Changes between Version 8 and Version 9 of UWSummerStatsWorkshop


Ignore:
Timestamp:
Jul 21, 2015, 6:58:23 PM (10 years ago)
Author:
iovercast
Comment:

--

Legend:

Unmodified
Added
Removed
Modified

  • UWSummerStatsWorkshop

    v8 v9  
    8181  * dirichlet vector with k components that sum to 1 
    8282 * [http://rpubs.com/eriqande/scot-cats Rrunstruct] - Usage for the R structure wrapper code 
     83* Tuesday PM 
     84 * Latent variables could make gibbs sampling easier? 
     85 * Haplotyping (Phase) 
     86  * Clark's method (search population for common haploytpes 
     87   * Id unambiguous individuals, construct known haplotypes, disambiguate unknown haplotypes from combinations of known haplotypes. 
     88   * Results may depend on order of observation, frequency is ignored, only matches exact haplotypes, doesn't measure uncertainty 
     89  * Bayesian method 
     90   * iterate through multiple times 
     91   * Use haplotype freq information 
     92   * account for uncertainty 
     93   * Haplotypes will look similar to ones you've seen before. 
     94  * Incorporating recombination is trickier. 
     95  * "Pseudo-Gibbs sampler" 
    8396==== quotes ==== 
    8497* "Out of all the tomorrows we might experience...." 
    8598* "Uncertainty is, intrinsically, personal." 
     99* "Random draws to mapped calls..."