Skip to content

Latest commit

 

History

History
59 lines (46 loc) · 2.88 KB

README_refr-genome-summary.md

File metadata and controls

59 lines (46 loc) · 2.88 KB
Raw

Summary of model organism genomes

We selected a diverse set of 10 model organisms to demonstrate the utility of iLoci for providing a descriptive overview of genome content and organization.

Data retrieval and processing

The fidibus script was used to perform the following tasks:

  • retrieve genome assemblies and annotations from NCBI RefSeq (the download task)
  • pre-process the data (the prep task)
  • compute iLoci (the iloci task)
  • collate sequences and annotations for various feature types (the breakdown task)
  • compile tables of summary statistics over the data (the stats task)

You can copy/paste the relevant command from below or alternatively execute the run-fidibus-stats.sh and run-fidibus-cleanup.sh workflow bash scripts in the wfscripts directory. Note that the scripts include also the work on the special interest genomes described in the manuscript, documentation of which is not repeated here.

fidibus --workdir=data \
        --numprocs=10 \
        --refr=Scer,Cele,Crei,Mtru,Agam,Dmel,Xtro,Drer,Mmus,Hsap \
        download prep iloci breakdown stats

This command could take up to a couple of hours to run: go enjoy lunch or read a paper while you wait! You will want to adjust the --numprocs parameter based on the number of available processors on your computer, as processing multiple genomes in parallel will complete the task more quickly. However, the mammalian genomes each take on the order of 30 minutes, so that is about the best runtime that can be expected.

Output data

The fidibus command creates a directory called data containing subdirectories, each dedicated to a specific genome. These subdirectories store intermediate and ancillary data files. Of particular interest are the following:

  • Xxxx.iloci.gff3: full annotation file, with iLoci specified and associated with their constituent gene features
  • Xxxx.miloci.gff3: annotation file containing just iLocus features, with overlapping siLoci and niLoci merged into miLoci
  • Xxxx.yyyy.tsv: statistics calculated on various data types (iLoci, miLoci, pre-mRNAs, mature mRNAs, exons, introns, coding sequences) and compiled into a tabular format for easy loading into R, Python, or similar data analysis environments

iLocus summary

Summaries of iLocus composition for these genomes, corresponding to Tables 1-3 in the manuscript, were computed with the following commands (script wfscripts/make-Tables1-3.sh).

fidibus-ilocus-summary.py  --workdir=data --outfmt=tsv Scer Cele Crei Mtru Agam Dmel Xtro Drer Mmus Hsap
fidibus-pilocus-summary.py --workdir=data --outfmt=tsv Scer Cele Crei Mtru Agam Dmel Xtro Drer Mmus Hsap
fidibus-milocus-summary.py --workdir=data --outfmt=tsv Scer Cele Crei Mtru Agam Dmel Xtro Drer Mmus Hsap