[Genabel-commits] r712 - in pkg/GenABEL: . R man

[noreply at r-forge.r-project.org]

Author: yurii
Date: 2011-05-05 15:06:37 +0200 (Thu, 05 May 2011)
New Revision: 712

Modified:
   pkg/GenABEL/CHANGES.LOG
   pkg/GenABEL/DESCRIPTION
   pkg/GenABEL/R/polygenic_hglm.R
   pkg/GenABEL/R/zzz.R
   pkg/GenABEL/man/polygenic_hglm.Rd
Log:
upgrade dev version to 1.6-7; small fix in polygenic_hglm example

Modified: pkg/GenABEL/CHANGES.LOG
===================================================================
--- pkg/GenABEL/CHANGES.LOG	2011-05-05 11:14:04 UTC (rev 711)
+++ pkg/GenABEL/CHANGES.LOG	2011-05-05 13:06:37 UTC (rev 712)
@@ -1,5 +1,14 @@
-*** v. 1.6-6 (2011.04.18)
+*** v. 1.6-7 (2011.05.05)
 
+upgrade version number
+
+fix in example of polygenic_hglm
+
+*** v. 1.6-6 (2011.04.29)
+
+Nicola Prirastu added LiLog and LiLogCC: procedures facilitating 
+analysis of binary outcomes using mixed models.  
+
 Added Xia Shen's procedure 'polygenic_hglm'. Features: quick 
 convergence, standard errors for fixed effects.
 

Modified: pkg/GenABEL/DESCRIPTION
===================================================================
--- pkg/GenABEL/DESCRIPTION	2011-05-05 11:14:04 UTC (rev 711)
+++ pkg/GenABEL/DESCRIPTION	2011-05-05 13:06:37 UTC (rev 712)
@@ -1,8 +1,8 @@
 Package: GenABEL
 Type: Package
 Title: genome-wide SNP association analysis
-Version: 1.6-6
-Date: 2011-03-31
+Version: 1.6-7
+Date: 2011-05-05
 Author: Yurii Aulchenko et al.
 Maintainer: Yurii Aulchenko <i.aoultchenko at erasmusmc.nl>
 Depends: R (>= 2.10.0), methods, MASS

Modified: pkg/GenABEL/R/polygenic_hglm.R
===================================================================
--- pkg/GenABEL/R/polygenic_hglm.R	2011-05-05 11:14:04 UTC (rev 711)
+++ pkg/GenABEL/R/polygenic_hglm.R	2011-05-05 13:06:37 UTC (rev 712)
@@ -3,6 +3,14 @@
 #' Estimation of polygenic model using a hierarchical generalized linear model 
 #' (HGLM; Lee and Nelder 1996. \code{hglm} package; Ronnegard et al. 2010).
 #' 
+#' The algorithm gives extended quasi-likelihood (EQL) estimates
+#' for restricted maximum likelihood (REML) (Ronnegard et al. 2010).
+#' Implemented is the inter-connected GLM interpretation of HGLM (Lee and Nelder 2001).
+#' Testing on fixed and random effects estimates are directly done using inter-connected
+#' GLMs. Testing on dispersion parameters (variance components) can be done by extracting
+#' the profile likelihood function value of REML.
+#' 
+#' 
 #' @param formula Formula describing fixed effects to be used in analysis, e.g. 
 #' y ~ a + b means that outcome (y) depends on two covariates, a and b. 
 #' If no covariates used in analysis, skip the right-hand side of the 
@@ -20,17 +28,7 @@
 #' 
 #' @author Xia Shen, Yurii Aulchenko
 #' 
-#' @details
-#'
-#' The algorithm gives extended quasi-likelihood (EQL) estimates
-#' for restricted maximum likelihood (REML) (Ronnegard et al. 2010).
-#' Implemented is the inter-connected GLM interpretation of HGLM (Lee and Nelder 2001).
-#' Testing on fixed and random effects estimates are directly done using inter-connected
-#' GLMs. Testing on dispersion parameters (variance components) can be done by extracting
-#' the profile likelihood function value of REML.
-#' 
 #' @references 
-#'
 #' Ronnegard, L, Shen, X and Alam, M (2010). hglm: A Package For Fitting 
 #' Hierarchical Generalized Linear Models. \emph{The R Journal}, \bold{2}(2), 20-28.
 #'
@@ -90,7 +88,7 @@
 #' # Self, SG and Liang KY (1987) Journal of the American Statistical Association.
 #' qchisq(((1 - .05) - .50)/.50, 1)
 #' # p-value
-#' pchisq(S, 1, lower.tail = FALSE)*2
+#' pchisq(S, 1, lower.tail = FALSE)/2
 #' 
 #' #' # ----- for binary traits
 #' h2dm <- polygenic_hglm(dm2 ~ sex + age, kin = gkin, df, family = binomial(link = 'logit'))

Modified: pkg/GenABEL/R/zzz.R
===================================================================
--- pkg/GenABEL/R/zzz.R	2011-05-05 11:14:04 UTC (rev 711)
+++ pkg/GenABEL/R/zzz.R	2011-05-05 13:06:37 UTC (rev 712)
@@ -1,6 +1,6 @@
 .onLoad <- function(lib, pkg) {
-	GenABEL.version <- "1.6-6"
-	cat("GenABEL v.",GenABEL.version,"(March 31, 2011) loaded\n")
+	GenABEL.version <- "1.6-7"
+	cat("GenABEL v.",GenABEL.version,"(May 05, 2011) loaded\n")
 	
 	# check for updates and news
 	address <- c(

Modified: pkg/GenABEL/man/polygenic_hglm.Rd
===================================================================
--- pkg/GenABEL/man/polygenic_hglm.Rd	2011-05-05 11:14:04 UTC (rev 711)
+++ pkg/GenABEL/man/polygenic_hglm.Rd	2011-05-05 13:06:37 UTC (rev 712)
@@ -5,7 +5,14 @@
     conv=1e-06, maxit=100, ...)}
 \description{Estimation of polygenic model}
 \details{Estimation of polygenic model using a hierarchical generalized linear model 
-(HGLM; Lee and Nelder 1996. \code{hglm} package; Ronnegard et al. 2010).}
+(HGLM; Lee and Nelder 1996. \code{hglm} package; Ronnegard et al. 2010).
+
+The algorithm gives extended quasi-likelihood (EQL) estimates
+for restricted maximum likelihood (REML) (Ronnegard et al. 2010).
+Implemented is the inter-connected GLM interpretation of HGLM (Lee and Nelder 2001).
+Testing on fixed and random effects estimates are directly done using inter-connected
+GLMs. Testing on dispersion parameters (variance components) can be done by extracting
+the profile likelihood function value of REML.}
 \author{Xia Shen, Yurii Aulchenko}
 \references{Ronnegard, L, Shen, X and Alam, M (2010). hglm: A Package For Fitting 
 Hierarchical Generalized Linear Models. \emph{The R Journal}, \bold{2}(2), 20-28.
@@ -77,7 +84,7 @@
 # Self, SG and Liang KY (1987) Journal of the American Statistical Association.
 qchisq(((1 - .05) - .50)/.50, 1)
 # p-value
-pchisq(S, 1, lower.tail = FALSE)*2
+pchisq(S, 1, lower.tail = FALSE)/2
 
 #' # ----- for binary traits
 h2dm <- polygenic_hglm(dm2 ~ sex + age, kin = gkin, df, family = binomial(link = 'logit'))