Skip to main content
ConductScienceScienceDex
Powered by ShareScore

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.

153

datasets available to search

ShareScore release 0.9.0

Reset

Dataset results

153 results for “Target prediction”

Learn how ShareScore rates datasets ↗
zenodo44/100

predicted microRNA target sites (miRanda)

<p>Target predictions based on the miRanda algorithm. The target sites are scored for likelihood of mRNA downregulation using mirSVR, a regression model that is trained on sequence and contextual features of the predicted miRNA::mRNA duplex. Expression profiles are derived from a comprehensive sequencing project of a large set of mammalian tissues and cell lines of normal and disease origin.</p> <p>This collection contains the following datasets from the August 2010 release of <a href="http://www.microrna.org">microRNA.org</a>:</p> <ul> <li>16228619 predicted microRNA target sites in 34911 distinct 3&#39;UTR from isoforms of 19898 human genes</li> <li>7459149 predicted microRNA target sites in 28287 distinct 3&#39;UTR from isoforms of 19231 mouse genes</li> <li>586068 predicted microRNA target sites in 6865 distinct 3&#39;UTR from isoforms of 6256 rat genes</li> <li>345671 predicted microRNA target sites in 12285 distinct 3&#39;UTR from isoforms of 10532 fruitfly genes</li> </ul>

opencc-byMay 2020View details →
zenodo44/100

KiSSim: Predicting off-targets from structural similarities in the kinome

<p><strong>KiSSim: Predicting off-targets from structural similarities in the kinome</strong></p> <p><strong>Project description.</strong></p> <p>KiSSim (Kinase Structural&nbsp;Similarity) is&nbsp;a novel fingerprint designed specifically for kinase pockets, allowing for similarity studies across the structurally covered kinome. The kinase fingerprint is based on the <a href="https://klifs.net/">KLIFS</a>&nbsp;pocket alignment, which defines 85 pocket residues for all kinase structures. This enables a residue-by-residue comparison without a computationally expensive alignment step.</p> <p>The pocket fingerprint encodes each pocket residue&rsquo;s spatial and physicochemical properties. The spatial properties describe the residue&rsquo;s position in relation to the kinase pocket center and important kinase subpockets, i.e. the hinge region, the DFG region, and the front pocket. The physicochemical properties encompass for each residue its size and pharmacophoric features, solvent exposure, and side chain orientation.</p> <p>Some datasets are not part of the `kissim_app` GitHub repository due to their size but can be downloaded from here to the respective kissim_app folders.</p> <p><strong>Data.</strong></p> <ul> <li>`20210902_KLIFS_HUMAN.tar.gz` --- save in `kissim_app/data/external/structures`</li> <li>`complete_SiteAlign.txt.gz` --- save in `kissim_app/data/external/sitealign`</li> </ul> <p><strong>Results.</strong></p> <ul> <li>`results.tar.bz2`--- save as `kissim_app/results`</li> </ul> <p>These are the KiSSim results:&nbsp;fingerprints,&nbsp;feature/fingerprint distances, kinase matrices, and kinase trees&nbsp;for structures in all (`all`), DFG-in (`dfg_in`), and DFG-out (`dfg_out`)&nbsp;conformation. In the case of the DFG-in conformation, we also have KiSSim runs with fingerprint subsets based on only residues that interact with certain ligands in KLIFS IFPs: Erlotinib (`dfg_in_IRE`), Imatinib (`dfg_in_STI`), Bosutinib (`dfg_in_DB8`), and Dopamapimod (`dfg_in_B96`). The folder contains README with a detailed file list.</p> <p><strong>Usage.</strong></p> <p>This dataset can be used to run the notebooks available on&nbsp;<a href="https://github.com/volkamerlab/kissim_app">https://github.com/volkamerlab/kissim_app</a>.</p> <ol> <li>Clone the kissim_app&nbsp;repository.</li> <li>Download the files provided here.</li> <li>If applicable, extract the archive content to the&nbsp;folders as indicated above and run the notebooks.</li> </ol> <pre><code class="language-bash">cd /path/to/your/download tar -xvf results.tar.bz2 -C /path/to/kissim_app/ tar -xvf 20210902_KLIFS_HUMAN.tar.bz2 -C /path/to/kissim_app/data/external/structures/ # In case you want the raw SiteAlign data mv complete_SiteAlign.txt.gz /path/to/kissim_app/data/external/sitealign</code></pre> <p><strong>Citation.</strong></p> <p>These&nbsp;datasets are&nbsp;part of the KiSSim publication: TBA</p>

openmit-licenseDec 2021View details →
zenodo44/100

Ion Implantation Sensor and Process Target Data for Predicting Ion Beam Tuning in Semiconductor Manufacturing

<h2><strong>Dataset Description:</strong></h2> <p>This dataset is designed to predict ion beam tuning setup processes in semiconductor manufacturing, in terms of tuning success or failure, and tuning duration. It is split into&nbsp;<strong><code>X</code></strong> and <code><strong>y</strong></code> to allow for supervised learning approaches.</p> <ul> <li><code><strong>X</strong></code> represents the current equipment condition and the process targets of the currently processed and the upcoming lot, as defined within recipes.</li> <li><code><strong>y</strong></code> represents the ion beam tuning setup report, which informs about the tuning success ratio and tuning duration. These setups are necessary, when switching between recipes to prepare the equipment for processing the next lot.&nbsp;<strong><code>y</code></strong> contains three labels, enabling classification of (1) tuning success or fail, and (2) prolonged tuning, as well as (3) estimation of tuning duration as a regression task.</li> </ul> <p>About <strong><code>X</code></strong>:</p> <p>Each lot is processed with a specific recipe to achieve the process target. The tuning takes place before the first wafer of the to-be-tuned recipe is processed. Each row in <strong><code>X</code></strong> includes logistical information such as the equipment used for processing and parsed recipe / process target information for the current and upcoming lot. The majority of data consists out of aggregated metrics of equipment-internally tracked sensor traces, recording physical parameters such as gas flows, temperatures, voltages and currents. When analyzed in conjunction with the processed recipe, these sensors provide insights into the current equipment condition.&nbsp;</p> <p>About <code><strong>y</strong></code>:</p> <p>The&nbsp;<code>setup_result</code> column indicates the success or failure of tuning - with <code>setup_result=0</code> indicating tuning success, while&nbsp;<code>setup_result=1</code> signals tuning failure. If the first tuning attempt fails, there may be follow-up attempts, but these are not included in this dataset. The&nbsp;<code>duration</code> column represents the tuning duration in seconds, as used for regression analysis. The&nbsp;<code>duration_interval</code> column is a binary label for prolonged tunings, i.e. <code>duration_interval=1</code> for instances, which take more than 6 minutes to tune.</p> <p>For reproducibility of the corresponding paper's results:</p> <ol> <li>The dataset contains the same carefully curated subset of features.</li> <li>The train_test_split() has already been performed, thus we provide&nbsp;<code>x_train</code> and <code>x_valid</code> separately.</li> <li>To reduce the effect of outliers in the data, the sensor data has already been scaled, as derived from&nbsp;<code>x_train</code>.</li> </ol> <p>In summary, these datasets (<code><strong>X</strong></code>, <code><strong>y</strong></code>) provide comprehensive information for predicting ion beam tuning in semiconductor manufacturing, making it a valuable resource for researchers and practitioners in the field.</p> <h2><strong>Python Code for Reproducibility:</strong></h2> <p>Furthermore, we share a jupyter notebook <code>ionbeamtuning.ipynb</code> with Python code to train the best performing model on the provided data, as described in the paper. To execute the code, you may need to install any missing packages specified in the <code>requirements.txt</code>, as indicated within the notebook.</p>

opencc-by-4.0Apr 2024View details →
zenodo44/100

Datasets for "Advancing Drug-Target Interactions Prediction: Leveraging a Large-Scale Dataset with a Rapid and Robust Chemogenomic Algorithm"

<p>All datasets required to reproduce the results of publication "Drug-Target Interactions Prediction at Scale: the Komet Algorithm with the LCIdb Dataset"</p>

opencc-by-4.0Feb 2024View details →
zenodo44/100

Predictive models for off-target binding profiles generation

<p>Models for predicting off-target binding, built with Conformal Prediction, and the <a href="http://cpsign-docs.genettasoft.com">CPSign software</a>. The dataset is part of an upcoming publication (Manuscript in preparation), which will provide more details.</p> <p>The dataset is a GZipped Tar archive, with the models as Java Archive (JAR) files. For every JAR-file, there is also a corresponding audit log, with the extension &quot;.audit.json&quot;, produced by the workflow software (<a href="http://scipipe.org">SciPipe</a>) used to train the models. This audit file contains all the shell commands used in the workflow that produced the models.</p>

opencc-by-4.0Jun 2018View details →
zenodo40/100

Data and analysis scripts of "Predictability awareness rather than mere predictability enhances the perceptual benefits for targets in auditory rhythms over targets following temporal cues"

<p>These are the data and the analysis scripts accompanying the publication&nbsp;</p> <p>&quot;Predictability awareness rather than mere predictability enhances the perceptual benefits for targets in auditory rhythms over targets following temporal cues&quot;.&nbsp;</p> <p>Check the readMe for an instruction.</p>

opencc-by-4.0Apr 2022View details →
zenodo40/100

Industry-scale Application and Evaluation of Deep Learning for Drug Target Prediction

<p>Artificial intelligence (AI) is undergoing a revolution thanks to the breakthroughs of machine learning algorithms in computer vision, speech recognition, natural language processing and generative modelling. Recent works on publicly available pharmaceutical data showed that AI methods are highly promising for Drug Target prediction. However, the quality of public data might be different than that of industry data due to different labs reporting measurements, different measurement techniques, fewer samples and less diverse and specialized assays. As part of a European funded project (ExCAPE), that brought together expertise from pharmaceutical industry, machine learning, and high-performance computing, we investigated how well machine learning models obtained from public data can be transferred to internal pharmaceutical industry data. Our results show that machine learning models trained on public data can indeed maintain their predictive power to a large degree when applied to industry data. Moreover, we observed that deep learning derived machine learning models outperformed comparable models, which were trained by other machine learning algorithms, when applied to internal pharmaceutical company datasets. To our knowledge, this is the first large-scale study evaluating the potential of machine learning and especially deep learning directly at the level of industry-scale settings and moreover investigating the transferability of publicly learned target prediction models towards industrial bioactivity prediction pipelines.</p>

opencc-by-4.0Jun 2019View details →
dryad40/100

In silico subcellular targeting predictions for cytosolic aminoacyl tRNA-synthetases (aaRS) in parasitic plants

<p>Eukaryotic nuclear genomes often encode distinct sets of protein translation machinery for function in the cytosol vs. organelles (mitochondria and plastids). This phenomenon raises questions about why multiple translation systems are maintained even though they are capable of comparable functions, and whether they evolve differently depending on the compartment where they operate. These questions are particularly interesting in land plants because translation machinery, including aminoacyl-tRNA synthetases (aaRS), is often dual-targeted to both the plastids and mitochondria. These two organelles have quite different metabolisms, with much higher rates of translation in plastids to supply the abundant, rapid-turnover proteins required for photosynthesis. Previous studies have indicated that plant organellar aaRS evolve more slowly compared to mitochondrial aaRS in other eukaryotes that lack plastids. Thus, we investigated the evolution of nuclear-encoded organellar and cytosolic translation machinery across a broad sampling of angiosperms, including non-photosynthetic (heterotrophic) plant species with reduced rates of plastid gene expression to test the hypothesis that translational demands associated with photosynthesis constrain the evolution of bacterial-like enzymes involved in organellar tRNA metabolism. Remarkably, heterotrophic plants exhibited wholesale loss of many organelle-targeted aaRS and other enzymes, even though translation still occurs in their mitochondria and plastids. These losses were often accompanied by apparent retargeting of cytosolic enzymes and tRNAs to the organelles, sometimes preserving aaRS-tRNA charging relationships but other times creating surprising mismatches between cytosolic aaRS and mitochondrial tRNA substrates. Our findings indicate that the presence of a photosynthetic plastid drives the retention of specialized systems for organellar tRNA metabolism.</p>

opencc-zeroAug 2023View details →
zenodo40/100

Molecularly targetable cell types in mouse visual cortex have distinguishable prediction error responses

<p>Raw data and code to reproduce figures in the manuscript &quot;Molecularly targetable cell types in mouse visual cortex have distinguishable prediction error responses&quot;</p> <p># README</p> <p>## Introduction</p> <p>This README provides essential information about the codebase for the manuscript titled &quot;Molecularly targetable cell types in mouse visual cortex have distinguishable prediction error responses.&quot; The code in this repository is self-contained and is expected to run smoothly given the appropriate versions of the required libraries/packages.</p> <p>## Directory structure and execution details</p> <p>### R code</p> <p>- Main Figures 2A-2D, 3A-3C, and 4A-4E, as well as supplemental figures S2A-S2H, S3A-S3E, S4L, and S5A-S5I, were generated using R. Execute the `R_figs_master.r` script located in the `r_code` directory.<br> - All figures will be saved within the `r_code/code_generated_figures` directory.<br> - Note: Exact UMAP representations might vary across different hardware and operating systems, likely due to an issue with the UWOT package ([Reference Issue](https://github.com/satijalab/seurat/issues/5514)). If figures appear outside their designated plot ranges, set &quot;FixAxes&quot; to &#39;FALSE&#39; in the `single_cell_variables.r` script.</p> <p>### MATLAB code</p> <p>- Main figures 1B, 1D-1F, and 6A-6H, as well as supplemental figures S1A-S1J and S6A-S6I, were generated using MATLAB (version 9.11.0.1809720 (R2021b) Update 1). Execute the `get_the_figs_matlab.m` script located in the `matlab_code` directory.<br> - All figures will be saved within the `matlab_code/code_generated_figures` directory.<br> - Required: [fca_readfcs, version 2020.06.22](https://ch.mathworks.com/matlabcentral/fileexchange/9608-fca_readfcs).</p> <p>### Python code</p> <p>- Figures 5B-5F panels were generated using Python (version 3.6.8). Run the `fig_5_analysis_code.py` script located in the `python_code` directory.<br> - All figures will be saved within the `python_code/code_generated_figures` directory.<br> - The preprocessed images located in `python_code/data_repository/Adamts2_processed`, `python_code/data_repository/Agmat_processed`, and `python_code/data_repository/Baz1a_processed` were generated using the ImageJ macro `python_code/cropped_to_processed_macro.ijm` from the raw images in `python_code/data_repository/Adamts2_cropped`, `python_code/data_repository/Agmat_cropped`, and `python_code/data_repository/Baz1a_cropped`.</p> <p>## Supplementary code (for reference only as raw data is not included)</p> <p>### Mapping code and genome construction code</p> <p>- Initial processing of Single-cell RNA-sequencing was performed with Cell Ranger, coordinated by the Python script:<br> &nbsp; `python_code/mapping_and_genome_construction/single_cell_mapping_pipeline.py`. Some components of this script are deprecated and were primarily used to pass .fastq files to Cell Ranger and organize the outputs.<br> - A custom genome was constructed to account for the expression of CaMPARI2 in the single-cell RNA-sequencing dataset:<br> &nbsp; `python_code/mapping_and_genome_construction/campari2_genome_construction.py`.<br> - Processing of Bulk RNA-sequencing, either single or paired-end, was executed through Python:<br> &nbsp; `python_code/mapping_and_genome_construction/bulk_single_end_mapping.py` and `python_code/mapping_and_genome_construction/bulk_paired_end_mapping.py`.<br> - A custom genome was constructed to account for the expression of various artificial promoter viruses:<br> &nbsp; `python_code/mapping_and_genome_construction/bulk_seq_genome_construction.py`.</p>

opencc-by-4.0Aug 2023View details →
zenodo40/100

Supplementary Datasets for "Genome-wide CRISPR off-target prediction and optimization using RNA-DNA interaction fingerprints"

<p>Supplementary Datasets for "Genome-wide CRISPR off-target prediction and optimization using RNA-DNA interaction fingerprints". The deposition contains training/testing&nbsp;datasets used in the article.</p>

opencc-by-4.0Oct 2023View details →
dryad40/100

In silico subcellular targeting predictions for cytosolic aminoacyl tRNA-synthetases (aaRS) in parasitic plants

Open the record for dataset details and reuse information.

publicOct 2024View details →
dryad40/100

'CellTrajectory' for cellular automata modelling of leukaemic stem cell dynamics in acute myeloid leukaemia: insights into predictive outcomes and targeted therapies

Open the record for dataset details and reuse information.

publicJan 2025View details →
dryad40/100

Does the tail show when the nose knows? AI-enhanced analysis of tail kinematics outperforms human experts at predicting when detection dogs find their target odor

Open the record for dataset details and reuse information.

publicMar 2025View details →
dryad40/100

'Biosim' for cellular automata modelling of leukaemic stem cell dynamics in acute myeloid leukaemia: insights into predictive outcomes and targeted therapies

Open the record for dataset details and reuse information.

publicJan 2025View details →
zenodo36/100

The dataset used in the article "Evidential Deep Learning for Guided Drug-Target Interaction Prediction"

<p>The file consists of two parts, the first part is the feature file used for model training and the second part is the saved model.</p> <p>bond_angle file is the feature file for the drugbank dataset, KIBA file is the feature file for the kiba dataset, davis is the feature file for the davis dataset. case_drugbank is the dungbank feature file for the independent test set and case_len_drug is the feature file for the patent validation dataset in the TKIs feature file.len_vs is the feature file for TKIs virtual screening dataset.</p> <p>In the runs folder, the models saved by random division on the Drugbank, KIBA, and Davis datasets are stored.</p> <p>The <a href="https://zenodo.org/api/records/14056305/draft/files/bert_weights_encoderMedium_10.h5/content" target="_blank" rel="noopener noreferrer">bert_weights_encoderMedium_10.h5</a> is the parameter for the small molecule pre-training model, MG-BERT.</p>

opencc-by-4.0Oct 2024View details →
zenodo36/100

Efficient Drug-Target Interactions Prediction Framework via Transferable Knowledge Fusing

<p><span>Drug-target integrations (DTI) prediction is a niche in drug discovery, streamlining the search for potential drugs. Computer-aided drug discovery (CADD) has gained traction for its precise predictions, efficiency, and adaptability across various situations. Yet, the computational demands of current top CADD models hinder their practical use due to heavy resource needs.</span></p> <p><span>In this research, we introduce TransFusE DTI, an effective framework for predicting DTIs that leverages pre-trained knowledge to construct models that optimize predictive accuracy while minimizing computational demands. The encoder uses a pre-extracted embedding vector from ProtBERT to reduce computational load and adapts a smaller ProtBERT model. It also includes target-related functional text to boost predictive accuracy. We evaluate the performance of TransFusE DTI using three widely-recognized benchmark datasets: BIOSNAP, DAVIS, and BindingDB, and compare its results to prior studies.</span></p> <p><span>Our results demonstrate that TransFusE DTI exhibits superior predictive performance on the BIOSNAP and BindingDB datasets. Notably, the model's parameter count is only 60% of that of the previous top-performing model by </span><span><a href="https://www.mdpi.com/1999-4923/14/8/1710"><span>Kang et al. (2022)</span></a></span><span>, and it operates efficiently with a learning rate of 26%. Furthermore, the model's video memory requirement is 11.2 GB, rendering it suitable for use on general-purpose Graphics Processing Units (GPUs). </span></p>

opencc-by-4.0Dec 2023View details →
zenodo36/100

Generative AI in the Advancement of Viral Therapeutics for Predicting and Targeting Immune-Evasive SARS-CoV-2 Mutations

<p>This dataset&nbsp;<strong>encompasses</strong> and describes the following features:</p> <ul> <li>Mutations in viruses like SARS-CoV-2 can make them escape vaccines and treatments.</li> <li>Accurately predicting these mutations is crucial for developing effective countermeasures.</li> <li>The study uses a type of AI called a Generative Adversarial Network (GAN) to analyze the virus's spike protein,&nbsp;which plays a key role in infection.</li> <li>The GAN generates protein sequences similar to natural ones,&nbsp;but which are also likely to evade immune responses.</li> <li>By analyzing these generated sequences,&nbsp;the researchers improve their AI model's ability to predict real-world escape mutations.</li> <li>This improved prediction could help design better vaccines and treatments, and prepare for future viral threats.</li> </ul>

opencc-by-4.0Jul 2024View details →
zenodo36/100

PIGNet: A physics-informed deep learning model toward generalized drug-target interaction predictions

<p>Training, test datasets of the paper &quot;PIGNet: A physics-informed deep learning model toward generalized drug-target interaction predictions&quot;.</p>

opencc-by-4.0Feb 2022View details →
zenodo36/100

Presentation scripts of "Predictability awareness rather than mere predictability enhances the perceptual benefits for targets in auditory rhythms over targets following temporal cues"

<p>These are the presentation scripts accompanying the publication&nbsp;</p> <p>&quot;Predictability awareness rather than mere predictability enhances the perceptual benefits for targets in auditory rhythms over targets following temporal cues&quot;.&nbsp;</p> <p>Check the readMe for an instruction.</p>

opencc-by-4.0Apr 2022View details →
zenodo36/100

CASP15 RNA target predictions with RoseTTAFold-NA

<p>Structure predictions of CASP-15 RNA targets made using RoseTTAFold-NA on December 9, 2022.</p>

opencc-by-4.0Jan 2023View details →

ScienceDex guides

Understand access before you commit

These curated guides explain access requirements, typical timelines, costs, and reuse considerations for widely used research datasets.

Compare curated 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.

allen-brain-atlas
neuroscienceopenDocumentation, web resources, and API references are available online.
Last verified 2026-04-30Open record

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.

abode-home-cage
behavioral-neuroscienceopenThe DataShare record exposes download links for annotations, documentation, license text, and the zipped per-snippet data directory.
Last verified 2026-04-30Open record

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.

dandi-nwb
electrophysiologyopenPublished Dandiset metadata and archive endpoints are available through the production DANDI API.
Last verified 2026-04-30Open record

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.

ibl
behavioral-neuroscienceopenPublic sessions can be searched and loaded from the IBL public data server through ONE.
Last verified 2026-04-29Open record

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.

openneuro
neuroscienceopenPublished datasets are available on demand over the internet.
Last verified 2026-04-29Open record

ScienceDex for faculty

Your academic CV and NIH biosketch

Make your CV in your school's format and your NIH biosketch from it, then download both as Word files.