The repository contains benchmarking scripts for testing String Decomposition Problem solutions described in the manuscript "The String Decomposition Problem and its Applications to Centromere Assembly".
AC uses chop_to_monomer.py script from the original alpha-CENTAURI pipeline to generate monomer positions in given long reads, after that it applies convert_identites.py script to choose best monomer for each place.
SD uses dynamic programming to generate best decomposition of each read into the given monomers.
All scripts, except AC_HMMer.snake and convert_identites.py, use Python3. Required libraries are listed in requirement.txt
ls scripts
AC_HMMer.snake snakemake file, runs AC pipeline (uses Python2)
convert_identities.py identifies best monomers for the decomposition (part of AC pipeline)
compare_tools_monomerfree.py generates statistics for Appendix "Monomer-free benchmarking"
generate_stats.py generates monoread-to-monocentromere alignment and calculates overall statistics (Table 1)
process_mismatches.py generates mismatches statistics based on generate_stats.py output
process_abnormal.py generates statistics for abnormal HORs based on generate_stats.py output
draw_identities.py MinAlignmentIdentity statistics
hor_analysis/ script for HOR analysis with HiFi reads from T2T project
parameters_tuning/ scripts for building logistic regression
All data used in this work can be found on T2T github page or on Figshare.
The String Decomposition Problem and its Applications to Centromere Assembly. Tatiana Dvorkina, Andrey V. Bzikadze, Pavel A. Pevzner bioRxiv 2019.12.26.888685; doi: https://doi.org/10.1101/2019.12.26.888685
In case of any issues please use issue tracker or email directly to [email protected]