Fix Issue rqtl#146: revise predict_snpgeno() so it handles homozygous…

… populations

DESCRIPTION

@@ -1,6 +1,6 @@
 Package: qtl2
-Version: 0.23-11
-Date: 2020-12-08
+Version: 0.23-12
+Date: 2020-12-09
 Title: Quantitative Trait Locus Mapping in Experimental Crosses
 Description: Provides a set of tools to perform quantitative
     trait locus (QTL) analysis in experimental crosses. It is a

NEWS.md

@@ -1,4 +1,4 @@
-## qtl2 0.23-11 (2020-12-08)
+## qtl2 0.23-12 (2020-12-09)
 
 ### Major changes
 
@@ -40,19 +40,22 @@
   `genoprob_to_alleleprob()` for general AIL crosses. We had not
   implemented the `geno2allele_matrix()` function.
 
-- Fix Issue #164, so `plot_pxg()` can handle a phenotype that is a
+- Fixed Issue #164, so `plot_pxg()` can handle a phenotype that is a
   single-column data frame.
 
-- Fix Issue #135, so `plot_scan1()` can take vector input (which is
+- Fixed Issue #135, so `plot_scan1()` can take vector input (which is
   then converted to a single-column matrix).
 
-- Fix Issue #157, to have `calc_genoprob()` give a better error
+- Fixed Issue #157, to have `calc_genoprob()` give a better error
   message about missing genetic map.
 
-- Fix Issue #178, to have `read_cross2()` give a warning not an error
+- Fixed Issue #178, to have `read_cross2()` give a warning not an error
   if incorrect number of alleles.
 
-- Fix Issue #180 re `scan1()` error if phenotypes' rownames have rownames.
+- Fixed Issue #180 re `scan1()` error if phenotypes' rownames have rownames.
+
+- Fixed Issue #146, revising `predict_snpgeno()` so that it works for
+  homozygous populations, like MAGIC lines or the Collaborative Cross.
 
 
 ## qtl2 0.22-11 (2020-07-09)

R/predict_snpgeno.R

@@ -56,8 +56,14 @@ function(cross, geno, cores=1)
        geno <- geno[ind,]
    }
 
+   phase_known <- is_phase_known(cross)
+
    # guess phase
-   ph <- guess_phase(cross, geno, cores=cores)
+   if(!phase_known) {
+       ph <- guess_phase(cross, geno, cores=cores)
+   } else {
+       ph <- cross$geno
+   }
 
    # subset cross chromosomes
    cross <- cross[,names(ph)]
@@ -72,8 +78,12 @@ function(cross, geno, cores=1)
    by_chr_func <- function(chr) {
        # ensure the same markers
        fg <- cross$founder_geno[[chr]]
-       ph1 <- ph[[chr]][,,1]
-       ph2 <- ph[[chr]][,,2]
+       if(phase_known) {
+           ph1 <- ph2 <- geno[[chr]]
+       } else {
+           ph1 <- ph[[chr]][,,1]
+           ph2 <- ph[[chr]][,,2]
+       }
        if(ncol(fg) != ncol(ph1) ||
           any(colnames(fg) != colnames(ph1))) {
            mar <- get_common_ids(setNames(colnames(fg), NULL), setNames(colnames(ph1), NULL))