Datasets

Open Source Data for Drug Discovery

Download, analyze, and build upon our growing library of free biological datasets to advance your AI models and traditional drug discovery efforts.

Explore Datasets

Datasets

We release curated datasets from our automated high-throughput platforms to accelerate drug discovery across multiple therapeutic modalities.

Functional Genomics

Chemical and genetic perturbation responses with multi-omic readouts
High-throughput transcriptomics using DRUG-seq (~10,000+ genes per sample)
High-content imaging with cellular morphology measurements
Data from primary cells, iPSC-derived cells, and standard cell lines

Antibody Developability

Comprehensive biophysical characterization across multiple assays
Hydrophobicity, aggregation, stability, and self-association measurements
Benchmark antibodies with diverse properties

Datasets

Functional Genomics

Details	GDPx3	GDPx2	GDPx1
DOWNLOAD NOW	*Download GDPx3*	*Download GDPx2*	*Download GDPx1*
RELEASE DATE	April 2025	December 2024	September 2024
CELL TYPES	A549 cells (human non-small cell carcinoma epithelial cell line), aortic smooth muscle cells, dermal fibroblast, aortic endothelial cells	Human melanocytes cells, aortic smooth muscle cells, dermal fibroblast, and skeletal muscle myoblasts	A549 cells (human non-small cell carcinoma epithelial cell line)
PERTURBATIONS	48 compounds, 4 concentrations, 2 time points, 4 replicates	85 compounds, 6 concentrations, 4 replicates	1,264 compounds, 2 concentrations, 4 replicates
READOUT	High Content Imaging (Cell Painting) with 2200 x 2200 image dimensions	Transcriptomic (DRUG-seq) with 2M reads sequencing depth	Transcriptomic (DRUG-seq) with 2M reads sequencing depth
PLATE DENSITY	384 well plate	384 well plate	384 well plate
AVAILABLE DATASET SIZE	220 GB	~200 GB / cell type	60 GB

Datasets

Antibody Developability

DETAILS	GDPa1
DOWNLOAD NOW	*Download GDPa1*
PREPRINT	Read the preprint with method details, results summary, and discussion
DATA PACKAGE	246 IgGs were characterized by the biophysical assays listed below. Data readouts are reported for each antibody that was tested for each assay and passed quality control criteria for that assay. Data is available in a tidy data format
ANTIBODY PRODUCTION	Antibodies were expressed in HEK293F and purified using Protein A chromatography prior to developability assessment for all assays. Antibodies tested on DLS-kD went through an additional polishing SEC step. A smaller subset of antibodies (20 IgGs) was produced in ExpiCHO and purified using Protein A chromatography.
DEVELOPABILITY ASSAYS	Titer by Valita Purity by rCE-SDS Aggregation by SEC Thermostability by nanoDSF and DSF Colloidal stability by SMAC Hydrophobicity by HIC Heparin binding by HAC Self association by AC-SINS Self association by DLS-kD Polyreactivity by bead-based method against CHO SMP and ovalbumin
DATASHEETS IN DATA FILE	Definitions of column headers in other datasheets Antibody sequences Assay data in “tidy data” format with one row per replicate Assay data summary statistics with average, standard deviation, and replicates for each assay Data for nanodsf vs dsf with the same ramp rate in “tidy data” format Prior literature data summarizing prior published results compared with GDPa1 data in the associated preprint

Go deeper on our data

Releasing the GDPx3 Cell Painting Dataset

PROPHET-Ab: A Data Platform for Antibody Developability

Releasing the GDPx2 Functional Genomics Dataset

Tell us what data you need. We’ll generate a large antibody developability dataset to power your AI model.

Need data? Let's talk.

Schedule a call