January 15, 2019, Tuesday

• CyVerse URLs is slow for loading MaizeCode data
  • Test was done on wormbase.org
  • Use the C. elegans as genome
  • https://s3.amazonaws.com/wormbase-modencode/wiggle_binaries/TSS_Kruesi.Reverse.bw
  • Amazon track: Misc, Transcription Start Sites, TSS (Kruesi) Reverse
• MaizeCode test data
  • https://data.sciapps.org/results/test/sig_pollen_rep1_R1.bw
  • https://de.cyverse.org/anon-files/iplant/home/maizecode/sci_data/results/MCrna-0.0.1_27dc8dba-9356-4091-a549-bf18d4e3a76a/sig_pollen_rep1_R1.bw
• Strategy
  • New CyVerse link is comparable to either AWS or local files
    • replace de.cyverse.org/anon-files with data.cyverse.org/dav-anon
  • Can we make the results folder on Halcott and put BW files in there when the workflow is done?
    • Or do it when the user is 'visualizing'? How do we know which genome it goes?
  • Currently the folder is owned by apache user, need to sort out why
  • Or we can do the transfer of BW files in a batch?
  • We are not going to host users BW files, we are just going to host MaizeCODE BW files
• About JBrowse
  • Can host multiple genomes without using Apollo
  • Need to change the config file
  • https://data.sciapps.org/view2/?data=
  • /var/www/html/jbrowse/JBrowse-1.12.1/
  • https://data.sciapps.org/view2/?data=data2/B73/B73_AGPv4_chrs_2016_11_01/apollo_data/
• For Alignment
  • We will need to ignore contigs and focus on pseudomolecular only
  • This means we will change the genome index files and realign all results
    • Or we will change the index but just filter the alignment file