Formatting data for ARGHelpR

Written by: Keaka Farleigh, Ph.D.
Date: September, 16th, 2026.
Date last modified: September, 17th, 2026

Purpose

To help you understand the input format for ARGHelpR and how to convert the output from ancestral recombination graph (ARG) estimation programs into this format.

Format expected by ARGHelpR

ARGHelpR uses bed files for analysis. There are four columns that are tab-delimited. The first column contains the chromosome, the second contains the start position, the third contains the end position, and the fourth column contains the ARG for that region.

The is natively output by ARGWeaver (Rasmussen et al., 2014), using the smc2bed function, but we have to convert output from other programs to bed format. We provide a couple scripts below to help. Note that these are python scripts.

The smc2bed will generate a bed file for each ARG analysis, as will the scripts below. You can use the command cat *.bed > my_args.bed to combine all of them if you ran them seperately for different regions of the genome.

Converting output from ARG-Needle

This is a script to convert output from ARG-Needle (Zhang et al., 2023).

import arg_needle_lib
import tskit

def export_arg_to_tree_bed(argn_path, chromosome, output_bed):
    # Load the inferred ARG from the .argn file
    arg_obj = arg_needle_lib.load_arg(argn_path)
    
    # Convert the ARG object to a tskit TreeSequence object
    # This automatically splits the genome into non-recombining interval segments
    ts = arg_needle_lib.arg_to_tskit(arg_obj)
    
    # Write intervals in bed format
    with open(output_bed, 'w') as bed_file:
        # Iterate through each local tree segment along the sequence length
        for tree in ts.trees():
            # Get start and end positions for the interval
            start = int(tree.interval.left)
            end = int(tree.interval.right)
            
            # Generate the Newick string phylogeny
            newick_str = tree.newick()
            
            # Write out bed file
            bed_file.write(f"{chromosome}\t{start}\t{end}\t{newick_str}\n")

# Example Usage:
export_arg_to_tree_bed("myarg.argn", "chr1", "myarg_chr1_trees.bed")

Then you can combine all of the ARGs for each chromosome for use in ARGHelpR (see cat command above).

Converting output from tsinfer

This is a script to convert output from tsinfer (Kelleher et al., 2019).

import tskit

def ts_to_bed(tree_sequence_path, chromosome, output_bed):
  
    # Load the ARG
    ts = tskit.load(tree_sequence_path)
    
    with open(output_bed, "w") as bed_file:
        # Iterate through every local tree along the genome
        for tree in ts.trees():
            # Get start and end positions for the interval
            start = int(tree.interval.left)
            end = int(tree.interval.right)
            
            # Generate the Newick string phylogeny
            newick_tree = tree.newick()
            
            # Write out bed file
            bed_file.write(f"{chromosome}\t{start}\t{end}\t{newick_tree}\n")

# Example Usage:
ts_to_bed("myarg.trees", "chr1", "myarg_chr1_trees.bed")

Then you can combine all of the ARGs for each chromosome for use in ARGHelpR (see cat command above).

Please reach out to Keaka Farleigh if you have any questions.

Literature Cited

Kelleher, J., Wong, Y., Wohns, A. W., Fadil, C., Albers, P. K., & McVean, G. (2019). Inferring whole-genome histories in large population datasets. Nature genetics, 51(9), 1330-1338.

Rasmussen, M. D., Hubisz, M. J., Gronau, I., & Siepel, A. (2014). Genome-wide inference of ancestral recombination graphs. PLoS genetics, 10(5), e1004342.

Zhang, B. C., Biddanda, A., Gunnarsson, Á. F., Cooper, F., & Palamara, P. F. (2023). Biobank-scale inference of ancestral recombination graphs enables genealogical analysis of complex traits. Nature Genetics, 55(5), 768-776.