Polish a nanopore assembly using Illumina reads
- Take as input the nanopore assembly, and fastq(s) of illumina reads for the assembly.
- Map the illumina reads to the draft assembly.
The quick way - run poligraph globally:
perl poligraph.pl --draft_assembly path/to/draft_assembly.fa --reads_fq path/to/reads.fq --base_dir dir_to_work_from --global --cortex_dir path/to/cortex --vcftools_dir path/to/vcftools --stampy_bin path/to/stampy.py
The slower (and better?) way - run in the 10kb windows - right now this is broken - open issue to fix!
perl poligraph.pl --draft_assembly path/to/draft_assembly.fa --reads_fq path/to/reads.fq --base_dir dir_to_work_from --cortex_dir path/to/cortex --vcftools_dir path/to/vcftools --stampy_bin path/to/stampy.py
Options:
--draft_assembly path/to/draft_assembly.fa, the assembly to be polished.
--reads_fq path/to/reads.fq, reads to use for polishing. If paired end fastqs, enclose in "".
--base_dir dir_to_work_from, default pwd, this will be treated as the head directory and the poligraph_corrected.fa will be output here. Creates the dir if it doesn't exist.
--window_size int, default 10,000
--num_procs int, default 20, number of windows to process in parallel.
--label string, adds a labelled level to directory structure (e.g. so can run with different cleaning methods in the dir and give them a different name).
--cortex_dir path/to/cortex.
--vcftools_dir path/to/vcftools.
--stampy_bin path/to/stampy.py.
--k int, kmer size for de bruijin graph construction, default 31.
--bc yes --pd no, exactly one of bc and pd must be yes. This is the method used by cortex to call variants.
--qthresh int, default 10.```