Quick Start

Launch a discovery program with Om — from a query

Browse curated SDK flows for unlocking datasets, running diligence jobs, or streaming data. Copy the code or launch the associated Colab to try it live with your API key.

Create an accountView API docs

Quick Start Flows

Select a flow to see the necessary endpoints, highlight notes, and launch a Colab notebook in one click.

Data access

Unlock & stream a protein dataset

Use the SDK to unlock a protein, inspect stats, and stream binders or points as CSV—perfect for quick model training experiments.

  • Unlock once per protein via `access_unlock` before streaming paid datasets.
  • Call stats endpoints to size pulls before spending credits.
  • Streaming helpers expose settlement headers (row counts, credits charged).
from omtx import OMTXClient

client = OMTXClient()

client.access_unlock(
    protein_uuid="550e8400-e29b-41d4-a716-446655440000",
    gene_name="BRCA1",
)

stats = client.data_access_points_stats(
    dataset="public",
    protein_uuid="550e8400-e29b-41d4-a716-446655440000",
)
print("Rows available:", stats.get("total_rows"))

stream = client.data_access_points_stream(
    dataset="public",
    protein_uuid="550e8400-e29b-41d4-a716-446655440000",
    limit=5_000,
    fmt="csv",
)

try:
    with open("brca1_public.csv", "wb") as fh:
        for chunk in stream.iter_bytes():
            fh.write(chunk)
finally:
    stream.close()

print("Charged rows:", stream.headers.get("X-Row-Count"))
View API docsRun in Colab