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.
37
datasets available to search
ShareScore release 0.9.0
Dataset results
37 results for “interpretable machine learning”
CLRD-GLPS: A Long-term Seasonal Dataset of Ruminant Livestock Distribution in China's Grazing Production Systems (2000-2021) Using Stacking-based Interpretable Machine Learning
<p>Advanced computational methods integrating ensemble learning with interpretable machine learning are essential for precision livestock management under increasing environmental constraints and food security pressures. This study develops a novel stacking-based interpretable machine learning (IML) framework that combines multiple algorithms with SHAP analysis techniques to generate the China's Long-term Ruminant Livestock Distribution in Grazing Livestock Production Systems (CLRD-GLPS) dataset. Our computational approach addresses critical challenges in livestock distribution modelling: livestock segmentation and spatial prediction accuracy. The framework integrates Random Forest, XGBoost, CatBoost, LightGBM, and Extra Trees through a two-layer stacking architecture, enhanced with SHAP (Shapley Additive Explanations) analysis for model interpretability. We also implemented interpretable machine learning for livestock production system segmentation to distinguish grazing from total livestock populations. The stacking ensemble demonstrated superior performance over individual algorithms, achieving R² values of 0.954-0.961 for cattle and 0.896-0.901 for sheep and goats, with improvements of up to 8.3% compared to best performance single-model approaches. Multi-scale validation confirmed computational robustness: livestock segmentation achieved R² = 0.80 at county level, while independent city-level validation of CLRD-GLPS datasets yielded R² = 0.76-0.80. SHAP interpretability analysis revealed distinct environmental drivers, with vegetation indices and topography primarily influencing cattle distribution, while snow conditions and elevation dominated sheep and goat patterns. This computational framework advances livestock distribution modelling through enhanced prediction accuracy, model stability, and interpretability, while the CLRD-GLPS dataset provides essential spatial-temporal information for rangeland sustainability assessments and evidence-based livestock management policies. This dataset is supported by the Second Tibetan Plateau Scientific Expedition and Research Program (STEP, grant no. 2019QZKK0906).</p>
Supplementary material for "Towards a metagenomics machine learning interpretable model for understanding the transition from adenoma to colorectal cancer"
<p>Supplementary files for "Towards a metagenomics machine learning interpretable model for understanding the transition from adenoma to colorectal cancer".</p>
Mapping a novel metric for Flash Flood Recovery using Interpretable Machine Learning
<p>This data is supplementary to the paper titled "Mapping a novel metric for Flash Flood Recovery using Interpretable Machine Learning". The file contains the main results.<br><br>For any queries, please visit <a href="https://hydrosense.iitd.ac.in" target="_blank" rel="noopener">Hydrosense Lab (IIT Delhi)</a>.</p>
Data from: Cornerstones are the key stones: Using interpretable machine learning to probe the clogging process in 2D granular hoppers
Open the record for dataset details and reuse information.
Data for NHESS manuscript by Biass et al. (2022): Insights into the vulnerability of vegetation to tephra fallouts from interpretable machine learning and big Earth observation data
<p>This repository contains the data produced in the context of the following paper:</p> <blockquote> <p>Biass S, Jenkins SF, Aberhard WH, Delmelle P, Wilson T (2022): Insights into the vulnerability of vegetation to tephra fallouts from interpretable machine learning and big Earth observation data, Accepted in NHESS</p> </blockquote> <p>Naming convention is: `run_date`_`landcover`_`impact_metrics`_`VI`_`anomaly`_test.pkl, where:</p> <ul> <li>Landcover is either crops, shrubs, herbaceous vegetation (grass), forests (trees) or all together</li> <li>Impact metrics is either minV (impact magnitude) or minT (impact duration)</li> <li>VI is the vegetation index (here, EVI)</li> <li>Anomaly is the impact indicator (here, cumulative difference index)</li> </ul> <p>Refer to the associated paper for more information on the methodology.</p> <p>Files are saved as .pkl and were generated by the <a href="https://explainerdashboard.readthedocs.io/en/latest/">explainerdashboard</a> library. They are the result of <a href="https://xgboost.readthedocs.io/en/stable/">XGBoost</a> runs that were optimised with <a href="https://optuna.org">Optuna</a> and analysed with the <a href="https://shap.readthedocs.io/en/latest/">SHAP</a> library. They contain:</p> <ol> <li>The explanatory variables and observed and computed target variables for all features</li> <li>The SHAP values</li> </ol> <p>To load the files, use <a href="https://explainerdashboard.readthedocs.io/en/latest/cli.html?highlight=load#explainerdashboard.explainers.BaseExplainer.from_file">this method</a>.</p> <p> </p>
RepliChrom: Interpretable machine learning predicts cancer-associated enhancer-promoter interactions using DNA replication timing
<p>This dataset accompanies the study "RepliChrom: Interpretable machine learning predicts cancer-associated enhancer-promoter interactions using DNA replication timing". The study introduces RepliChrom, a computational framework designed to predict enhancer–promoter interactions (EPIs) by leveraging multi-scale replication timing (RT) signals. This approach addresses the fundamental challenge of distinguishing gene targets regulated by distal enhancers from those activated by proximal transcriptional activity-a key problem in understanding the causal basis of complex diseases.</p> <p>Despite recent advances in high-throughput technologies such as Hi-C, ChIA-PET, and Hi-TrAC that allow genome-wide reconstruction of 3D chromatin architecture, the role of DNA replication timing in mediating these spatial interactions remains underexplored. RepliChrom fills this gap by using cell-type-specific RT profiles as predictive features for chromatin interaction inference.</p> <p>To support model development, training, and evaluation, we provide a comprehensive multi-omics dataset covering six human cell lines (K562, GM12878, HeLaS3, IMR90, NHEK, and HUVEC ), encompassing:</p> <p>Hi-C datasets: Processed chromatin interaction loops used to define positive and negative enhancer–promoter interaction pairs.</p> <p>ChIA-PET datasets: Interaction data anchored around transcription factor binding, including POLR2A and CTCF, used for model validation across different interaction types.</p> <p>Hi-TrAC datasets: Targeted chromatin accessibility-derived interaction data, offering complementary validation of the model on alternate platforms.</p> <p>Replication Timing (RT) data: Processed RT signal profiles for each cell line, used to extract multi-scale temporal features as inputs for RepliChrom.</p> <p>These datasets enable reproducibility of the model training process and serve as benchmark resources for future research into DNA replication–mediated regulation of 3D genome architecture.</p> <p><strong>Included Files</strong></p> <p>Hi-C_datasets.zip (2.23 MB): Processed Hi-C interaction pairs for six cell types.</p> <p>ChIA-PET_datasets.zip (818.18 KB): CTCF and POLR2A ChIA-PET interactions across multiple lines.</p> <p>Hi-TrAC_datasets.zip (196 bytes): Hi-TrAC-based chromatin interaction training datasets across multiple lines.</p> <p>Cellline_RT_data.zip (86.17 MB): Replication timing signal data across multiple human cell types for multi-scale replication timing feature extraction.</p> <p><strong>Usage</strong><br>All datasets are intended for academic, non-commercial use. The provided files can be directly used to reproduce the training and evaluation of RepliChrom, and may also support broader applications in enhancer–promoter modeling, replication-timing analysis, and 3D genomics studies. For detailed usage instructions and code implementation, please refer to the GitHub repository: https://github.com/DaoFuying/RepliChrom</p>
Supporting Information for "Machine Learning interpretation of the correlation between infrared emission features of interstellar polycyclic aromatic hydrocarbons"
<p>This is the supporting information for the article "Machine Learning interpretation of the correlation between infrared emission features of interstellar polycyclic aromatic hydrocarbons". It contains:</p> <p>In Supporting_Information.pdf:</p> <p>1. A map of the correlation between descriptors.</p> <p>2. A spectral distribution.</p> <p>3. An explanation of the file example_code.zip.</p> <p>In example_code.zip:</p> <p>An example code of a machine learning model based on ECFP and corresponding input files.</p>
[Replication package] Explainable Human-Machine Teaming using Model Checking and Interpretable Machine Learning
<p>Anonymized replication package of submission #1917: "Explainable Human-Machine Teaming using Model Checking and Interpretable Machine Learning".</p> <p>See README.md for further instructions.</p>
Dataset for publication: "Using explainable machine learning to interpret the effects of policies on air pollution: COVID-19 lockdown in London"
<p>This online repository offers supplementary datasets supporting the findings in the research article titled "Using explainable machine learning to interpret the effects of policies on air pollution: COVID-19 lockdown in London," published in Environmental Science & Technology. The dataset contains air quality data from different monitoring sites in London between 2016 and 2020 and weather data for the same period. Additionally, the dataset incorporates 136 features relating to London's Middle Layer Super Output Areas (MSOAs) in the year 2019, which can be used to identify the key factors contributing to the heterogeneous changes in air quality levels at different spatial locations during the pandemic. Please refer to the metadata file for detailed data sources and descriptions. </p>
Code for: Comparison and interpretability of machine learning models to predict severity of chest injury
<p><span><span><span><span><span><span><span><span><span><span><span><b>Objective:</b> Trauma quality improvement programs and registries improve care and outcomes for injured patients. Designated trauma centers calculate injury scores using dedicated trauma registrars; however, many injuries arrive at non-trauma centers, leaving a substantial amount of data uncaptured. We propose automated methods to identify severe chest injury using machine learning (ML) and natural language processing (NLP) methods from the electronic health record (EHR) for quality reporting.</span></span></span></span></span></span></span></span></span></span></span></p> <p><span><span><span><span><span><span><span><span><span><span><span><b>Materials and Methods:</b> A level I trauma center was queried for patients presenting after injury between 2014 and 2018. Prediction modeling was performed to classify severe chest injury using a reference dataset labeled by certified registrars. Clinical documents from trauma encounters were processed into concept unique identifiers for inputs to ML models: logistic regression with elastic net regularization (EN), extreme gradient boosted machines (XGB), and convolutional neural networks (CNN). The optimal model was identified by examining predictive and face validity metrics using global explanations.</span></span></span></span></span></span></span></span></span></span></span></p> <p><span><span><span><span><span><span><span><span><span><span><span><b>Results:</b> Of 8,952 encounters, 542 (6.1%) had a severe chest injury. CNN and EN had the highest discrimination, with an area under the receiver operating characteristic curve of 0.93 and calibration slopes between 0.88 and 0.97. CNN had better performance across risk thresholds with fewer discordant cases. Examination of global explanations demonstrated the CNN model had better face validity, with top features including "contusion of lung" and "hemopneumothorax." </span></span></span></span></span></span></span></span></span></span></span></p> <p><span><span><span><span><span><span><span><span><span><span><span><b>Discussion: </b>The CNN model featured optimal discrimination, calibration, and clinically relevant features selected. </span></span></span></span></span></span></span></span></span></span></span></p> <p><span><span><span><span><span><span><span><span><span><span><span><b>Conclusion:</b> NLP and ML methods to populate trauma registries for quality analyses are feasible.</span></span></span></span></span></span></span></span></span></span></span></p>
Data for "Generative and interpretable machine learning for aptamer design and analysis of in vitro sequence selection"
<p>Once decompressed, the file contains a folder which contains:</p> <ul> <li>The files "s100_Nth.fasta" (where "N" is 5, 6, 7 or 8), which are the output of the SELEX experiment described in the paper with DOI: <a href="https://doi.org/10.1002/cbic.201900265">10.1002/cbic.201900265</a>. They are standard fasta files, and the descriptor of each sequence is of the form "seqX-Y", where "X" is an increasing label, and "Y" is the number of times "seqX" has been obtained (number of counts of "seqX").</li> <li>The file "Aptamer_Exp_Results.csv", which contains the sequences tested experimentally for the paper "Generative and interpretable machine learning for aptamer design and analysis of in vitro sequence selection" (preprint available at https://doi.org/10.1101/2022.03.12.484094), with the following experimental results for each sequence: (i) whether the sequence was able to bind thrombin ('B' for binders, 'NB' for non-binders); (ii) the thrombin exosite used for binding ('I' for exosite I, 'II' for exosite II, 'n/a' for sequences not tested).</li> </ul> <p>Examples of usage of the data are available at https://github.com/adigioacchino/RBMsForAptamers.</p>
VirMAD - Interpretable Machine Learning for Bacterial Defense and Viral Antidefense Pair Inference
Open the record for dataset details and reuse information.
Dataset for the paper 'Predicting mechanical properties of polycrystalline nanopillars by interpretable machine learning'
<div> <div>This dataset contains the data produced for the above paper. The dataset consists of:</div> <div> </div> <div>- input nanopillars before deformation (molecular_dynamics/nanopillars)</div> <div>- stress-strain curves acquired by deforming the nanopillars (molecular_dynamics/stress_strain_curves)</div> <div>- weights of the CNNs trained to predict mechanical properties of the nanopillars (machine_learning/train_CNN)</div> <div>- Grad-CAM fields of the predictions (machine_learning/train_CNN)</div> <br> <div>Codes used for creating and analyzing the dataset are available at https://github.com/tekoivisto/nanopillar-ML</div> </div>
Quantifying the Effects of Vagus Nerve Stimulation on Gastric Myoelectric Activity in Ferrets Using an Interpretable Machine Learning Approach
<p>Raw data</p>
Code for: Comparison and interpretability of machine learning models to predict severity of chest injury
Open the record for dataset details and reuse information.
Driving factors of HONO and their synergy in Hohhot, a semi-arid city in northern China: Insights from interpretable machine learning approaches
<p>Research data includes air pollutants observation data and meteorological parameters data.</p>
Replication Data for: Interpretable machine learning prediction of fire emission and comparison with FireMIP process-based models
<p>The target and predictor variables used in the developed ML model.</p>
A Multi-Omics Interpretable Machine Learning Model Reveals Modes of Action of Small Molecules
GEO Series GSE129144. Mus musculus. 54 samples. Type: Expression profiling by high throughput sequencing; Genome binding/occupancy profiling by high throughput sequencing.
Revealing the Grammar of Small RNA Secretion Using Interpretable Machine Learning III
GEO Series GSE230010. Homo sapiens. 3 samples. Type: Other.
Revealing the Grammar of Small RNA Secretion Using Interpretable Machine Learning I
GEO Series GSE230006. Homo sapiens. 16 samples. Type: Non-coding RNA profiling by high throughput sequencing.
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.