: a folder for each dataset, containing <ul> <li><code>{dataset}_preprocessing.ipynb</code>: a Jupyter notebook for downloading and preprocessing the datasets. In particular, this generates the custom label->ID mapping for entities and relations, and the numerical tensor of <code>(h_ID,r_ID,t_ID)</code> triples for all edges in the graph, which can be used to compute graph topological metrics (e.g., using <a href="https://github.com/graphcore-research/kg-topology-toolbox">kg-topology-toolbox</a>) and compare them with the edge prediction accuracy.</li> <li><code>test_ranks.csv</code>: csv table with columns <code>["h", "r", "t"]</code> specifying the head, relation, tail IDs of the test triples, and columns <code>["DistMult", "TransE", "RotatE", "TripleRE", "ConvE"]</code> with the rank of the ground-truth tail in the ordered list of predictions made by the five KGE models;</li> <li><code>entity_dict.csv</code>: list of entity labels, ordered by entity ID (as generated in the preprocessing notebook);</li> <li><code>relation_dict.csv</code>: list of relation labels, ordered by relation ID (as generated in the preprocessing notebook).</li> </ul> </li> <li><code>train</code>: code to reproduce training (and validation) of the five KGE models, using the <a href="https://github.com/graphcore-research/bess-kge">BESS-KGE</a> distribution framework. <ul> <li><code>train/scripts</code>: executable scripts, with specifications of the final hyperparameters for all models and datasets.</li> </ul> </li> <li><code>notebooks</code>: Jupyter notebooks for data analysis and generation of all the figures in the paper.</li> </ul> <p>The separate <code>top_100_tail_predictions.zip</code> archive contains, for each of the test queries in the corresponding <code>test_ranks.csv</code> table, the IDs of the top-100 tail predictions made by each of the five KGE models, ordered by decreasing likelihood. The predictions are released in a <code>.npz</code> archive of numpy arrays (one array of shape <code>(n_test_triples, 100)</code> for each of the KGE models). </p> </div> </div>
NASA GES-DISC Knowledge Graph for Link Prediction
<p>This dataset includes a knowledge graph of NASA GES-DISC collections, featuring interconnected nodes for datasets, data center, projects, platforms, instruments, science keywords, and publications. Designed for link prediction tasks, it aids machine learning research in satellite observation, remote sensing, and climate change. The dataset is in CSV format, ready for graph databases and ML frameworks.</p> <p> </p>
Data from: Predictive links between petal color and pigment quantities in natural Penstemon hybrids
<p class="MsoNormal">Flowers have evolved remarkable diversity in petal color, in large part due to pollinator-mediated selection. This diversity arises from specialized metabolic pathways that generate conspicuous pigments. Despite the clear link between flower color and floral pigment production, studies determining predictive relationships between pigmentation and petal color are currently lacking. In this study, we analyze a dataset consisting of hundreds of natural <em>Penstemon</em> hybrids that exhibit variation in flower color, including blue, purple, pink, and red. For each individual hybrid, we measured anthocyanin pigment content and petal spectral reflectance. We found that floral pigment quantities are correlated with hue, chroma, and brightness as calculated from petal spectral reflectance data: hue is related to the relative amounts of delphinidin vs. pelargonidin pigmentation, whereas brightness and chroma are correlated with the total anthocyanin pigmentation. We used a partial least squares regression approach to identify predictive relationships between pigment production and petal reflectance. We find that pigment quantity data provide robust predictions of petal reflectance, confirming a pervasive assumption that differences in pigmentation should predictably influence flower color. Moreover, we find that reflectance data enables accurate inferences of pigment quantities, where the full reflectance spectra provide much more accurate inference of pigment quantities than spectral attributes (brightness, chroma, and hue). Our predictive framework provides readily interpretable model coefficients relating spectral attributes of petal reflectance to underlying pigment quantities. These relationships represent key links between genetic changes affecting anthocyanin production and ecological functions of petal coloration.</p>
Supplementary dataset for "Evaluating Graph Neural Networks for Link Prediction: Current Pitfalls and New Benchmarking"
<p>The supplementary dataset for the paper "Evaluating Graph Neural Networks for Link Prediction: Current Pitfalls and New Benchmarking". We include the splits for cora, citeseer, and pubmed, the hard negative samples, and the node2vec embeddings. We also include a jupyter file <em>read_data.ipynb</em> to show how to read the non-txt file.</p> <ul> <li>heart_test_samples.npy, heart_valid_samples.npy: the heard negative samples</li> <li>*-n2v-embedding.pt: node2vec embeddings</li> <li>test_samples_index.pt, valid_samples_index.pt: the node index of the selected samples in ogbl-ppa under HeaRT</li> <li>gnn_feature: the input feature of cora, citeseer, pubmed</li> </ul> <p>More details for our code and how to use the dataset are on the code repository: https://github.com/Juanhui28/HeaRT .</p>
Data from: Predictive links between petal color and pigment quantities in natural Penstemon hybrids
Open the record for dataset details and reuse information.
FI_WD20K & SI_WD20K: Datasets For Hyper-Relational Inductive Link Prediction
<p>The datasets correspond to the work presented in <a href="https://arxiv.org/abs/2107.04894">"Improving Inductive Link Prediction Using Hyper-Relational Facts"</a>.</p> <p> </p> <p> </p>
Dataset for "Towards Better Evaluation for Dynamic Link Prediction"
<p>These are the datasets used in <em>Towards Better Evaluation for Dynamic Link Prediction</em></p> <p>For preparing the datasets, we closely follow the baseline methods' data preparation strategy.<br> The original networks are saved as <network>.csv.</p> <p>The networks are formatted as follows:<br> * Each edge is denoted in one line.<br> * Each line has the following format: source_node, destination_node, timestamp, edge_label, comma-separated arrays of edge features.<br> * Please note that if there is no edge label available, the edge_label column will be filled with 0s only for loading purpose; these labels are not used in the link prediction task.<br> * The first line denotes the network format.<br> * Edge features should include at least one feature. If there is no edge feature available, a 0 value is used for all the edges.</p> <p>The network edge-lists are pre-processed for different methods to use them (Specifically, for preprocessing the data, we use the scripts available in "preprocess_data.py" file of the corresponding baseline).<br> Ater preprocessing the network edge-list, there are three files that are used by the models:<br> * <ml_network>.csv: this file contains the timestamp edge-list.<br> * <ml_network>.npy: this file contains the edge features in the dense `npy` format that has the features in binary format.<br> * <ml_network_node>.npy: this file contains the node features in the dense `npy` format that contains the node features in binary format.<br> Please note that when the edge features or node features are absent, we use a vector of zeros is used as the node/edge features in line with the baseline methods.</p>
Exploring links between climatic predictability and the evolution of within- and transgenerational plasticity
<p>In variable environments, phenotypic plasticity can increase fitness by providing tight environment-phenotype matching. However, adaptive plasticity is expected to evolve only when the future selective environment can be predicted based on the prevailing conditions. That is, the juvenile environment should be predictive of the adult environment (within-generation plasticity) or the parental environment should be predictive of the offspring environment (transgenerational plasticity). Moreover, environmental predictability can also shape transient responses such as stress responses in an adaptive direction. Here, we test links between environmental predictability and the evolution of adaptive plasticity by combining time series analyses and a common garden experiment using temperature as a stressor in a temperate butterfly (Melitaea cinxia). Time series analyses revealed that across-season fluctuations in temperature over 48 years are overall predictable. However, within the growing season, temperature fluctuations showed high heterogeneity across years with low autocorrelations and the timing of temperature peaks was asynchronous. Most life-history traits showed strong within-generation plasticity for temperature and traits such as body size and growth rate broke the temperature-size rule. Evidence for transgenerational plasticity, however, was weak and detected for only two traits each in an adaptive and non-adaptive direction. We suggest that the low predictability of temperature fluctuations within the growing season likely disfavours the evolution of adaptive transgenerational plasticity but instead favours strong within-generation plasticity.</p>
Integrated multi-omics analysis of early lung adenocarcinoma links tumor biological features with predicted indolence or aggressiveness
<p>This is a collection of datasets presented in the manuscript "Multi-omics data analysis identifies correlations between tumor biology features and predicted behaviors in early lung adenocarcinoma". This study provides a comprehensive profiling of LUAD indolence and aggressiveness at the biological bulk and single cell levels, as well as at the clinical and radiomics levels. This hypothesis generating study uncovers several potential future research avenues. It also highlights the importance and power of data integration to improve our systemic understanding of LUAD and to help reduce the gap between basic science research and clinical practice.</p>
Freebase Datasets for Robust Evaluation of Knowledge Graph Link Prediction Models
<p><strong>Freebase</strong> is amongst the largest public cross-domain knowledge graphs. It possesses three main data modeling idiosyncrasies. It has a strong <strong>type system</strong>; its properties are purposefully represented in <strong>reverse pairs</strong>; and it uses <strong>mediator objects</strong> to represent multiary relationships. These design choices are important in modeling the real-world. But they also pose nontrivial challenges in research of embedding models for knowledge graph completion, especially when models are developed and evaluated agnostically of these idiosyncrasies. We make available several variants of the Freebase dataset by inclusion and exclusion of these data modeling idiosyncrasies. This is the first-ever publicly available <strong>full-scale</strong> Freebase dataset that has gone through <strong>proper preparation</strong>. </p><p> </p><p>Dataset Details</p><p>The dataset consists of the four variants of Freebase dataset as well as related mapping/support files. For each variant, we made three kinds of files available:</p><ul><li>Subject matter triples file<ul><li><i>fb+/-CVT+/-REV</i> One folder for each variant. In each folder there are 5 files: train.txt, valid.txt, test.txt, entity2id.txt, relation2id.txt Subject matter triples are the triples belong to subject matters domains—domains describing real-world facts.<ul><li>Example of a row in train.txt, valid.txt, and test.txt: <ul><li>2, 192, 0</li></ul></li><li>Example of a row in entity2id.txt:<ul><li>/g/112yfy2xr, 2</li></ul></li><li>Example of a row in relation2id.txt:<ul><li>/music/album/release_type, 192</li></ul></li><li>Explaination<ul><li>"/g/112yfy2xr" and "/m/02lx2r" are the MID of the subject entity and object entity, respectively. "/music/album/release_type" is the realtionship between the two entities. 2, 192, and 0 are the IDs assigned by the authors to the objects.</li></ul></li></ul></li></ul></li><li>Type system file<ul><li><i>freebase_endtypes</i>: Each row maps an edge type to its required subject type and object type.<ul><li>Example<ul><li>92, 47178872, 90</li></ul></li><li>Explanation<ul><li>"92" and "90" are the type id of the subject and object which has the relationship id "47178872".</li></ul></li></ul></li></ul></li><li>Metadata files<ul><li><i>object_types</i>: Each row maps the MID of a Freebase object to a type it belongs to.<ul><li>Example<ul><li>/g/11b41c22g, /type/object/type, /people/person</li></ul></li><li>Explanation<ul><li>The entity with MID "/g/11b41c22g" has a type "/people/person"</li></ul></li></ul></li><li><i>object_names</i>: Each row maps the MID of a Freebase object to its textual label.<ul><li>Example<ul><li>/g/11b78qtr5m, /type/object/name, "Viroliano Tries Jazz"@en</li></ul></li><li>Explanation<ul><li>The entity with MID "/g/11b78qtr5m" has name "Viroliano Tries Jazz" in English.</li></ul></li></ul></li><li><i>object_ids</i>: Each row maps the MID of a Freebase object to its user-friendly identifier.<ul><li>Example<ul><li>/m/05v3y9r, /type/object/id, "/music/live_album/concert"</li></ul></li><li>Explanation<ul><li>The entity with MID "/m/05v3y9r" can be interpreted by human as a music concert live album.</li></ul></li></ul></li><li><i>domains_id_label</i>: Each row maps the MID of a Freebase domain to its label.<ul><li>Example<ul><li>/m/05v4pmy, geology, 77</li></ul></li><li>Explanation<ul><li>The object with MID "/m/05v4pmy" in Freebase is the domain "geology", and has id "77" in our dataset.</li></ul></li></ul></li><li><i>types_id_label</i>: Each row maps the MID of a Freebase type to its label.<ul><li>Example<ul><li>/m/01xljxh, /government/political_party, 147</li></ul></li><li>Explanation<ul><li>The object with MID "/m/01xljxh" in Freebase is the type "/government/political_party", and has id "147" in our dataset.</li></ul></li></ul></li><li><i>entities_id_label</i>: Each row maps the MID of a Freebase entity to its label.<ul><li>Example<ul><li>/g/11b78qtr5m, Viroliano Tries Jazz, 2234</li></ul></li><li>Explanation<ul><li>The entity with MID "/g/11b78qtr5m" in Freebase is "Viroliano Tries Jazz", and has id "2234" in our dataset.</li></ul></li><li><i>properties_id_label</i>: Each row maps the MID of a Freebase property to its label.<ul><li>Example<ul><li>/m/010h8tp2, /comedy/comedy_group/members, 47178867</li></ul></li><li>Explanation<ul><li>The object with MID "/m/010h8tp2" in Freebase is a property(relation/edge), it has label "/comedy/comedy_group/members" and has id "47178867" in our dataset.</li></ul></li></ul></li><li><i>uri_original2simplified</i> and <i>uri_simplified2original</i>: The mapping between original URI and simplified URI and the mapping between simplified URI and original URI repectively.<ul><li>Example<ul><li><i>uri_original2simplified</i><ul><li>"<a href="http://rdf.freebase.com/ns/type.property.unique">http://rdf.freebase.com/ns/type.property.unique</a>": "/type/property/unique"</li></ul></li><li><i>uri_simplified2original</i><ul><li>"/type/property/unique": "<a href="http://rdf.freebase.com/ns/type.property.unique">http://rdf.freebase.com/ns/type.property.unique</a>"</li></ul></li></ul></li><li>Explanation<ul><li>The URI "<a href="http://rdf.freebase.com/ns/type.property.unique">http://rdf.freebase.com/ns/type.property.unique</a>" in the original Freebase RDF dataset is simplified into "/type/property/unique" in our dataset.</li><li>The identifier "/type/property/unique" in our dataset has URI <a href="http://rdf.freebase.com/ns/type.property.unique">http://rdf.freebase.com/ns/type.property.unique</a> in the original Freebase RDF dataset.</li></ul></li></ul></li></ul></li></ul></li></ul>
Exploring links between climatic predictability and the evolution of within- and transgenerational plasticity
Open the record for dataset details and reuse information.
Data from: Link prediction in real-world multiplex networks via layer reconstruction method
Networks are invaluable tools to study real biological, social and technological complex systems in which connected elements form a purposeful phenomenon. A higher resolution image of these systems shows that the connection types do not confine to one but to a variety of types. Multiplex networks encode this complexity with a set of nodes which are connected in different layers via different types of links. A large body of research on link prediction problem is devoted to finding missing links in single-layer (simplex) networks. In recent years, the problem of link prediction in multiplex networks has gained the attention of researchers from different scientific communities. Although most of these studies suggest that prediction performance can be enhanced by using the information contained in different layers of the network, the exact source of this enhancement remains obscure. Here, it is shown that similarity w.r.t. structural features (eigenvectors) is a major source of enhancements for link prediction task in multiplex networks using the proposed Layer Reconstruction Method and experiments on real-world multiplex networks from different disciplines. Moreover, we characterize how low values of similarity w.r.t. structural features result in cases where improving prediction performance is substantially hard.
Data from: Linking functional diversity and ecosystem processes: a framework for using functional diversity metrics to predict the ecosystem impact of functionally unique species
1.Functional diversity (FD) metrics are widely used to assess invasion ecosystem impacts, but we have limited theory to predict how FD should respond to invasion. A key challenge to effectively using FD metrics is the complexity of conceptualizing alterations to multi-dimensional trait space, making it difficult to select a priori the most appropriate metric for specific ecological questions. 2.Here, we provide expectations on how invasion should change four commonly used FD metrics—functional richness (FRic), evenness (FEve), divergence (FDiv), and dispersion (FDis)—and then test these expectations in a lab decomposition experiment. We simulate invasion of a forest by understory plants by adding leaf litter from 18 natives and nonnatives to a representative canopy tree litter mixture to test changes in FD and decomposition. 3.All four metrics changed predictably with invasion. Species that were more functionally unique or when added at greater proportions had larger impacts on FD. Overall, FRic, FEve, and FDiv were poor choices for understanding impacts of nonnative species. FDis was the only metric that both changed predictably with addition of understory litter and correlated intuitively with changes in carbon mineralization. Furthermore, ranking species based upon how much they changed FDis of the litter mixture provided a fair assessment of which species had the largest impact on decomposition. As such, functional dispersion may be a key tool for predicting a priori which nonnatives will have the greatest impact on ecosystem processes. 4.Synthesis: We highlight the need to assess the suitability of each FD metric for the specific ecological question at hand. Our work reveals the pitfalls of considering multiple metrics or randomly choosing a single metric without suitability assessments. At the same time, it suggests a framework for metric assessment that should help lead to selection of a metric or metrics that provide robust a priori insights into how invasion by nonnative species can impact ecosystem processes.
Accretion in the recurrent nova T CrB: Linking the superactive state to the predicted outburst
<p>UBV photometry of the recurrent nova T CrB </p>
Dataset and results for paper: Predicting missing links in directed networks: An investment-profit index
<p>Dataset and raw results for paper: Predicting missing links in directed networks: An investment-profit index. File "All_indices_12_dataset.mat" contains results of all 9 indices in 12 datasets. File "IP_sigma_12_dataset.mat" contains AUC and precision values when parameter sigma changes in IP index.</p>
Dynamic Predictions of the Links Between Psychological and Physical Health of Older Patients in Nursing Home
ClinicalTrials.gov study NCT05729906. IPD Sharing: UNDECIDED. Countries: 1. Publications: 17.
Predictive Toxicity Test Linked to Radiotherapy After Mastectomy and Immediate Implant Reconstruction
ClinicalTrials.gov study NCT04342546. IPD Sharing: NO. Countries: 1. Publications: 6.
Link-HF: Multisensor Non-invasive Telemonitoring System for Prediction of Heart Failure Exacerbation
ClinicalTrials.gov study NCT03037710. IPD Sharing: NO. Countries: 1. Publications: 1.
Data from: Link prediction in real-world multiplex networks via layer reconstruction method
Open the record for dataset details and reuse information.
Data from: Linking functional diversity and ecosystem processes: a framework for using functional diversity metrics to predict the ecosystem impact of functionally unique species
Open the record for dataset details and reuse information.
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.