Python scripts for running GraphChainer experiments

• Install miniconda
• conda env create -f environment.yml
• conda activate GraphChainerScripts
• Compile and place the aligner's binaries on the bin/ folder (they are assumed to be called GraphAligner, GraphChainer, minigraph, minichain respectively)
• Place Badread's repo in this directory git clone https://github.com/rrwick/Badread.git

Instructions

• Every python script has its own instructions/helper
• The scripts should be run in the order [generate_sim_reads.py], run_experiment.py, compute_summary.py, compute_metrics.py (if real reads are used the first script is skipped)
• generate_sim_reads.py : Takes a graph as input and generates simulated reads from a random path of the graph using GraphChainer random path generator and Bardread simulator.
• run_experiment.py : Runs the four tools on the specified graph and fastq files.
• compute_summary.py : Computes csv tables with the metrics used in the paper, it receives as input the graph and the fastq file, or in the case of simulated reads, the files output by generate_sim_reads.py.
• compute_metrics.py : Plots a graph with the summary (csv) files output by compute_sumary.py