Getting started

bindcurve supports Python 3.10 through 3.14. Install it into your project environment with either uv or pip:

uv

uv add bindcurve

pip

python -m pip install bindcurve

Fit an inhibition curve

bindcurve accepts long-form observations. The required columns are compound_id, concentration, and response:

import bindcurve as bc
import pandas as pd

observations = pd.DataFrame(
    {
        "compound_id": ["example"] * 7,
        "concentration": [0.01, 0.1, 0.5, 1.0, 2.0, 10.0, 100.0],
        "response": [99.0, 90.9, 66.7, 50.0, 33.3, 9.1, 1.0],
    }
)

data = bc.DoseResponseData.from_dataframe(observations)
results = bc.fit(
    data,
    model="ic50",
    fixed={"ymin": 0.0, "ymax": 100.0},
)

print(results.summary()[["compound_id", "IC50"]])

  compound_id  IC50
0     example   1.0

Concentrations must use one consistent unit. The fitted IC₅₀ is reported in that same unit. See the logistic-model theory for the model definition and interpretation.