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.
199
datasets available to search
ShareScore release 0.9.0
Dataset results
199 results for “active learning”
Docked structures from "Optimizing active learning for free energy calculations"
<p>This archive contains the docked TYK2 structures used in the paper "Optimizing active learning for free energy calculations" (<a title="Persistent link using digital object identifier" href="https://doi.org/10.1016/j.ailsci.2022.100050" target="_blank" rel="noreferrer noopener"><span><span>https://doi.org/10.1016/j.ailsci.2022.100050</span></span></a>). AM1-BCC charges are stored in the field "AM1Cache" in the SD file. The charges can be extracted using the code sample below. </p> <p> </p> <pre><code>from rdkit import Chem import base64 import pickle suppl = Chem.SDMolSupplier("10k_most_similar_tyk2_charged.sdf", removeHs=False) for mol in suppl: am1 = mol.GetProp("AM1Cache") am1_charges = pickle.loads(base64.b64decode(mol.GetProp("AM1Cache"))) assert len(am1_charges) == mol.GetNumAtoms(), "Charge cache has different number of charges than mol atoms"</code></pre>
Supplementary data for "Machine learning-based prediction of activity and substrate specificity for OleA enzymes in the thiolase superfamily"
<p>Supplementary data for "Machine learning-based prediction of activity and substrate specificity for OleA enzymes in the thiolase superfamily"</p>
Systematic Data Analysis and Diagnostic Machine Learning Reveal Differences between Compounds with Single- and Multitarget Activity
<p>The deposited files contain balanced data sets of multi-target (MT) and single-target (ST) compounds (CPDs) used for machine learning studies (https://dx.doi.org/10.1021/acs.molpharmaceut.0c00901). The first file (st_mt_data.tsv) contains 15,142 MT- and 15,081 ST-CPDs and the second (st_dt_data.tsv) 1828 DT- and 1776 ST-CPDs. For each CPD, a nonstereo_aromatic_SMILES representation, the original ChEMBL_cid, UniProt (target) IDs, and CPD category (CPD_CAT) (i.e. DT/MT/ST) is provided. DT stands for 'diverse-target' and denotes a subset of MT-CPDs (as detailed in the publication). In addition, a CPD is tagged “Y” if it continued to be present in the data set after removal of 50% randomly selected CPDs or 50% CPD nearest neighbors (NN), respectively.</p>
DFT Calculated xyz and log Files as well as csv Files for Machine Learning in Support of "Tailoring Phosphine Ligands for Improved C H Activation: Insights from Δ-Machine Learning"
<p>Transition metal complexes have played crucial roles in various homogeneous catalytic processes due to their exceptional versatility. This adaptability stems not only from the central metal ions but also from the vast array of choices of the ligand spheres, which form an enormously large chemical space. For example, Rh complexes, with a well-designed ligand sphere, are known to be efficient in catalyzing the C-H activation process in alkanes. To investigate the structure-property relation of the Rh complex and identify the optimal ligand that minimizes the calculated reaction energy ΔE of an alkane C-H activation, we have applied a Δ-Machine Learning method trained on various features to study 1,743 pairs of reactants (Rh(PLP)(Cl)(CO)) and intermediates (Rh(PLP)(Cl)(CO)(H)(propyl)). Our findings demonstrate that the models exhibit robust predictive performance when trained on features derived from electron density (R<sup>2 </sup>= 0.816), and SOAPs (R<sup>2 </sup>= 0.819), a set of position-based descriptors. Leveraging the model trained on xTB-SOAPs that only depend on the xTB-equilibrium structures, we propose an efficient and accurate screening procedure to explore the extensive chemical space of bisphosphine ligands. By applying this screening procedure, <a>we identify ten newly selected reactant-intermediate pairs with an average ΔE </a>of 33.2 kJ mol<sup>-1</sup>, remarkably lower than the average ΔE of the original data set of 68.0 kJ mol<sup>-1</sup>. This underscores the efficacy of our screening procedure in pinpointing structures with significantly lower energy levels.</p> <p>_______________________________________________________________________</p> <p>The dataset contains three file types:</p> <p>Version 1.0:</p> <ol> <li>xyz files of the final optimized Rh-phosphine complexes; one set for the starting materials denoted as "molecule-XXXX_4-times" and one set for the intermediates after C-H activation denoted as "molecule-XXXX_6-times"</li> <li>Gaussian16 log files for the optimization process; one set for the starting materials denoted as "molecule-XXXX_4-times" and one set for the intermediates after C-H activation denoted as "molecule-XXXX_6-times"</li> <li>csv files containing the per molecule features used for training the different machine learning models. The name of the csv files indicates which property was predicted and which model was used</li> </ol> <p>New in version 1.1 (other data is unchanged):</p> <ol> <li>Gaussian16 log files for the ten newly identified bisphosphine ligands; one set for the product material denoted as "LXX_6-times-axial" and one set for the transition state for the C-H activation denoted as "LXX_C-H-activation_TS"</li> </ol>
LungVis1.0: Active learning AI-powered 3D imaging ecosystem for spatial profiling of lung geometry and pulmonary nanoparticle delivery
<p>The imaging dataset was obtained by light sheet fluorescence microscopy on tissue cleared murine lungs. It includes whole lung autofluorence image, particle fluorescence image, and artifical intelligence nnU-Net generated lung airway segments. The dataset provides 78 healthy murine lung strucutre and airway geometry for C57BL/6 mice and offers comprehensive delivery features including qualitative and quantitative analysis on the temporal and spatial inter- and intra-acinar deposition patterns and NP regional dosimetry for four commonly-used routes of pulmonary delivery,namely intranasal liquid aspiration, intratracheal liquid instillation, ventilator-assisted and nose-only aerosol inhalation.</p> <p>Raw LSFM imaging data collection was carried out between 2017-2021, the AI code and generated airway segmention were performed in 2021-2022, the whole datasets were then compiled in 2023. </p> <p>Please ensure to cite our paper for any reuse or reanalysis. Yang, L., Liu, Q., Kumar, P. <em>et al.</em> LungVis 1.0: an automatic AI-powered 3D imaging ecosystem unveils spatial profiling of nanoparticle delivery and acinar migration of lung macrophages. <em>Nat Commun</em> <strong>15</strong>, 10138 (2024). https://doi.org/10.1038/s41467-024-54267-1</p> <p>For any inquiries, please feel free to contact us at lin.yang@helmholtz-munich.de </p>
Supporting data for: "Data-driven discovery of cardiolipin-selective small molecules by computational active learning"
<p>This repository contains supporting data and code for the paper titled "Data-driven discovery of cardiolipin-selective small molecules by computational active learning" by Bernadette Mohr, Kirill Shmilovich, Isabel Kleinwächter, Dirk Schneider, Andrew L.Ferguson, and Tristan Bereau.</p>
Reconstruction of Natural Visual Scenes from Primary Visual Cortical Neural Activity Using Adversarial Learning
<p>This data set and code goes with "Reconstruction of Natural Visual Scenes from Primary Visual Cortical Neural Activity Using Adversarial Learning"</p>
Enhanced Sequence-Activity Mapping and Evolution of Artificial Metalloenzymes by Active Learning
<p>This entry contains data, pretrained models and supplementary files for our enzyme engineering study:</p> <p><strong>Title:</strong> Enhanced Sequence-Activity Mapping and Evolution of Artificial Metalloenzymes by Active Learning<br><strong>Journal:</strong> ACS Central Science</p> <p>If you use any of the data or code in the <a href="https://github.com/lasgroup/ml-protein-design-sav-gold">repository</a>, please cite the paper.</p> <p>The contents of this entry are:</p> <ol> <li>Sequence embeddings needed for reproducing the code are found in data.zip. Unzip the contents of the folder to <code>/data</code> in the code structure. </li> <li>NGS sequencing analysis and raw data in <code>NGS analysis.zip</code></li> <li>10% subset of structures generated with the Rosetta software in <code>structures.zip</code></li> <li>Pretrained and saved models for plotting, clustering and further prediction are saved in <code>models.zip</code></li> <li>Raw assay data including our designed libraries by active learning are found in <code>assay_and_ML_data.zip</code></li> </ol>
Figure 6. Activities dealing with microlearning-Micro Learning: A Modernized Education System
<p>There are many unique activities involved in microlearning, which provides a high rate of successful knowledge transformation. To the inquiry made in research on the activities dealing with microlearning that support knowledge transformation, figure 6 shows that mind mapping has the highest view of respondents, up to 72%, the nearest to it is story telling with 50%. The variation displayed in the figure is significant. 44% of the respondents opt for tagging followed by text with 38%.</p>
Noninvasive detection of macrophage activation with single-cell resolution through machine learning
<p>Data related to the article "Noninvasive detection of macrophage activation with<br> single-cell resolution through machine learning".</p> <p>The package contains 2 folders:<br> - RawData: This package contains raw data and examples of processing to extract the<br> variables employed to train and assess the models.<br> - Variables: This package contains the extracted data from the various experiments<br> showed in the article.</p>
Dataset for "Resonance with subthreshold oscillatory drive organizes activity and optimizes learning in neural networks"
<p>This is the experimental dataset for the paper "Resonance with subthreshold oscillatory drive organizes activity and optimizes learning in neural networks", doi/10.1073/pnas.1716933115.</p>
New Ideas for Brain Modelling 4-Figure 4. LHS relates to neuron binding ensemble mass, with central column activated. RHS relates to hierarchy, with a direct mapping. The two red lines show where the ensemble is missing and so it needs to be learned. The blue lines show extra neurons from the hierarchy back to the ensemble, but can be removed as error. The other paired black squares represent where the patterns match and can oscillate together.
<p>This paper continues the research that considers a new cognitive model based strongly on the human brain, last updated in Greer (2016). In particular, it considers figure 4 of that paper (Figure below) and how it might be useful in practice. The paper also describes some new methods in the areas of image processing and behaviour simulation. The image processing introduces a most classical form of pattern cross-referencing, while the behaviour equations used feedback for a memory-type of cross-referencing. The work is all based on earlier research by the author and the new additions are intended to fit in with the overall design. For image processing, a grid-like structure is used with ‘full linking’, if you like. Each cell in the classifier grid stores a list of all other cells it gets associated with and this is used as the learned image that new input is compared with. For the behaviour metric, a new prediction equation is suggested, as part of a simulation, that uses feedback and history to dynamically determine its current state and course of action. While the new methods are from widely different topics, both can be compared with the binary-analog type of interface that is the main focus of the paper. Sensory input may be static and binary, but cross- references result in variable comparisons that make the input more dynamic. It is suggested that the simplest of linking between a tree and ensemble can explain neural binding and variable signal strengths.</p>
Improving Open Source Face Detection by Combining an Adapted Cascade Classification Pipeline and Active Learning
<p>The <em><strong>EAVISE Open Source Face Detection Dataset</strong></em> consists of several items that were used to generate the improved frontal face detection model using LBP features and AdaBoost for OpenCV3.2.</p> <ul> <li>The annotations of the FDDB dataset, converted to the OpenCV format for doing a correct evaluation.</li> <li>The final trained model (IterativeHardPositives+ model) which is included in the OpenCV 3.2 framework.</li> </ul>
Linked collectors and determiners for: The description of a new species of gall-inducing wasp: a learning activity for primary school students.
Natural history specimen data linked to collectors and determiners held within, "The description of a new species of gall-inducing wasp: a learning activity for primary school students". Claims or attributions were made on Bionomia by volunteer Scribes, <a href="https://bionomia.net/dataset/8fa4af52-5a18-43b0-9cd7-f9887f71a1f9">https://bionomia.net/dataset/8fa4af52-5a18-43b0-9cd7-f9887f71a1f9</a> using specimen data from the dataset aggregated by the Global Biodiversity Information Facility, <a href="https://gbif.org/dataset/8fa4af52-5a18-43b0-9cd7-f9887f71a1f9">https://gbif.org/dataset/8fa4af52-5a18-43b0-9cd7-f9887f71a1f9</a>. Formatted as a Frictionless Data package.
Supplementary materials of the active learning methodology for online engineering education
<p>These are the supplementary materials for the active learning methodology for engineering education integrating online and mobile learning. The methodology is focused on teaching electronics, physical computing, basic robotics, and programming. The purpose of the methodology was to provide active learning, experimentation, and reflection in online classes to the students increasing their motivation and self-efficacy.</p> <p>The file contains the following elements:</p> <ol> <li>Rubric employed to evaluate the student-created videos and blogs.</li> <li>The survey's questions of the different courses that employ online and mobile learning modalities.</li> <li>A set of URLs with examples of the blogs and videos constructed by the students.</li> </ol>
Towards a Diverse Next-Generation Energy Workforce: Teaching Artificial Photosynthesis and Electrochemistry in Elementary Schools through Active Learning
<p>Artificial photosynthesis is a promising approach to generate important commodity chemicals using abundant chemical feedstocks and renewable energy sources. Despite its importance, affordable and effective hands-on classroom activities that demonstrate artificial photosynthesis and teach key concepts, especially for primary school students, is lacking. This will be a critical step in the development of the next-generation energy workforce, especially one that is diverse in race and gender. To aid in this effort, we present an artificial photosynthesis lesson plan based on active-learning techniques that uses safe and highly accessible materials (baking soda, tap water, plastic jars, Ni coil, alligator clips, and a solar cell) to perform solar-powered water splitting. The efficacy of the lesson plan in teaching basic concepts of artificial photosynthesis was evaluated with pre- and post-test data, which shows a statistically significant improvement in overall student understanding. Importantly, the data show that the lesson plan presented here is effective at narrowing the performance gap between minority students and overly represented groups. This study aids in the development and education of a demographically diverse energy workforce through an active learning-based lesson plan for primary school students.</p>
Deep Representation Learning of Physical Activity and Sleep Patterns During Pregnancy Identifies post-hoc Inferences Associated with Prematurity
<p><strong>Running title</strong>: series2signal gestational age "clock" for pregnancy monitoring</p> <p><strong>Summary</strong>: </p> <p>Preterm birth (PTB) is the leading cause of infant mortality globally. While research has focused on the development of predictive models for PTB, cost-effective interventions have remained understudied. Physical activity and sleep present unique opportunities for interventions in low- and middle-income populations. However, objective measurement of physical activity and sleep remains challenging and self-reported metrics suffer from low-resolution and accuracy that decays over time. In this study, we use physical activity data collected using a wearable device comprising over 181, 944 hours of data across N = 1, 083 patients. Using a new state-of-the art deep learning time-series classification architecture, we first develop a ”clock” of healthy dynamics in physical activity patterns during pregnancy by using gestational age (GA) as a surrogate for progression of pregnancy. We also developed a novel interpretability algorithm that integrates unsupervised clustering, model error analysis, feature attribution, and automated actigraphy analysis, allowing for model interpretation with respect to sleep, activity, and static clinical variables. Our model performs significantly better than 7 other machine learning and AI methods for modeling the progression of pregnancy based on measures of physical activity and sleep.</p> <p>Importantly, we found that deviations from this normal ”clock” of physical activity and sleep changes during pregnancy are strongly associated with pregnancy outcomes. When our model underestimates GA, there are 0.52 fewer preterm births than expected (P = 1.01e − 67) and when our model overestimates GA, there are 1.44 times (P = 2.82e − 39) more preterm births than expected. Model error is negatively correlated with interdaily stability (P = 0.043), indicating that our model assigns a more advanced GA when an individual’s daily rhythms are less precise. Supporting this, our model attributes higher importance to sleep periods in predicting higher-than-actual GA, relative to lower-than-actual GA (P = 1.01e − 21). Combining prediction with interpretability allows us to robustly signal when activity behaviors increase or decrease the likelihood of preterm birth and advocates for the future development of clinical decision support through passive monitoring and suggestions around exercise habits and sleep patterns, which are easily implemented in low- and middle-income countries (LMICs). Beyond this particular application, the presented pipeline can be used to analyze high-fidelity time-series data in other translational studies utilizing wearable devices.</p> <p> </p> <p><strong>Data description (brief)</strong>: the raw wearables data is available as .mtn files with the GA encoded in the filename after the underscore. The processed data with sleep annotations can be loaded using the pickle module for serialized objects in python. See https://github.com/nealgravindra/wearables for examples.</p>
Dataset of paper Privacy Orientation during Online Teaching-Learning Activities: Practices Adopted and Lessons Learned
<p>Dataset of the paper accepted for publication in the Journal on Interactive Systems (JIS).</p> <p>DA SILVA, M.; VITERBO, J.; SALGADO, L. C. C.; MOURÃO, E. Privacy Orientation during Online Teaching-Learning Activities:<br> Practices Adopted and Lessons Learned. Journal on Interactive Systems, Porto Alegre, RS, 2023.</p>
Associated data underlying the publication "Flipped Classroom Real-World Activities for Learning Open Computing Concepts"
<p>Various “open” concepts in Computing, such as open standards, open data, open licenses or system interoperability are becoming more important in the professional lives of software engineers. However, students usually do not receive a systematic education about these concepts ; rather they sporadically learn about a subset of these topics. This paper presents the revised version of the Open Computing course in the University of Zagreb, Faculty of Electrical Engineering and Computing (FER), which teaches a clear set of current topics focused on open data and correlating concepts. The new e-learning course is carried out using the “flipped classroom” educational method ; students construct their knowledge in a set of real-world mini-activities throughout the course, instead of passively learning from the official course resources. In this paper, we discuss our flipped classroom activities, their relation to revised Bloom’s taxonomy of educational objectives, and present the evaluation of the first course instance us- ing this model. Students’ feedback shows they welcome this change in approach, finding the course useful and interesting, and preferring this method to the traditional full lecture setting.</p>
Dataset - Speeding up high-throughput characterization of materials libraries by active learning: autonomous electrical resistance measurements
<p>With the trend towards multinary materials and the associated increase in measurement time, there is a clear need for increasing the efficiency of measurement procedures. In systems requiring long materials characterization times, the implementation of active learning can help decreasing the measurement duration significantly. This dataset is part of the publication in Digital Discovery under the same title and holds the algorithm as well as the data used to test its performance. The algorithm leverages an active learning approach with a Gaussian process model capable of selecting the next measurement area of a library of materials based on the highest uncertainty. Ten materials libraries were manufactured by magnetron sputtering, the composition was measured with EDX and the electrical resistance was measured using the described test stand. The code can also be found on <a href="https://gitlab.ruhr-uni-bochum.de/fthelen/auto-resist-meas">Gitlab</a>.</p>
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.