datasets
Training and evaluation data, with the modality, task and licence stated up front. Listed live from the Hugging Face Hub.
GDPa1
GDPa1: Antibody developability dataset
Contains the assay data for 242 antibodies across 10 assays as described in our latest preprint, PROPHET-Ab: A high-throughput platform for biophysical antibody developability assessment to enable AI/ML model training.
Example usage
Using pandas:
import pandas as pd
# Login using e.g. `huggingface-cli login` to access this dataset
df = pd.read_csv("hf://datasets/ginkgo-datapoints/GDPa1/GDPa1_v1.2_20250814.csv")
Using Hugging… See the full description on the dataset page: https://huggingface.co/datasets/ginkgo-datapoints/GDPa1.vn-provinces-gini
Vietnam Gini coefficient
Vietnam Gini coefficient. Geographic labels are English (UN/GSO style ASCII romanization). Tables cover provinces, regions and national total where present. Province names follow ar_core.vn_geo (historical 63-province system).
Figures
Hero
Comparison
Color key
Files
provinces (441 rows)
data/provinces.csv
data/provinces.dta
data/provinces.xlsx
regions (42 rows)
data/regions.csv
data/regions.dta
data/regions.xlsx… See the full description on the dataset page: https://huggingface.co/datasets/letrinhan/vn-provinces-gini.GDPx2
GDPx2
GDPx2 Functional Genomics Dataset: DRUG-seq + Chemical Perturbation in 4 Primary Cell Types. This dataset contains the supplementary data for the bioRxiv preprint "Mapping the Transcriptional Landscape of Drug Responses in Primary Human Cells Using High-Throughput DRUG-seq".
Experimental Design
Experiments were performed in a 384-well format assay. Four primary human cell lines were used: aortic smooth muscle cells, skeletal myoblasts, dermal fibroblasts… See the full description on the dataset page: https://huggingface.co/datasets/ginkgo-datapoints/GDPx2.GDPx1
GDPx1
GDPx1 Functional Genomics Dataset: DRUG-seq + Chemical Perturbation in A549 Cells
Experimental Design
The Library of Pharmacologically Active Compounds (LOPAC 1280), containing 1,264 compounds, was tested at two concentrations (100 nM and 1000 nM) using four technical replicates in A549 cells (a human lung carcinoma epithelial cell line). DMSO was used as an inert control, and Digitonin as a transcriptional control. This generated 28 384-well plates that were… See the full description on the dataset page: https://huggingface.co/datasets/ginkgo-datapoints/GDPx1.GDPx3
GDPx3
GDPx3 is a high-content imaging dataset (Cell Painting) generated by screening 40 compounds across four concentrations at two time points, plus controls. This data was collected in A549, dermal fibroblasts, aortic smooth muscle cells, and aortic endothelial cells. GDPx3 offers a wealth of phenotypic information ready for your AI models.
Experimental Design
Experiments were performed in a 384-well format assay. Four cell lines – A549 (a human lung carcinoma… See the full description on the dataset page: https://huggingface.co/datasets/ginkgo-datapoints/GDPx3.darwin-a2ap-analysis
Darwin A2AP Model Analysis Dataset
Last Updated: 2025-11-02T06:09:38.572521
Summary
Total models analyzed: 1000
A2AP compatible models (score ≥ 60): 737
Average A2AP score: 68.7
Top 10 A2AP Compatible Models
model_id a2ap_compatibility_score merge_difficulty
Qwen/Qwen3-30B-A3B-Instruct-2507 96.0 Easy
Qwen/Qwen3-30B-A3B-Instruct-2507 96.0 Easy… See the full description on the dataset page: https://huggingface.co/datasets/ginipick/darwin-a2ap-analysis.darwin-a2ap-analysis-20250915_163155
Darwin A2AP Model Analysis Results
Generated on: 2025-09-15T16:31:55.831490
Summary
Total models analyzed: 20
A2AP compatible models (score ≥ 60): 12
Top 10 A2AP Compatible Models
model_id a2ap_compatibility_score merge_difficulty
baidu/ERNIE-4.5-21B-A3B-Thinking 75.0 Medium
Qwen/Qwen3-Next-80B-A3B-Instruct 75.0 Medium
Qwen/Qwen3-Next-80B-A3B-Thinking… See the full description on the dataset page: https://huggingface.co/datasets/ginipick/darwin-a2ap-analysis-20250915_163155.dsc204-datadarwin-a2ap-analysis-20250915_151507
Darwin A2AP Model Analysis Results
Generated on: 2025-09-15T15:15:08.105743
Summary
Total models analyzed: 100
A2AP compatible models (score ≥ 60): 46
Analysis version: 1.4-fixed-trending
Top 10 A2AP Compatible Models
model_id a2ap_compatibility_score merge_difficulty
TinyLlama/TinyLlama-1.1B-Chat-v1.0 80.0 Medium
baidu/ERNIE-4.5-21B-A3B-Thinking 75.0 Medium… See the full description on the dataset page: https://huggingface.co/datasets/ginipick/darwin-a2ap-analysis-20250915_151507.