Moore Notes 2 9 15

Group Call

• Participants: Katie, Jonathan, Tom, Stephen, Guillaume, Dongying, Ladan, Stacia, Patrick, Josh

• Upcoming meetings
  • JGI users meeting
  • ASM in New Orleans
  • AAAS this week (Katie speaking at main meeting, and citizen science satellite meeting)

• MicrobeCensus changes have been pushed recently
  • Tom will check affects on ShotMap

• ShotMap final loose ends
  • Target this Friday for these
  • Submit a week or so later

• Tara Oceans update
  • PATRIC genomes as reference db
    • JE: check for phylogenetically novel genomes
  • Genome clusters
  • Gene clusters within genome clusters
  • DIAMOND for read alignment
    • Faster than RAPSEARCH2 with a lot of reads
    • A lot of compute is indexing of huge protein db
  • Josh working on prioritizing using environmental variables

• Stephen: Strain variation in gene copy number in metagenomes (slides)
  • Used PhyEco markers to identify genome clusters
    • Bacterial specific markers perform best (Bacterial and Archaeal are probably too conserved)
    • Ribosomal proteins have high sequence conservation (too high? maybe due to codon bias?)
  • Perpetual updating
  • Discussion of this paper: http://www.cell.com/cell/abstract/S0092-8674(15)00013-6
    • JE: how does this relate to mapping errors (e.g., repeats)?
    • Future directions for Stephen given this paper
      • They did not generate methods that are easily used by others
      • Create data resources (e.g., genome clusters, pangenomes)
      • Analyze more data, better data
      • Use genes outside KEGG
      • Look at SNPs
      • Unanswered biological questions: Co-variation of genes (genomic islands), biogeography, core genes, gene flux across species
      • JE: Look more broadly at covariation of genes (present/absent, not just copy number variation)
    • Applications to metatranscriptome analysis, "phylogenetic RNA-seq"