Ignatieva, A., Favero, M., Koskela, J., Sant, J., Myers, S.R. The length of haplotype blocks and signals of structural variation in reconstructed genealogies. bioRxiv, doi.org/10.1101/2023.07.11.548567

DoLoReS: Detection of Localised Recombination Suppression

To run the script, first install the necessary packages

pip install -r requirements.txt

Then run

python -m run-dolores -h

to see the full list of options.

The example code can be run using the provided ARG (10Mb of human chr21 simulated using stdpopsim and reconstructed using Relate):

python -m run-dolores -C chr21 -n chr21_100_relate -t example

The input name should correspond to the location of the ARG in name.trees (or name.trees.tsz) in tskit format, name.poplabels (if using genotype IDs for clade calculations), and name.popsize (a comma-separated file with two columns, column one giving the start time of each epoch in generations, and column two the corresponding haploid population size).

The script produces plots of the results (Q-Q plot of the (one-sided) Test 1 p-values, plot of Test 1 p-value outliers, and a Manhattan plot for Test 1 and Test 2) and an output file (.csv) with the following entries:

• name: Input name of trees to process
• genetic_map: Recombination map used
• total_clades: Total number of clades in the ARG (of size > 2)
• clade_num: Index of clade
• clade_id: ID of clade
• nlog10p_test1: p-value for Test 1
• nlog10p_test2: p-value for Test 2
• cladesize: Size of clade
• span: Genomic span of clade (bp)
• start: Start position of clade
• end: End position of clade
• mut_span: Genomic span of clade measured from leftmost to rightmost supporting mutation (bp)
• left_mut: Position of leftmost supporting mutation
• right_mut: Position of rightmost supporting mutation
• num_mutations: Number of supporting mutations
• merged: How many clades have been merged together to produce this one
• chunk_index: Chunk index where clade begins
• tree_index: Tree index where clade begins
• node_id: Node ID of clade MRCA in tree where clade begins

Genomic span of edges

This script computes p-values for (a subset of) edges in the input ARG. Run

python -m run-edgespans -h

to see the full list of options.

The example code can be run using the provided ARG (10Mb of human chr21 simulated using stdpopsim with the SMC' model):

python -m run-edgespans -C chr21 -n chr21_100_smcprime -t example -T smcprime -b 1000

The script produces a Q-Q plot and histogram of the (two-sided) p-values.