Skip to content

Create reference intervals

Legacy Documentation - PHG Version 1

This section contains documentation for PHG version 1, which is no longer actively developed. It is preserved here for archival and historical reference only. If you are looking to use the Practical Haplotype Graph, please refer to the PHG v2 documentation, which reflects the current version of the software.

SCRIPT PURPOSE

Generate reference intervals for the Practical Haplotype Graph (PHG)

NOTES

  • This script assumes it is run inside the PHG Docker container with predefined I/O paths
  • The .gff file is assumed to be in JGI format: gene models have the "gene" name, and an "ID=..." field is present in annotation

RUNNING THE SCRIPT

#!bash

docker run --rm                                                                    \
-v /your_data_folder/:/tempFileDir/data                                            \
maizegenetics/phg                                                                  \
/CreateReferenceIntervals.sh -f your_reference.fasta -a your_reference.gene.gff3 [ ... optional parameters]

REQUIRED PARAMETERS

#!bash

   -f <file name>  
      name of fasta file containing the reference sequence  
   -a <file name>  
      name of genome annotation file in .gff format containing gene model annotation

OPTIONAL PARAMETERS

#!bash

  -k <integer>  
     Length of kmer used for determining repetitive regions  
     Default: 11
  -e <integer>  
     Number of bases by which to expand gene models for initial reference interval selection  
     Default: 1000
  -m <integer>  
     Distance (in bp) between genes below which gene models are merged  
     Default: 100
  -p <double>  
     Proportion of kmers to be considered repetitive.  
     This determines the high kmer count tail which is considered repetitive (e.g. the top 0.05 most frequent)  
     Default: 0.1
  -n <integer>  
     Number of kmer copies (genome-wide) above which a kmer is considered repetitive. Overrides -p  
     Default: none, -p is used by default
  -l <integer>  
     The number of bases to consider when evaluating if a location in the genome is repetitive  
     Default: 100
  -s <integer>  
     The step size (in bp) by which to proceed outward from a gene model when evaluating flanking regions  
     Default: 10

SCRIPT RESULTS

Output location (subfolder in the input data folder)

  • genomic_intervals_unique-timestamp

Relevant output contents

  • reference_intervals_run.log -- a log file summarizing parameters for the run
  • your_fasta.gene.expand.trimmed.summary_report.tsv -- a summary of seed gene model expansion
  • your_fasta.kmer_count.tsv -- a complete list of kmer counts for kmers with count > 1
  • your_fasta.gene.expand.trimmed.bed -- the final reference intervals, in BED format