Source code for uchrom.strc.tad.utils

# TAD utilities: BED file I/O and substructure operations

import numpy as np
import pandas as pd




def load_tad_from_bed(bed_path):
    """Load TAD regions from BED file."""
    regions = []
    with open(bed_path, 'r') as f:
        for line in f:
            line = line.strip()
            if not line or line.startswith('#'):
                continue
            parts = line.split('\t')
            if len(parts) >= 3:
                chrom = parts[0]
                start = int(parts[1])
                end = int(parts[2])
                regions.append((chrom, start, end))
    return regions





def tad_regions_to_bin_indices(regions, bins_df, chrom=None):
    """Convert TAD genomic regions to bin indices."""
    bin_indices = []
    for rgn_chrom, rgn_start, rgn_end in regions:
        if chrom is not None and rgn_chrom != chrom:
            continue
        mask = (
            (bins_df['chrom'] == rgn_chrom) &
            (bins_df['start'] >= rgn_start) &
            (bins_df['end'] <= rgn_end)
        )
        matching_bins = bins_df[mask]
        if len(matching_bins) > 0:
            start_idx = matching_bins.index[0]
            end_idx = matching_bins.index[-1]
            bin_indices.append((start_idx, end_idx))
    return bin_indices





def substructures_from_tads(tad_indices, structure_coords):
    """Extract substructures based on TAD boundaries."""
    substructures = []
    for start_idx, end_idx in tad_indices:
        subcoords = structure_coords[start_idx:end_idx+1].copy()
        substructures.append(subcoords)
    return substructures





def merge_substructures(substructures, tad_indices, total_bins):
    """Merge TAD substructures back into full structure."""
    full_coords = np.zeros((total_bins, 3))
    for (start_idx, end_idx), subcoords in zip(tad_indices, substructures):
        full_coords[start_idx:end_idx+1] = subcoords
    return full_coords





def save_tads_to_bed(tad_indices, bins_df, output_path):
    """Save detected TAD regions to BED file."""
    with open(output_path, 'w') as f:
        for start_idx, end_idx in tad_indices:
            chrom = bins_df.iloc[start_idx]['chrom']
            start = bins_df.iloc[start_idx]['start']
            end = bins_df.iloc[end_idx]['end']
            f.write(f"{chrom}\t{start}\t{end}\n")
    print(f"TADs saved to {output_path}")