TODO LIST

[Sebastien-Raguideau]

Before release we want :

• Documentation:
  • explain config file
  • regenerate workflow figures
    • rule name consistency, upper case or not
    • informative rule name
    • generate desman workflow fig
  • document workflow figures : add some infos on key rules, input/output/what it does
• Installation:
  • deal with COG database: rule to download it, or specify where to get it
  • gtdb install....
  • deal with desman
  • learn how to create a new releases for the new panda version
  • create a new conda install, with right version of pandas, referring to new desman release
  • create a conda based install (Seb)
  • either a conda repos, where everything is taken care of (submodule pull, compilation of spades tk)
  • or just an env and add some manual installation steps
  • document installations
  • document a conda env install and additional steps needed (compilation, etc)
  • document an install from source of all dependencies.
• Figures/Results:
  • colored .gfa concatenated over SCG (try to generate corresponding fig from command line) (Seb)
  • gtdb output (Seb)
  • barplot of coverage for each strain of a mag (Kim)
  • for each mag : a tree of strains (Seb)
  • number of strain(y) versus coverage of bins (x) (Seb)
  • summary : nb strains, coverage each strain, total mag coverage, divergence,nb mags , nb bins, quality of assembly. (Seb)
• Cosmetic changes:
  • make it so we don't call a script named start.py (Seb) DONE
  • make it so that we can call start.py from anywhere, not just the STRONG repos (Seb) DONE
  • change folders so that users can either get all the data they need, or all the results they needs : figs, nb strains/strains freq, trees ... etc
  • change bayespath output, these folder are a mess, we want informative files names, gamma is not, we want a meaningfull summary. We want less file in the same folder. A bit overwhelming.
• misc
  • change scg_data : get the mean AA length for all cogs. DONE

latest version /mnt/gpfs/chris/Projects/STRONG_Runs/Benchmarking/ResultsI/Synth_G45_S10

[chrisquince]

Another enhancement allow alternative definition of MAG rather than 75% single copy core genes which is our default.