Minutes from Extra R-Genetics Telecon

Date: Friday, December 16, 2005 Time: 3:30 PM Present: Greg, David, Nitin, Ross.

1. possible next teleconference January 6.
2. Bioconductor moving to slots for phenodata which is a good and useful idea, and metadata which will use MIAME http://www.mged.org/Workgroups/MIAME/miame.html We can recycle or pervert that sufficiently for documenting genesets - there is an array specific section which we could generalise to a genotyping platform descriptor perhaps.
3. How to break the r-genetics project into components/libraries. Could be informed/driven by use cases for common end applications - eg affy snpchip case control or family data. What goes Where? Specifically, what goes into GeneticsBase? (read, write, hwe, ld, freqtables,..)

   What's in a library?

   suggested pattern:

   GeneticsPed (Mendel inconsistencies, relatedness, viewped, GRR,...) (David mentioned a ?potential developer - will report back.)

   GeneticsTDT? (fbat...)

   GeneticsHap? (snphap...)

   GeneticsCaco? (armitage, haplo.score, simdata, power,..)

4. Clean fbat up.

   remove reduntant checks at start of each function.

   remove readdata/hwe - use GeneticsBase.

   Move Mendel into GeneticsPed? Read data from GeneticsBase and then prepare a Pedigree object - derived from geneset. Optionally perform tests for appropriate data, mendel checks etc so these don't need to be repeated anywhere else.

5. Preparation for an alpha release - vignettes for fbat, descriptive.
6. How should we present r-genetics publicly for the alpha release - formally housed at bioconductor as a subproject seemed very wise and sensible.

   Vince is happy to accommodate us.

   It will certainly help users find us.

   We can maintain our sourceforge development distribution and join the bioc release cycle for beta and production builds, but we could use a separate dynamic development site.