Visit the GitHub repository for the full framework code:
https://github.com/silicobio/hoike1## 1. Look up the condition samples for a tissue that exists in the normal reference set.
2user_target_tissue = "Skin"
3condition_subset = dataset.condition_df[dataset.condition_df["tissue_type"] == user_target_tissue].reset_index(drop=True)
4normal_baseline_array = dataset.normal_profiles[user_target_tissue]
5
6## 2. Generate with sampling-time normalization consistent with diffusion training.
7generated_df = generate_synthetic_condition_data_consistent(
8 normal_profile=normal_baseline_array,
9 jepa=jepa_model,
10 diffusion=diff_model,
11 scheduler=scheduler,
12 gene_cols=dataset.gene_cols,
13 num_samples=2500,
14 value_cap=condition_value_cap,
15 sampling_noise_scale=1.1,
16)
