# Constructed data: 1,000 proteins; 100 changed 2-fold (log2 fold change ±1), 900 unchanged; missing values more likely at
# low intensity. Python 3.13, NumPy, pandas. Seed 1, the first one tried.
import numpy as np, pandas as pd
from model import study, N
rng = np.random.default_rng(1)
x, eff = study(rng)
d = pd.DataFrame(x.T.round(3), columns=[f"control_{i + 1}" for i in range(N)] + [f"treated_{i + 1}" for i in range(N)])
d.insert(0, "protein", [f"P{i + 1:04d}" for i in range(x.shape[1])])
d.to_csv("intensities.csv", index=False)
pd.DataFrame({"protein": d.protein, "true_log2fc": eff}).to_csv("truth.csv", index=False)