Moore Notes 1 8 14

Group Call

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

• What conferences are people attending in 2014?
  • Tom considering: ASM (Boston in May), ISME (S. Korea in August), SMBE (Costa Rica in June)
  • Katie attending SMBE and Arrowhead
  • Josh will probably go to ESA (August)
  • Patrick and/or Stephen interested in human microbiome Keystone meeting (Montana in April) http://www.keystonesymposia.org/index.cfm?e=web.Meeting.Program&meetingid=1269

• PICRUSTs update
  • Last time talked about KO number based analysis
  • Moved analysis over to module identifiers (much bigger, so slower to run even with top 100 PICRUSTs predictions) http://edhar.genomecenter.ucdavis.edu/~gjospin/picrust_test/picrust_test/Jan_6_2014_modules/top100/
    • Correlation is better than previous analysis, but still some large residuals, e.g., compare to: http://edhar.genomecenter.ucdavis.edu/~gjospin/picrust_test/picrust_test/Nov_18_2013/top100/Genome_Freq_Apr_Day_top100_Apr_Day_mg_top100.pdf

• Any news on Tara Oceans data?
  • Some data sets in NCBI, but not recently released and not too many locations
  • Tara group did send a query to PLoS Biology about publishing an overview paper (no data in that paper, so said they were submitting data papers elsewhere)

• EFI: http://enzymefunction.org
  • Katie talked with John Gerlt (Director) and Matt Jacobson (PI)
  • They are interested in our novel and unannotated gene families:
    • SFams with no PFAM annotation, good QC, and >1 sequence (Tom: maybe focus on groups of sequence similar SFams)
    • Microbiome assemblies
  • Blue waters compute time
    • All-vs-all BLAST
    • Adding metagenomes to SFams
  • JE: more novel functions outside human micro biome
    • Patrick: Bacteriodetes has particularly low level of annotation

• Shattuck data storage: server is getting full
  • Intermediate data generated by shot map is huge
    • Some things could be deleted (can easily regenerate later if needed), e.g.,
      • Translated reads and output of splitting on stops (only takes an hour to redo this)
      • Rapsearch raw search results
  • Stephen: How to implement?
    • Could write a post-processing script
    • Katie: would be nice to have the post-processing script automatically run be default
  • Tom: What about advanced compression techniques (better than gzip)
    • JE: DNA and other supposed to be somewhat superior (Gale Rosen)
    • Stephen: bzip is better, but slower

• Guillaume: check on opening ports on shattuck for SFam distribution

• Jonathan: Not using WordPress website so will delete

• No call on 1/22. Next week: Maybe Stephen