wiki:PombasePlanningMeetingMinutes20131120

Planning Meeting Wednesday 20 November 2013 - Minutes


Agenda for this meeting


Wednesday 20 November, 14:00, Cambridge

Present: Val, Midori, Jürg, Antonia, Mark, Dan, Paul, Steve
Remote: Kim

Action item review

Outstanding action, sort database names for blast options

  • AI: Val to make a jira ticket for blast database options (DONE)

Survey results:

Browser - Slow speeds: Paul says that each track takes about a second to load. At the moment they are working to implement (ETA around a month?) a new track loader where data is combined so that loading is quicker.

The poly data is slow to load because it is in bedgraph format. Bigwig should be quicker. Dan says they can try to mangle it into a different format and see what works best. Mark says that there might be a problem converting to bigwig as bedgraphs can be defined with a single base, bigwigs must have at least two bases so is only good for ranges and might therefore not work very well for polyA data.

  • AI: Ensembl to look toward a better format to represent data which is presented at the single nucleotide level and cannot be indexed

Browser - Contig is below tracks:

  • AI: Mark put the contig/gene view above the above the data tracks.

Browser - Viewing

  • AI Mark to zoom out detail browser view once to include context around gene

Blast Long term Sanger -> variation data. Currently developed at Sanger. Expected date will probably be quite long?

  • AI: MArk Add NCBI blast link with protein sequence submission side by side with ensembl blast (at least until a usable blast facility is available)

hosting datasets

The ftp directory we asked people to upload to disappeared. Dan now reinstated it.

  • AI: Midori to email to ask for data resubmissions (DONE).
  • AI: Val to email Paco for nucleosome positioning data (I can't find this paper?).

Split cell vs population phenotypes

Rationale is for querying and also to make display more clear. We should probably make some good documentation about this. For querying it is important because a cell might be inviable or cut, which is inviable, but the penetrance might only be 10 percent so the population is not inviable. This is important for people who might want to know if they can grow a mutant etc.

Grant related

Discussed some possibilities for next grant display of network data

i ) Cytoskape extension support display of interaction data ii) Pathways Cell designer -> Reactome Paul: Reactome would provide infrastructure, we would provide curation. Hard to run parallel for ourselves. iii) Pathways Using GO and annotation extensions Complement with another small bbsrc grant? Automate pathway generation.

  • AI: Val ask Paul Thomas /CJM to explain details of modular/LEGO annotation
  • AI: Val to Summarize tasks need to be done for the grant proposal.

Curation

Discussed gene summaries. Can solicit these using a gene page banner once firs lot are implemented

Last modified 7 years ago Last modified on Nov 20, 2013, 5:41:09 PM