AMBITIOUS BIO

Corpus: a whole-body dataset for biological AI

Corpus by Ambitious Bio is a donor-matched atlas of human tissue built for teams training biological AI and doing drug discovery. Ambitious Bio, based in Cambridge, Massachusetts, collects tissue prospectively from consented postmortem donors under consistent protocols, so every sample is comparable across tissues and donors.

Each donor contributes about 30 tissue sites today, with a design target of more than 50 per body. Every sample gets deep mass-spectrometry proteomics (Orbitrap Astral), detecting roughly 10,000–12,000 proteins per tissue, with multi-omics planned for every sample. Because all tissues come from the same donor, models can learn how molecules vary across organs within one body, not just across populations.

In its first cohort, Corpus expanded the Human Protein Atlas's protein parts list by a median of 56% per shared tissue, and in every tissue compared it detected at least one potential drug target the Human Protein Atlas had missed. Collection is continuous, with plans for about 2,500 donors in 2027 and 10,000 in 2028, and diseased donors added from 2027.

Corpus is licensed commercially for AI training, target discovery, safety and biodistribution work. A free synthetic demo dataset is on Hugging Face at https://huggingface.co/datasets/ambitious-bio/corpus-synthetic-demo. Ambitious Bio was founded by CEO Elizabeth Hudson. Licensing and access: hello@ambitious.bio.

Corpus + Human Protein Atlas

More of the body’s protein parts list

  • Human Protein Atlas alone
  • Human Protein Atlas + Corpus
Detected Ensembl gene IDs per tissue
TissueHuman Protein Atlas aloneHuman Protein Atlas + Corpus
adipose tissue4,85013,450
adrenal gland9,25014,350
appendix9,65014,500
bone marrow7,95013,650
breast8,85013,850
bronchus9,20014,300
cervix8,55013,450
colon10,35014,200
duodenum9,80014,550
endometrium9,25013,850
epididymis9,10014,350
esophagus8,30014,300
fallopian tube9,20014,000
gallbladder9,55014,050
heart7,95014,200
kidney10,25014,750
liver7,85013,700
lung9,20014,550
lymph node7,90014,050
nasopharynx8,95013,800
oral mucosa7,70014,200
ovary6,60013,450
pancreas9,25014,400
parathyroid gland7,30013,950
prostate8,20013,200
rectum9,85014,100
salivary gland8,55014,300
seminal vesicle8,45013,200
skeletal muscle7,35013,500
skin9,65014,650
small intestine9,80014,200
smooth muscle6,75013,400
spleen7,10013,850
stomach9,90015,050
testis11,05014,900
thyroid8,95014,100
tonsil9,60014,100
urinary bladder9,05014,050
vagina7,40012,600

Adding Corpus expands the inventory in all 39 shared tissues.

Matched gene-level comparison. One Ensembl gene ID counted once per tissue.

Download the table (CSV)

Press

News and announcements from Ambitious.