Moore Notes 6 17 09

From OpenWetWare
Jump to navigationJump to search

Group call (Katie and Jonathan missing)

(Sam taking notes)

  • Progress report due June 30; Jessica Green has started an outline on the OWW page.
    • Everyone should put a paragraph (and possibly, figure) for their work.
  • Sept 10-11 are the dates for our meeting; Jessica Green cannot make it on Sept 9. (Apologies to Jonathan.)
  • Elaine sent to Jessica Green an e-mail question:
    • How to translate 97% similarity cutoff for 16S RNA to another gene, like rpoB or PKS, where there is much less similarity?
    • For example, at 50% similarity, there is still a lot of diversity in the sequences Elaine is looking at, which would have been much reduced in 16S RNA sequences.
    • Jonathan responded that iSEEM proposed to solve this problem and it is in the section on the grant on OTUs.
    • Jessica is confused as to where this is in the grant.
    • Steve noted the section titled "Phylotype OTU Weighting: Sequence conservation and sequence fragments".
    • Steve also noted that Martin has a summary on the wiki somewhere about the scores for different gene families based on something like how conserved they are.
    • It would be useful to discuss this with Martin and Dongying to understand what they did exactly and how it relates to this question.
    • Steve proposes that since the 97% value does not come from some analytical process and was more or less arbitrarily chosen by people, Elaine could just pick any value that seems reasonable to her.
    • Josh notes that the question is really what threshold value is analogous to this threshold for other gene families. How do you figure this out? What does a "corresponding" or "analogous" threshold really mean?
    • Jessica: from looking over what Dongying wrote in the "Gene Family" notes on the wiki, it's not apparent where we are in the project on determining "relative sequence conservation scores" and "position specific conservation scores", and how this relates to evenness versus family size versus universality. Should discuss in the next call!
    • How do we determine OTUs for different gene families? Jessica has an idea about using something like the species-area relationships to do this.
    • Josh: we could determine thresholds so that we get the same numbers of OTUs on average. There may be other ways to interpret what "corresponding thresholds" means -- what do we really want?
    • James: Elaine has samples from PKS genes -- how can you estimate how many OTUs are there so that you get a similar answer to what you would get if you pulled out 16S RNA genes?
  • Jonathan told Jessica he has a "cool" idea for a project for iSEEM:
    • 1. Measure community similarity in metagenomic data with "Steve's" or other metric;
    • 2. Compared to similarity data as measured from transcriptome and proteomic data?
    • Jessica asked him where this data is housed, and Jonathan says he doesn't know.
    • Jessica: There are a bunch of people who have gotten money to get transcriptome and proteomic data. What does this mean? What kind of inferences could we make by looking at biodiversity patterns in this data?
    • If anyone has any thoughts or wants to chat or has paper references about this topic, please come forward.
    • Maybe this is something we can talk about at our next call? What exactly is a transcriptome and a proteome? What are the differences?
    • Josh: via wikipedia: "Transcriptome" is the set of expressed genes, and "Proteome" is the set of proteins coded by the expressed genes.
  • Josh suggests that we look at the new Harte paper in Ecology Letters; he'll post it on the wiki.
    • Let's discuss in future call after more of us have read it.
  • Steve suggests that we schedule a group call for next week since we have a lot to discuss, and some of it depends strongly on members of iSEEM who are not present this week. Sam will send out an e-mail to everyone about this.