hugging-science/dataset-quest-index · Datasets at Hugging Face

id stringlengths 32 32	created_at timestamp[us]date 2025-09-18 11:40:40 2025-09-20 08:04:57	dataset_name stringlengths 6 107	dataset_url stringlengths 21 139	description stringlengths 25 847	approx_size float64 0 1k	size_unit stringclasses 3 values	field stringlengths 3 99	user stringlengths 1 25
7bdf8e576b9b4c3daaf9d587b429b2f6	2025-09-18T11:40:40	PET/CT Imaging	https://github.com/microsoft/lymphoma-segmentation-dnn/tree/main; https://microsoft.github.io/aiforgoodlab/resource.html?d=PETCT-Imaging-15	Lymphoma lesion segmentation and quantitation from FDG PET/CT images using deep neural networks	0	GB	Healthcare Imaging	anonymous
2c8ab04f729d4bbca922acbaf56dde3d	2025-09-18T13:14:22	MSnLib	https://www.nature.com/articles/s41592-025-02813-0	A large-scale, open multi-stage fragmentation mass spectral library for 30,000 unique small molecules.	21.6	GB	Natural products discovery, drug discovery, medicinal chemistry, structural chemistry, Metabolomics	Ludovico Comito
460f8634d2324de9a4912e5ea28a7133	2025-09-18T14:13:58	Reporting of Surgically Removed Lymph Nodes for Breast Tumors	https://meridian.allenpress.com/aplm/article/146/11/1308/487735/Reporting-of-Surgically-Removed-Lymph-Nodes-for	evidence-based pathology reporting of different types of cancers, with the inclusion of all parameters deemed to be relevant for best patient care and future data collection	0	GB	Healthcare	Pranjal Dubey
47d7e22550c440f5951961f3a6c0ce33	2025-09-18T14:16:51	Nuclear Fusion Dataset	https://www.kaggle.com/datasets/adebusayoadewunmi/nuclearfusion-data/data	This data was simulated to give a view of what a nuclear fusion experiment data would be like and it does not hold any representation of any real-life experiment. But the features were selected based on the creator's understanding of some components that would be necessary for a successful ignition.	35	MB	Magnetic Field Fluctutations, Leak Percentage, Plasma Instabilities, etc.	anonymous
96d131a55c084040b83a9de4bf07a1fb	2025-09-18T14:20:19	Dataset of Solution-based Inorganic Materials Synthesis Procedures Extracted from the Scientific Literature	https://doi.org/10.6084/m9.figshare.16583387.v4	here are 20,037 hydrothermal synthesis reactions and 15,638 precipitation synthesis reactions. Each record corresponds to a synthesis procedure extracted from a paragraph and is represented as an individual JSON objec	161	MB	Material Synthesis	Pranjal Dubey
4a78a154811845ba902184a1d37fa3f4	2025-09-18T14:23:31	Human Reference Interactome (HuRI)	https://interactome-atlas.org/	Human protein–protein interactome from large scale yeast two hybrid screens (HI-III-20). ~52k curated binary PPIs across ~8.3k proteins with QC and assay metadata. Good for link prediction, graph learning, and protein function modeling. Check source for redistribution limits.	162	MB	interactomics; proteomics; systems biology; bioinformatics	-
44510458008b4e65b983f34f96d518a9	2025-09-18T14:23:33	Human Reference Interactome (HuRI)	https://interactome-atlas.org/	Human protein–protein interactome from large scale yeast two hybrid screens (HI-III-20). ~52k curated binary PPIs across ~8.3k proteins with QC and assay metadata. Good for link prediction, graph learning, and protein function modeling. Check source for redistribution limits.	162	MB	interactomics; proteomics; systems biology; bioinformatics	-
dccf6b447e934e5eba2bb0e0f31def88	2025-09-18T14:28:47	Human Reference Interactome (HuRI)	https://www.interactome-atlas.org/	Human protein–protein interactome from large scale yeast two hybrid screens (HI-III-20). ~52k curated binary PPIs across ~8.3k proteins with QC and assay metadata. Good for link prediction, graph learning, and protein function modeling. Check source for redistribution limits.	162	MB	interactomics; proteomics; systems biology; bioinformatics	-
e2773a85b36d4b0e9876bbbd1eedaddc	2025-09-18T14:29:32	Human Reference Interactome (HuRI)	https://www.interactome-atlas.org/	Human protein–protein interactome from large scale yeast two hybrid screens (HI-III-20). ~52k curated binary PPIs across ~8.3k proteins with QC and assay metadata. Good for link prediction, graph learning, and protein function modeling. Check source for redistribution limits.	162	MB	interactomics; proteomics; systems biology; bioinformatics	Samuel
cacc9d4e686c43cdbd700e1ebfbef5df	2025-09-18T14:43:12	Human Reference Interactome (HuRI)	https://interactome-atlas.org/	Human protein–protein interactome from large scale yeast two hybrid screens (HI-III-20). ~52k curated binary PPIs across ~8.3k proteins with QC and assay metadata. Good for link prediction, graph learning, and protein function modeling. Check source for redistribution limits.	162	MB	interactomics; proteomics; systems biology; bioinformatics	Samuel
e7d8117398a04d87bb3870cbe0afcb91	2025-09-18T15:00:43	Awadhi Dataset	https://github.com/kmi-linguistics/awadhi	POS Tagged Awadhi Dataset	26	MB	Linguistics	Rishav Chandra Varma
121c2b53fdd547d68e8202d518b1b876	2025-09-18T15:06:13	Bhojpuri - Voice Text Dataset	https://huggingface.co/datasets/KonthouKabiAI/bhojpuri-lr-v3	Audio - Text pairs of Bhojpuri Dataset of 24.6k pairs	7	GB	Linguistics	Rishav Chandra Varma
3d0cac830c6b4eff8545468af7815f88	2025-09-18T15:08:36	English-Bhojpuri SMT System: Insights from the Karaka Model	https://github.com/shashwatup9k/BHLTR/tree/master	Bhojpuri POS Tagged Dataset and Parallel Corpora Dataset	20	MB	Linguistics	Rishav Chandra Varma
c7ef4f68480c4cb189086f31c7f1badb	2025-09-18T16:48:26	Protein Data Bank (PDB)	https://www.rcsb.org/	The Protein Data Bank (PDB) is the global archive of experimentally determined 3D structures of proteins, nucleic acids, and complex assemblies, widely used for structural biology and drug discovery research. Check source for redistribution limits. Data is fragmented across many file formats, schema changes over time, no consistent ML-friendly embeddings. Metadata parsing requires domain-specific parsers.	2	TB	Structural Biology, Bioinformatics, Drug Discovery	anonymous
fa7f8dd520e24de486ea4f8127c7432a	2025-09-18T16:49:36	IFBench	https://huggingface.co/collections/allenai/ifbench-683f590687f61b512558cdf1	One of the most used benchmarks for instruction following, yet HF dataset card is quite dry. Not directly a science dataset but given its importance for the industry I believe it deserves a better HF page.	200	MB	LLM	Lucas Fernandes Martins
b0eb9affc50449dca24863b8c20baf15	2025-09-18T16:58:40	PubChem BioAssay Dataset	https://pubchem.ncbi.nlm.nih.gov/source/	Large-scale dataset containing bioactivity results for millions of chemical compounds tested in high-throughput screening assays. Provides assay outcomes, compound structures, and associated metadata. Check source for redistribution limits. Critical resource for training ML models in drug discovery, virtual screening, and activity prediction. Enables foundation models for bioactivity and chemical property prediction.	150	GB	Chemistry, Drug Discovery, Bioinformatics	Naga Adithya Kaushik
314286105a074c64a8d1709efd9fe16d	2025-09-18T17:03:48	NASA Exoplanet Archive	https://exoplanetarchive.ipac.caltech.edu/	Catalog of confirmed exoplanets, stellar parameters, orbital characteristics, and light curves collected from multiple NASA missions. Provides both tabular metadata and time-series photometry.	10	TB	Astrophysics, Space Science	Naga Adithya Kaushik
a4df3c0954bf4e45ad3fcb88901627e9	2025-09-18T17:50:43	NASA’s Solar Dynamics Observatory (SDO)	https://huggingface.co/nasa-ibm-ai4science/Surya-1.0	Surya is a 366M-parameter transformer model pretrained on 9 years (≈218 TB) of multi-instrument data from NASA’s Solar Dynamics Observatory (SDO), including 8 Atmospheric Imaging Assembly (AIA) channels and 5 Helioseismic and Magnetic Imager (HMI) products.	218	TB	Astrophysics, Space Science, Heliophysics	anonymous
aed2ada9cd274485956f2df52fe8b03e	2025-09-18T18:25:10	BindingDB	https://www.bindingdb.org/rwd/bind/chemsearch/marvin/Download.jsp	BindingDB is a FAIRsharing knowledgebase that accelerates biomedical research and innovation by extracting, accumulating, organizing, and annotating protein-ligand interaction datasets.	3	GB	ligand-prot interactions	anonymous
a2f68ce72b914f34bc18ad6f154e0233	2025-09-18T19:37:03	BubbleML_2	https://huggingface.co/datasets/hpcforge/BubbleML_2	BubbleML_2 is a high-fidelity dataset of boiling simulations in 2D for three fluids (FC-72, Liquid N2 and R515B). It provides paired time-series fields stored in HDF5 (.hdf5) files together with metadata (.json).	2.35	TB	PDEs	Sheikh Md Shakeel Hassan
a183190b77584a0a8ce79908c5319133	2025-09-18T20:01:26	Rubin's Legacy Survey of Space and Time (LSST)	https://data.lsst.cloud/	The Rubin Observatory’s Legacy Survey of Space and Time (LSST) is a 10-year astronomical survey scheduled to begin in 2025 at Cerro Pachón in Chile, using an 8.4-meter telescope and the world’s largest 3.2-gigapixel digital camera. It will scan the southern sky every few nights in six optical bands (u, g, r, i, z, y), producing about 15 terabytes of data per night and ultimately a dataset exceeding tens of petabytes. By building both a deep static map of the universe and a time-domain “movie of the sky,” LSST will enable discoveries ranging from the nature of dark matter and dark energy to the inventory of solar system bodies, the structure of the Milky Way, and the detection of rare, fast-changing cosmic transients.	1,000	TB	Astronomy, Cosmology, Physics	anonymous
73a136bf76e74e25b5153c92922b64a4	2025-09-18T20:03:17	Rubin's Legacy Survey of Space and Time (LSST)	https://data.lsst.cloud/	The Rubin Observatory’s Legacy Survey of Space and Time (LSST) is a 10-year astronomical survey scheduled to begin in 2025 at Cerro Pachón in Chile, using an 8.4-meter telescope and the world’s largest 3.2-gigapixel digital camera. It will scan the southern sky every few nights in six optical bands (u, g, r, i, z, y), producing about 15 terabytes of data per night and ultimately a dataset exceeding tens of petabytes. By building both a deep static map of the universe and a time-domain “movie of the sky,” LSST will enable discoveries ranging from the nature of dark matter and dark energy to the inventory of solar system bodies, the structure of the Milky Way, and the detection of rare, fast-changing cosmic transients.	1,000	TB	Astronomy, Cosmology, Physics	Steven Dillmann
e4eb7e1ce61a41348d49b1d86b3558a9	2025-09-18T20:05:36	Multimodal Universe	https://huggingface.co/MultimodalUniverse	The Multimodal Universe dataset is a large scale collection of multimodal astronomical data, including images, spectra, and light curves, which aims to enable research into foundation models for astrophysics and beyond.	100	TB	Astronomy, Cosmology, Physics	Steven Dillmann
a92f734164304dd799e9914f7dded1e1	2025-09-18T20:11:53	Chandra Source Catalog (CSC)	https://cda.cfa.harvard.edu/cscweb/index.do	The Chandra Source Catalog (CSC) is a virtual X-ray astrophysics facility that enables both detailed individual source studies and statistical studies of large samples of X-ray sources detected in ACIS and HRC-I imaging observations obtained by the Chandra X-ray Observatory. The catalog provides carefully-curated, high-quality, and uniformly calibrated and analyzed tabulated positional, spatial, photometric, spectral, and temporal source properties, as well as science-ready X-ray data products.	40	GB	Astronomy, Cosmology, Physics	Steven Dillmann
08b4872696ea4451aa65363ff22902c3	2025-09-18T20:12:12	Chandra Source Catalog (CSC)	https://cda.cfa.harvard.edu/cscweb/index.do	The Chandra Source Catalog (CSC) is a virtual X-ray astrophysics facility that enables both detailed individual source studies and statistical studies of large samples of X-ray sources detected in ACIS and HRC-I imaging observations obtained by the Chandra X-ray Observatory. The catalog provides carefully-curated, high-quality, and uniformly calibrated and analyzed tabulated positional, spatial, photometric, spectral, and temporal source properties, as well as science-ready X-ray data products.	40	TB	Astronomy, Cosmology, Physics	Steven Dillmann
83c03815e5784d1fb3aceb7d3c98b25d	2025-09-19T01:33:58	Smith42/galaxies	https://huggingface.co/datasets/Smith42/galaxies	8.6 million galaxy images from the DESI Legacy Survey, crossmatched with galaxy metadata.	1.1	TB	Astronomy	Mike Smith
5d30e4f6963f4241895c2cb1abfa4c6a	2025-09-19T02:34:11	Multimodal Universe	https://github.com/MultimodalUniverse/MultimodalUniverse	The Multimodal Universe dataset is a large scale collection of multimodal astronomical data, including images, spectra, and light curves, which aims to enable research into foundation models for astrophysics and beyond.	120	TB	Astronomy	Mike Smith
2f4525f502d14595b7a16c417407fd53	2025-09-19T02:41:17	ROVER-Gen	https://huggingface.co/datasets/cheryyunl/ROVER-Gen	Logical puzzle solving and instruction following dataset for large vision models. Given an image and the task (prompt), the dataset consists of a final result (image) after executing the task on this image.	0	GB	logical puzzle solving, visual instruction following	SB
5e633ca577df48368ca9a95832f081d9	2025-09-19T02:48:20	bird-species	https://huggingface.co/datasets/Ez-Clap/bird-species	This dataset contains images of Belgian birds and their species. Dataset is split into train (6.78k rows), validation (215 rows) and test (215 rows).	163	MB	bird-species	SB
28b52169b90c495e870e898681491c13	2025-09-19T02:54:33	Wild Life Dataset	https://lila.science/datasets/wcscameratraps	This data set contains approximately 1.4M camera trap images representing around 675 species from 12 countries, making it one of the most diverse camera trap data sets available publicly. Data were provided by the Wildlife Conservation Society. The most common classes are tayassu pecari (peccary), meleagris ocellata (ocellated turkey), and bos taurus (cattle).	0	MB	wildlife	SB
67c59f7c54774d119ac61d65be1903bd	2025-09-19T06:57:31	ECOTOX	https://cfpub.epa.gov/ecotox/	ECOTOX is a comprehensive Knowledgebase providing single chemical environmental toxicity data on aquatic and terrestrial species.	0	GB	chemistry	Rahul Maity
fc394bc9e68f496382c259081b497c8e	2025-09-19T08:40:06	emg2pose	https://github.com/facebookresearch/emg2pose	A dataset of Surface electromyography (sEMG) recordings paired with ground-truth, motion-capture recordings of the hands. Data loading, baseline model training, and baseline model evaluation code are provided.	431	GB	Surface Electromyography	MatteoFasulo
c585ff9115574c759db6cde676465c75	2025-09-19T08:42:11	Generic non-invasive neuromotor interface for HCI	https://fb-ctrl-oss.s3.amazonaws.com/generic-neuromotor-interface-data	Surface electromyography (sEMG) data and training models associated with the paper "A generic non-invasive neuromotor interface for human-computer interaction". The dataset contains sEMG recordings from 100 participants in each of the three tasks described in the paper: discrete_gestures, handwriting, and wrist.	150	GB	Surface Electromyography	MatteoFasulo
3915860e8db542adaee2ae21e1275376	2025-09-19T08:56:58	clinical_trials_gov	https://clinicaltrials.gov	Clinical Trial Study information is crucial for drug development/repurposing and while the information is publicly available, it comes in really bad formats which varies for different diseases (as each has different endpoint) and different studies (Phase I/II/III). The information is available within data vendors but behind proprietary walls	0	GB	medicine	anonymous
5499c9b6ab8d4a36b86eef244e3a6970	2025-09-19T08:57:53	clinical_trials_gov	https://clinicaltrials.gov	Clinical Trial Study information is crucial for drug development/repurposing and while the information is publicly available, it comes in really bad formats which varies for different diseases (as each has different endpoint) and different studies (Phase I/II/III). The information is available within data vendors but behind proprietary walls	0	GB	medicine	anonymous
98b0da12a1c9439198fb6aefb2becd8c	2025-09-19T09:04:55	clinical_trials_gov	https://clinicaltrials.gov	Clinical Trial Study information is crucial for drug development/repurposing and while the information is publicly available, it comes in really bad formats which varies for different diseases (as each has different endpoint) and different studies (Phase I/II/III). The information is available within data vendors but behind proprietary walls	4	GB	healthcare	Piotr Kaniewski
64d8d78da2b3451781c828ac199d7ed8	2025-09-19T09:14:12	pubmed_literature	https://pubmed.ncbi.nlm.nih.gov/download/	Biomedical Literature is crucial for drug discovery and while its available at API service, its not available in a single place in a structured format.	30	GB	biomedicine	Piotr Kaniewski
71a05fedb7ac4519a05af58171e1fb61	2025-09-19T09:16:30	daily_med_labels	https://dailymed.nlm.nih.gov/dailymed/	daily med contains labels of fda approved drugs/medicines yet its not availalbe in a structured format that doesnt require manual scraping	30	GB	healthcare	Piotr Kaniewski
7c8408a366e54a4892e90627b0597678	2025-09-19T09:38:28	ontology_mapping	https://www.ebi.ac.uk/ols4/	Ontologies are essential in knowledge discovery/bioinformatics; yet mapping between them is a non trivial task	10	GB	bioinformatics	Piotr Kaniewski
c2849283ae1a41d9b6c7a20c98098588	2025-09-19T14:32:01	GeneReviews	https://huggingface.co/datasets/Tonic/GeneReviews	GeneReviews Dataset Extraction This project extracts text and metadata from GeneReviews® chapters downloaded from NCBI Bookshelf and creates a structured dataset in Hugging Face format. Overview GeneReviews® is an international point-of-care resource for clinicians, providing clinically relevant and medically actionable information for inherited conditions. This project processes the XML files from the GeneReviews database and creates a structured dataset suitable for machine learning and research applications. 📈 Dataset Statistics: Total Records: 929 GeneReviews chapters Average Abstract Length: 899.6 characters Average Content Length: 56,377.9 characters Total References: 13,683 references across all chapters Average References per Chapter: 14.7 Chapters with >100 references: 12 chapters Total Keywords: 9,616 Unique Keywords: 6,824	16	MB	Clinical reviews of genetic conditions	Joseph [open/acc] Pollack
6ae0293a58324f64a91e33d895c61e14	2025-09-19T18:51:33	GV-Rep	https://zenodo.org/records/11502840	A large-scale Genetic variants (GVs) dataset featuring a comprehensive collection of 7 million GV records with detailed annotations and one clinician verified dataset. Paper: https://arxiv.org/abs/2407.16940 , Github: https://github.com/bowang-lab/genomic-FM	30	GB	Genomics	Mohammed Hamdy
75e0431bb75241baa8f876820ff0fedc	2025-09-19T19:03:50	Biology-Instructions	https://github.com/hhnqqq/Biology-Instructions	A large-scale multi-omics biological sequences-related instruction-tuning dataset including DNA, RNA, proteins, and multi-molecules. Paper: https://arxiv.org/abs/2412.19191	100	MB	Multi-omics	Mohammed Hamdy
71fcbca937a0443a920606be6dcc4060	2025-09-19T19:05:48	International Brain Lab Data	http://internationalbrainlab.com/data; https://www.biorxiv.org/content/10.1101/2023.07.04.547681v4	Multiple Brain-Related Datasets. A key challenge is that this dataset contains data from multiple experiments and different modalities. The dataset comprises 459 Neuropixel experimental sessions encompassing 699 distinct recordings (probe insertions). These recordings were obtained from 139 subjects performing the IBL task across 12 different laboratories. Spike-sorting analysis yielded 621,733 units, of which 75,708 were classified as good quality. In total, 241 brain regions were recorded(from their website).	50	TB	Multisite Neural Recording Local Field Potential(LFP)	Ken Fahmidur Rahman
a77b360636984b4ea3391c1757cdca50	2025-09-19T19:34:53	Insect-1M	https://uark-cviu.github.io/projects/insect-foundation/	Large-scale Insect Dataset comprising more than a million images with hierarchical labels from the high to the low taxonomy level, including class, order, family, genus, and species. Paper: https://arxiv.org/abs/2502.09906	100	GB	Entomology	Mohammed Hamdy
92b0add1ee9b446fb234f13438ea93e0	2025-09-20T08:04:57	ToT-Biology	https://huggingface.co/datasets/moremilk/ToT-Biology	The ToT-Biology dataset emphasizes mechanistic understanding and explanatory biological reasoning, rather than just providing correct answers. It aims to train AI models in interpretability and logical deduction within the biological realm. Spanning a wide range of biological complexities, it starts with foundational concepts in cell biology, genetics, and ecology, and progresses to advanced areas like systems biology, synthetic biology, and computational biophysics. The dataset incorporates real-world applications across diverse fields such as medicine, agriculture, environmental science, and bioengineering, ensuring practical relevance alongside rigorous theoretical exploration.	80	MB	Biology	SHIVAM DUBEY

7bdf8e576b9b4c3daaf9d587b429b2f6

2025-09-18T11:40:40

PET/CT Imaging

https://github.com/microsoft/lymphoma-segmentation-dnn/tree/main; https://microsoft.github.io/aiforgoodlab/resource.html?d=PETCT-Imaging-15

Lymphoma lesion segmentation and quantitation from FDG PET/CT images using deep neural networks

0

GB

Healthcare Imaging

anonymous

2c8ab04f729d4bbca922acbaf56dde3d

2025-09-18T13:14:22

MSnLib

https://www.nature.com/articles/s41592-025-02813-0

A large-scale, open multi-stage fragmentation mass spectral library for 30,000 unique small molecules.

21.6

GB

Natural products discovery, drug discovery, medicinal chemistry, structural chemistry, Metabolomics

Ludovico Comito

460f8634d2324de9a4912e5ea28a7133

2025-09-18T14:13:58

Reporting of Surgically Removed Lymph Nodes for Breast Tumors

https://meridian.allenpress.com/aplm/article/146/11/1308/487735/Reporting-of-Surgically-Removed-Lymph-Nodes-for

evidence-based pathology reporting of different types of cancers, with the inclusion of all parameters deemed to be relevant for best patient care and future data collection

0

GB

Healthcare

Pranjal Dubey

47d7e22550c440f5951961f3a6c0ce33

2025-09-18T14:16:51

Nuclear Fusion Dataset

https://www.kaggle.com/datasets/adebusayoadewunmi/nuclearfusion-data/data

This data was simulated to give a view of what a nuclear fusion experiment data would be like and it does not hold any representation of any real-life experiment. But the features were selected based on the creator's understanding of some components that would be necessary for a successful ignition.

35

MB

Magnetic Field Fluctutations, Leak Percentage, Plasma Instabilities, etc.

anonymous

96d131a55c084040b83a9de4bf07a1fb

2025-09-18T14:20:19

Dataset of Solution-based Inorganic Materials Synthesis Procedures Extracted from the Scientific Literature

https://doi.org/10.6084/m9.figshare.16583387.v4

here are 20,037 hydrothermal synthesis reactions and 15,638 precipitation synthesis reactions. Each record corresponds to a synthesis procedure extracted from a paragraph and is represented as an individual JSON objec

161

MB

Material Synthesis

Pranjal Dubey

4a78a154811845ba902184a1d37fa3f4

2025-09-18T14:23:31

Human Reference Interactome (HuRI)

https://interactome-atlas.org/

Human protein–protein interactome from large scale yeast two hybrid screens (HI-III-20). ~52k curated binary PPIs across ~8.3k proteins with QC and assay metadata. Good for link prediction, graph learning, and protein function modeling. Check source for redistribution limits.

162

MB

interactomics; proteomics; systems biology; bioinformatics

-

44510458008b4e65b983f34f96d518a9

2025-09-18T14:23:33

Human Reference Interactome (HuRI)

https://interactome-atlas.org/

Human protein–protein interactome from large scale yeast two hybrid screens (HI-III-20). ~52k curated binary PPIs across ~8.3k proteins with QC and assay metadata. Good for link prediction, graph learning, and protein function modeling. Check source for redistribution limits.

162

MB

interactomics; proteomics; systems biology; bioinformatics

-

dccf6b447e934e5eba2bb0e0f31def88

2025-09-18T14:28:47

Human Reference Interactome (HuRI)

https://www.interactome-atlas.org/

Human protein–protein interactome from large scale yeast two hybrid screens (HI-III-20). ~52k curated binary PPIs across ~8.3k proteins with QC and assay metadata. Good for link prediction, graph learning, and protein function modeling. Check source for redistribution limits.

162

MB

interactomics; proteomics; systems biology; bioinformatics

-

e2773a85b36d4b0e9876bbbd1eedaddc

2025-09-18T14:29:32

Human Reference Interactome (HuRI)

https://www.interactome-atlas.org/

Human protein–protein interactome from large scale yeast two hybrid screens (HI-III-20). ~52k curated binary PPIs across ~8.3k proteins with QC and assay metadata. Good for link prediction, graph learning, and protein function modeling. Check source for redistribution limits.

162

MB

interactomics; proteomics; systems biology; bioinformatics

Samuel

cacc9d4e686c43cdbd700e1ebfbef5df

2025-09-18T14:43:12

Human Reference Interactome (HuRI)

https://interactome-atlas.org/

Human protein–protein interactome from large scale yeast two hybrid screens (HI-III-20). ~52k curated binary PPIs across ~8.3k proteins with QC and assay metadata. Good for link prediction, graph learning, and protein function modeling. Check source for redistribution limits.

162

MB

interactomics; proteomics; systems biology; bioinformatics

Samuel

e7d8117398a04d87bb3870cbe0afcb91

2025-09-18T15:00:43

Awadhi Dataset

https://github.com/kmi-linguistics/awadhi

POS Tagged Awadhi Dataset

26

MB

Linguistics

Rishav Chandra Varma

121c2b53fdd547d68e8202d518b1b876

2025-09-18T15:06:13

Bhojpuri - Voice Text Dataset

https://huggingface.co/datasets/KonthouKabiAI/bhojpuri-lr-v3

Audio - Text pairs of Bhojpuri Dataset of 24.6k pairs

7

GB

Linguistics

Rishav Chandra Varma

3d0cac830c6b4eff8545468af7815f88

2025-09-18T15:08:36

English-Bhojpuri SMT System: Insights from the Karaka Model

https://github.com/shashwatup9k/BHLTR/tree/master

Bhojpuri POS Tagged Dataset and Parallel Corpora Dataset

20

MB

Linguistics

Rishav Chandra Varma

c7ef4f68480c4cb189086f31c7f1badb

2025-09-18T16:48:26

Protein Data Bank (PDB)

https://www.rcsb.org/

The Protein Data Bank (PDB) is the global archive of experimentally determined 3D structures of proteins, nucleic acids, and complex assemblies, widely used for structural biology and drug discovery research. Check source for redistribution limits. Data is fragmented across many file formats, schema changes over time, no consistent ML-friendly embeddings. Metadata parsing requires domain-specific parsers.

2

TB

Structural Biology, Bioinformatics, Drug Discovery

anonymous

fa7f8dd520e24de486ea4f8127c7432a

2025-09-18T16:49:36

IFBench

https://huggingface.co/collections/allenai/ifbench-683f590687f61b512558cdf1

One of the most used benchmarks for instruction following, yet HF dataset card is quite dry. Not directly a science dataset but given its importance for the industry I believe it deserves a better HF page.

200

MB

LLM

Lucas Fernandes Martins

b0eb9affc50449dca24863b8c20baf15

2025-09-18T16:58:40

PubChem BioAssay Dataset

https://pubchem.ncbi.nlm.nih.gov/source/

Large-scale dataset containing bioactivity results for millions of chemical compounds tested in high-throughput screening assays. Provides assay outcomes, compound structures, and associated metadata. Check source for redistribution limits. Critical resource for training ML models in drug discovery, virtual screening, and activity prediction. Enables foundation models for bioactivity and chemical property prediction.

150

GB

Chemistry, Drug Discovery, Bioinformatics

Naga Adithya Kaushik

314286105a074c64a8d1709efd9fe16d

2025-09-18T17:03:48

NASA Exoplanet Archive

https://exoplanetarchive.ipac.caltech.edu/

Catalog of confirmed exoplanets, stellar parameters, orbital characteristics, and light curves collected from multiple NASA missions. Provides both tabular metadata and time-series photometry.

10

TB

Astrophysics, Space Science

Naga Adithya Kaushik

a4df3c0954bf4e45ad3fcb88901627e9

2025-09-18T17:50:43

NASA’s Solar Dynamics Observatory (SDO)

https://huggingface.co/nasa-ibm-ai4science/Surya-1.0

Surya is a 366M-parameter transformer model pretrained on 9 years (≈218 TB) of multi-instrument data from NASA’s Solar Dynamics Observatory (SDO), including 8 Atmospheric Imaging Assembly (AIA) channels and 5 Helioseismic and Magnetic Imager (HMI) products.

218

TB

Astrophysics, Space Science, Heliophysics

anonymous

aed2ada9cd274485956f2df52fe8b03e

2025-09-18T18:25:10

BindingDB

https://www.bindingdb.org/rwd/bind/chemsearch/marvin/Download.jsp

BindingDB is a FAIRsharing knowledgebase that accelerates biomedical research and innovation by extracting, accumulating, organizing, and annotating protein-ligand interaction datasets.

3

GB

ligand-prot interactions

anonymous

a2f68ce72b914f34bc18ad6f154e0233

2025-09-18T19:37:03

BubbleML_2

https://huggingface.co/datasets/hpcforge/BubbleML_2

BubbleML_2 is a high-fidelity dataset of boiling simulations in 2D for three fluids (FC-72, Liquid N2 and R515B). It provides paired time-series fields stored in HDF5 (.hdf5) files together with metadata (.json).

2.35

TB

PDEs

Sheikh Md Shakeel Hassan

a183190b77584a0a8ce79908c5319133

2025-09-18T20:01:26

Rubin's Legacy Survey of Space and Time (LSST)

https://data.lsst.cloud/

The Rubin Observatory’s Legacy Survey of Space and Time (LSST) is a 10-year astronomical survey scheduled to begin in 2025 at Cerro Pachón in Chile, using an 8.4-meter telescope and the world’s largest 3.2-gigapixel digital camera. It will scan the southern sky every few nights in six optical bands (u, g, r, i, z, y), producing about 15 terabytes of data per night and ultimately a dataset exceeding tens of petabytes. By building both a deep static map of the universe and a time-domain “movie of the sky,” LSST will enable discoveries ranging from the nature of dark matter and dark energy to the inventory of solar system bodies, the structure of the Milky Way, and the detection of rare, fast-changing cosmic transients.

1,000

TB

Astronomy, Cosmology, Physics

anonymous

73a136bf76e74e25b5153c92922b64a4

2025-09-18T20:03:17

Rubin's Legacy Survey of Space and Time (LSST)

https://data.lsst.cloud/

The Rubin Observatory’s Legacy Survey of Space and Time (LSST) is a 10-year astronomical survey scheduled to begin in 2025 at Cerro Pachón in Chile, using an 8.4-meter telescope and the world’s largest 3.2-gigapixel digital camera. It will scan the southern sky every few nights in six optical bands (u, g, r, i, z, y), producing about 15 terabytes of data per night and ultimately a dataset exceeding tens of petabytes. By building both a deep static map of the universe and a time-domain “movie of the sky,” LSST will enable discoveries ranging from the nature of dark matter and dark energy to the inventory of solar system bodies, the structure of the Milky Way, and the detection of rare, fast-changing cosmic transients.

1,000

TB

Astronomy, Cosmology, Physics

Steven Dillmann

e4eb7e1ce61a41348d49b1d86b3558a9

2025-09-18T20:05:36

Multimodal Universe

https://huggingface.co/MultimodalUniverse

The Multimodal Universe dataset is a large scale collection of multimodal astronomical data, including images, spectra, and light curves, which aims to enable research into foundation models for astrophysics and beyond.

100

TB

Astronomy, Cosmology, Physics

Steven Dillmann

a92f734164304dd799e9914f7dded1e1

2025-09-18T20:11:53

Chandra Source Catalog (CSC)

https://cda.cfa.harvard.edu/cscweb/index.do

The Chandra Source Catalog (CSC) is a virtual X-ray astrophysics facility that enables both detailed individual source studies and statistical studies of large samples of X-ray sources detected in ACIS and HRC-I imaging observations obtained by the Chandra X-ray Observatory. The catalog provides carefully-curated, high-quality, and uniformly calibrated and analyzed tabulated positional, spatial, photometric, spectral, and temporal source properties, as well as science-ready X-ray data products.

40

GB

Astronomy, Cosmology, Physics

Steven Dillmann

08b4872696ea4451aa65363ff22902c3

2025-09-18T20:12:12

Chandra Source Catalog (CSC)

https://cda.cfa.harvard.edu/cscweb/index.do

The Chandra Source Catalog (CSC) is a virtual X-ray astrophysics facility that enables both detailed individual source studies and statistical studies of large samples of X-ray sources detected in ACIS and HRC-I imaging observations obtained by the Chandra X-ray Observatory. The catalog provides carefully-curated, high-quality, and uniformly calibrated and analyzed tabulated positional, spatial, photometric, spectral, and temporal source properties, as well as science-ready X-ray data products.

40

TB

Astronomy, Cosmology, Physics

Steven Dillmann

83c03815e5784d1fb3aceb7d3c98b25d

2025-09-19T01:33:58

Smith42/galaxies

https://huggingface.co/datasets/Smith42/galaxies

8.6 million galaxy images from the DESI Legacy Survey, crossmatched with galaxy metadata.

1.1

TB

Astronomy

Mike Smith

5d30e4f6963f4241895c2cb1abfa4c6a

2025-09-19T02:34:11

Multimodal Universe

https://github.com/MultimodalUniverse/MultimodalUniverse

The Multimodal Universe dataset is a large scale collection of multimodal astronomical data, including images, spectra, and light curves, which aims to enable research into foundation models for astrophysics and beyond.

120

TB

Astronomy

Mike Smith

2f4525f502d14595b7a16c417407fd53

2025-09-19T02:41:17

ROVER-Gen

https://huggingface.co/datasets/cheryyunl/ROVER-Gen

Logical puzzle solving and instruction following dataset for large vision models. Given an image and the task (prompt), the dataset consists of a final result (image) after executing the task on this image.

0

GB

logical puzzle solving, visual instruction following

SB

5e633ca577df48368ca9a95832f081d9

2025-09-19T02:48:20

bird-species

https://huggingface.co/datasets/Ez-Clap/bird-species

This dataset contains images of Belgian birds and their species. Dataset is split into train (6.78k rows), validation (215 rows) and test (215 rows).

163

MB

bird-species

SB

28b52169b90c495e870e898681491c13

2025-09-19T02:54:33

Wild Life Dataset

https://lila.science/datasets/wcscameratraps

This data set contains approximately 1.4M camera trap images representing around 675 species from 12 countries, making it one of the most diverse camera trap data sets available publicly. Data were provided by the Wildlife Conservation Society. The most common classes are tayassu pecari (peccary), meleagris ocellata (ocellated turkey), and bos taurus (cattle).

0

MB

wildlife

SB

67c59f7c54774d119ac61d65be1903bd

2025-09-19T06:57:31

ECOTOX

https://cfpub.epa.gov/ecotox/

ECOTOX is a comprehensive Knowledgebase providing single chemical environmental toxicity data on aquatic and terrestrial species.

0

GB

chemistry

Rahul Maity

fc394bc9e68f496382c259081b497c8e

2025-09-19T08:40:06

emg2pose

https://github.com/facebookresearch/emg2pose

A dataset of Surface electromyography (sEMG) recordings paired with ground-truth, motion-capture recordings of the hands. Data loading, baseline model training, and baseline model evaluation code are provided.

431

GB

Surface Electromyography

MatteoFasulo

c585ff9115574c759db6cde676465c75

2025-09-19T08:42:11

Generic non-invasive neuromotor interface for HCI

https://fb-ctrl-oss.s3.amazonaws.com/generic-neuromotor-interface-data

Surface electromyography (sEMG) data and training models associated with the paper "A generic non-invasive neuromotor interface for human-computer interaction". The dataset contains sEMG recordings from 100 participants in each of the three tasks described in the paper: discrete_gestures, handwriting, and wrist.

150

GB

Surface Electromyography

MatteoFasulo

3915860e8db542adaee2ae21e1275376

2025-09-19T08:56:58

clinical_trials_gov

https://clinicaltrials.gov

Clinical Trial Study information is crucial for drug development/repurposing and while the information is publicly available, it comes in really bad formats which varies for different diseases (as each has different endpoint) and different studies (Phase I/II/III). The information is available within data vendors but behind proprietary walls

0

GB

medicine

anonymous

5499c9b6ab8d4a36b86eef244e3a6970

2025-09-19T08:57:53

clinical_trials_gov

https://clinicaltrials.gov

Clinical Trial Study information is crucial for drug development/repurposing and while the information is publicly available, it comes in really bad formats which varies for different diseases (as each has different endpoint) and different studies (Phase I/II/III). The information is available within data vendors but behind proprietary walls

0

GB

medicine

anonymous

98b0da12a1c9439198fb6aefb2becd8c

2025-09-19T09:04:55

clinical_trials_gov

https://clinicaltrials.gov

Clinical Trial Study information is crucial for drug development/repurposing and while the information is publicly available, it comes in really bad formats which varies for different diseases (as each has different endpoint) and different studies (Phase I/II/III). The information is available within data vendors but behind proprietary walls

4

GB

healthcare

Piotr Kaniewski

64d8d78da2b3451781c828ac199d7ed8

2025-09-19T09:14:12

pubmed_literature

https://pubmed.ncbi.nlm.nih.gov/download/

Biomedical Literature is crucial for drug discovery and while its available at API service, its not available in a single place in a structured format.

30

GB

biomedicine

Piotr Kaniewski

71a05fedb7ac4519a05af58171e1fb61

2025-09-19T09:16:30

daily_med_labels

https://dailymed.nlm.nih.gov/dailymed/

daily med contains labels of fda approved drugs/medicines yet its not availalbe in a structured format that doesnt require manual scraping

30

GB

healthcare

Piotr Kaniewski

7c8408a366e54a4892e90627b0597678

2025-09-19T09:38:28

ontology_mapping

https://www.ebi.ac.uk/ols4/

Ontologies are essential in knowledge discovery/bioinformatics; yet mapping between them is a non trivial task

10

GB

bioinformatics

Piotr Kaniewski

c2849283ae1a41d9b6c7a20c98098588

2025-09-19T14:32:01

GeneReviews

https://huggingface.co/datasets/Tonic/GeneReviews

GeneReviews Dataset Extraction This project extracts text and metadata from GeneReviews® chapters downloaded from NCBI Bookshelf and creates a structured dataset in Hugging Face format. Overview GeneReviews® is an international point-of-care resource for clinicians, providing clinically relevant and medically actionable information for inherited conditions. This project processes the XML files from the GeneReviews database and creates a structured dataset suitable for machine learning and research applications. 📈 Dataset Statistics: Total Records: 929 GeneReviews chapters Average Abstract Length: 899.6 characters Average Content Length: 56,377.9 characters Total References: 13,683 references across all chapters Average References per Chapter: 14.7 Chapters with >100 references: 12 chapters Total Keywords: 9,616 Unique Keywords: 6,824

16

MB

Clinical reviews of genetic conditions

Joseph [open/acc] Pollack

6ae0293a58324f64a91e33d895c61e14

2025-09-19T18:51:33

GV-Rep

https://zenodo.org/records/11502840

A large-scale Genetic variants (GVs) dataset featuring a comprehensive collection of 7 million GV records with detailed annotations and one clinician verified dataset. Paper: https://arxiv.org/abs/2407.16940 , Github: https://github.com/bowang-lab/genomic-FM

30

GB

Genomics

Mohammed Hamdy

75e0431bb75241baa8f876820ff0fedc

2025-09-19T19:03:50

Biology-Instructions

https://github.com/hhnqqq/Biology-Instructions

A large-scale multi-omics biological sequences-related instruction-tuning dataset including DNA, RNA, proteins, and multi-molecules. Paper: https://arxiv.org/abs/2412.19191

100

MB

Multi-omics

Mohammed Hamdy

71fcbca937a0443a920606be6dcc4060

2025-09-19T19:05:48

International Brain Lab Data

http://internationalbrainlab.com/data; https://www.biorxiv.org/content/10.1101/2023.07.04.547681v4

Multiple Brain-Related Datasets. A key challenge is that this dataset contains data from multiple experiments and different modalities. The dataset comprises 459 Neuropixel experimental sessions encompassing 699 distinct recordings (probe insertions). These recordings were obtained from 139 subjects performing the IBL task across 12 different laboratories. Spike-sorting analysis yielded 621,733 units, of which 75,708 were classified as good quality. In total, 241 brain regions were recorded(from their website).

50

TB

Multisite Neural Recording Local Field Potential(LFP)

Ken Fahmidur Rahman

a77b360636984b4ea3391c1757cdca50

2025-09-19T19:34:53

Insect-1M

https://uark-cviu.github.io/projects/insect-foundation/

Large-scale Insect Dataset comprising more than a million images with hierarchical labels from the high to the low taxonomy level, including class, order, family, genus, and species. Paper: https://arxiv.org/abs/2502.09906

100

GB

Entomology

Mohammed Hamdy

92b0add1ee9b446fb234f13438ea93e0

2025-09-20T08:04:57

ToT-Biology

https://huggingface.co/datasets/moremilk/ToT-Biology

The ToT-Biology dataset emphasizes mechanistic understanding and explanatory biological reasoning, rather than just providing correct answers. It aims to train AI models in interpretability and logical deduction within the biological realm. Spanning a wide range of biological complexities, it starts with foundational concepts in cell biology, genetics, and ecology, and progresses to advanced areas like systems biology, synthetic biology, and computational biophysics. The dataset incorporates real-world applications across diverse fields such as medicine, agriculture, environmental science, and bioengineering, ensuring practical relevance alongside rigorous theoretical exploration.

80

MB

Biology

SHIVAM DUBEY

Datasets:

hugging-science
/

dataset-quest-index

Space using hugging-science/dataset-quest-index 1