Documentation till partial overlap

R/colocboost_output.R

@@ -794,17 +794,17 @@ get_model_info <- function(cb_obj, outcome_names = NULL) {
   model_coveraged <- cb_obj$cb_model_para$coveraged
   jk_update <- cb_obj$cb_model_para$real_update_jk
   outcome_proximity_obj <- lapply(cb_obj$cb_model, function(cb) cb$obj_path)
-  outcome_coupled_obj <- lapply(cb_obj$cb_model, function(cb) cb$obj_single)
+  outcome_coupled_best_update_obj <- lapply(cb_obj$cb_model, function(cb) cb$obj_single)
   outcome_profile_loglik <- lapply(cb_obj$cb_model, function(cb) cb$profile_loglike_each)
-  names(outcome_proximity_obj) <- names(outcome_coupled_obj) <-
+  names(outcome_proximity_obj) <- names(outcome_coupled_best_update_obj) <-
     names(outcome_profile_loglik) <- outcome_names
   ll <- list(
     "model_coveraged" = model_coveraged,
     "n_updates" = n_updates,
     "profile_loglik" = profile_loglik,
     "outcome_profile_loglik" = outcome_profile_loglik,
     "outcome_proximity_obj" = outcome_proximity_obj,
-    "outcome_coupled_obj" = outcome_coupled_obj,
+    "outcome_coupled_best_update_obj" = outcome_coupled_best_update_obj,
     "jk_update" = jk_update
   )
   return(ll)

R/colocboost_plot.R

@@ -144,13 +144,18 @@ colocboost_plot <- function(cb_output, y = "log10p",
     # - begin plotting
     coloc_cos <- cb_plot_input$cos
     outcomes <- cb_plot_input$outcomes
+    if (length(y)==1) outcome_idx <- 1
     if (is.null(outcome_idx)) {
       if (is.null(coloc_cos)) {
         # - no colocalized effects, draw all outcomes in this region
         if (length(cb_plot_input$outcomes) == 1) {
           message("There is no fine-mapped causal effect in this region!. Showing margianl for this outcome!")
         } else {
-          message("There is no colocalization in this region!. Showing margianl for all outcomes!")
+          if (length(y) == 1){
+            message("There is no colocalization in this region!. Showing VCP = 0!")
+          } else {
+            message("There is no colocalization in this region!. Showing margianl for all outcomes!")
+          }
         }
         outcome_idx <- 1:length(y)
       } else {
@@ -195,10 +200,12 @@ colocboost_plot <- function(cb_output, y = "log10p",
       } else {
         do.call(plot, args)
       }
-      mtext(outcomes[iy],
-        side = cb_plot_init$outcome_legend_pos, line = 0.2, adj = 0.5,
-        cex = cb_plot_init$outcome_legend_size, font = 1
-      )
+      if (length(y)!=1){
+        mtext(outcomes[iy],
+              side = cb_plot_init$outcome_legend_pos, line = 0.2, adj = 0.5,
+              cex = cb_plot_init$outcome_legend_size, font = 1
+        )
+      }
       if (add_vertical) {
         for (iii in 1:length(add_vertical_idx)) {
           abline(v = add_vertical_idx[iii], col = "#E31A1C", lwd = 1.5, lty = "dashed")
@@ -209,6 +216,7 @@ colocboost_plot <- function(cb_output, y = "log10p",
       if (!is.null(coloc_cos)) {
         n.coloc <- length(coloc_cos)
         coloc_index <- cb_plot_input$coloc_index
+        if (length(y)==1){coloc_index = lapply(coloc_index, function(i) 1 )}
         legend_text <- list(col = vector())
         legend_text$col <- head(cb_plot_init$col, n.coloc)
 
@@ -312,6 +320,7 @@ get_input_plot <- function(cb_output, plot_cos_idx = NULL,
   variables <- cb_output$data_info$variables
   Z <- cb_output$data_info$z
   coef <- cb_output$data_info$coef
+  vcp <- list(as.numeric(cb_output$vcp))
 
   # if finemapping
   if (cb_output$data_info$n_outcomes == 1) {
@@ -336,7 +345,7 @@ get_input_plot <- function(cb_output, plot_cos_idx = NULL,
   names(coloc_hits) <- names(coloc_cos)
   if (!is.null(coloc_cos)) {
     # extract vcp each outcome
-    vcp <- lapply(1:length(analysis_outcome), function(iy) {
+    cos_vcp <- lapply(1:length(analysis_outcome), function(iy) {
       pos <- which(sapply(coloc_index, function(idx) iy %in% idx))
       if (length(pos) != 0) {
         w <- do.call(cbind, cb_output$cos_details$cos_vcp[pos])
@@ -372,8 +381,9 @@ get_input_plot <- function(cb_output, plot_cos_idx = NULL,
     } else {
       warnings("No colocalized effects in this region!")
     }
-    coloc_index <- NULL
-    vcp <- rep(0, cb_output$data_info$n_variables)
+    coloc_index <- select_cs <- NULL
+    vcp <- list(rep(0, cb_output$data_info$n_variables))
+    cos_vcp <- lapply(1:length(analysis_outcome), function(iy) rep(0, cb_output$data_info$n_variables) )
   }
   # extract x axis
   if (variant_coord) {
@@ -398,6 +408,7 @@ get_input_plot <- function(cb_output, plot_cos_idx = NULL,
     "x" = x,
     "Zscores" = Z,
     "vcp" = vcp,
+    "cos_vcp" = cos_vcp,
     "coef" = coef,
     "cos" = coloc_cos,
     "cos_hits" = coloc_hits,
@@ -528,18 +539,22 @@ plot_initial <- function(cb_plot_input, y = "log10p",
   } else if (y == "z_original") {
     plot_data <- cb_plot_input$Zscores
     ylab <- "Z score"
+  } else if (y == "cos_vcp") {
+    plot_data <- cb_plot_input$cos_vcp
+    ylab <- "CoS-specific VCP"
+    args$ylim <- c(0, 1)
   } else if (y == "vcp") {
     plot_data <- cb_plot_input$vcp
-    ylab <- "VCP"
     if (length(cb_plot_input$outcomes) == 1) {
       ylab <- "VPA"
     }
+    ylab <- "VCP"
     args$ylim <- c(0, 1)
-  } else if (y == "coef") {
+  }else if (y == "coef") {
     plot_data <- cb_plot_input$coef
     ylab <- "Coefficients"
   } else {
-    stop("Invalid y value! Choose from 'z' or 'z_original'")
+    stop("Invalid y value! Choose from 'log10p', 'z_original', 'vcp', 'coef', or 'cos_vcp'!")
   }
   if (!exists("xlab", args)) args$xlab <- "variables"
   if (!exists("ylab", args)) args$ylab <- ylab
@@ -607,3 +622,4 @@ plot_initial <- function(cb_plot_input, y = "log10p",
 
   return(args)
 }
+

R/colocboost_workhorse.R

@@ -80,7 +80,7 @@ colocboost_workhorse <- function(cb_data,
       # - if all outcomes do not have signals, STOP
       message(paste0(
         "Using multiple testing correction method: ", func_multi_test,
-        ". Stop ColocBoost since no outcomes ", focal_outcome_idx, " have association signals."
+        ". Stop ColocBoost since no outcomes ", focal_outcome_idx, "have association signals."
       ))
     } else {
       message(paste0(

R/data.R

@@ -1,76 +1,92 @@
 #' Individual level data for 5 traits
 #'
-#' An example dataset with simulated genotypes and outcomes for 5 traits
+#' An example dataset with simulated genotypes and traits for 5 traits
 #'
 #' @format ## `Ind_5traits`
 #' A list with 3 elements
 #' \describe{
 #'   \item{X}{List of genotype matrices}
-#'   \item{Y}{List of outcomes}
-#'   \item{TrueCausalVariants}{List of causal variants}
+#'   \item{Y}{List of traits}
+#'   \item{true_effect_variants}{List of causal variants}
 #' }
-#' @source See Cao et. al. 2025 for details. TO-DO-LIST
+#' @source The Ind_5traits dataset contains 5 simulated phenotypes alongside corresponding genotype matrices.
+#' The dataset is specifically designed for evaluating and demonstrating the capabilities of ColocBoost in multi-trait colocalization analysis 
+#' with individual-level data. See Cao et. al. 2025 for details.
+#' 
 #' @family colocboost_data
 "Ind_5traits"
 
 #' Summary level data for 5 traits
 #'
-#' An example dataset with simulated statistics and LD for 5 traits
+#' An example dataset with simulated statistics for 5 traits
 #'
 #' @format ## `Sumstat_5traits`
 #' A list with 2 elements
 #' \describe{
 #'   \item{sumstat}{Summary statistics for 5 traits}
-#'   \item{TrueCausalVariants}{List of causal variants}
+#'   \item{true_effect_variants}{List of causal variants}
 #' }
-#' @source See Cao et. al. 2025 for details. TO-DO-LIST
+#' @source The Sumstat_5traits dataset contains 5 simulated summary statistics, 
+#' where it is directly derived from the Ind_5traits dataset using marginal association. 
+#' The dataset is specifically designed for evaluating and demonstrating the capabilities of ColocBoost 
+#' in multi-trait colocalization analysis with summary association data. See Cao et. al. 2025 for details. 
+#' 
 #' @family colocboost_data
 "Sumstat_5traits"
 
 
-#' Summary level data for 5 traits
+#' Individual level data for 2 traits and 2 causal variants with heterogeneous effects
 #'
-#' An example dataset with simulated statistics and LD for 5 traits
+#' An example dataset with simulated genotypes and traits for 2 traits and 2 common causal variants with heterogeneous effects
 #'
 #' @format ## `Heterogeneous_Effect`
 #' A list with 3 elements
 #' \describe{
 #'   \item{X}{List of genotype matrices}
-#'   \item{Y}{List of outcomes}
-#'   \item{TrueCausalVariants}{List of causal variants}
+#'   \item{Y}{List of traits}
+#'   \item{variant}{Incides of two causal variants}
 #' }
-#' @source See Cao et. al. 2025 for details. TO-DO-LIST
+#' @source The Heterogeneous_Effect dataset contains 2 simulated phenotypes alongside corresponding genotype matrices.
+#' There are two causal variants, both of which have heterogeneous effects on two traits.
+#' See Figure 2b in Cao et. al. 2025 for details. 
+#' 
 #' @family colocboost_data
 "Heterogeneous_Effect"
 
 
-#' Summary level data for 5 traits
+#' Individual level data for 2 traits and 2 causal variants with weaker effects for focal trait
 #'
-#' An example dataset with simulated statistics and LD for 5 traits
+#' An example dataset with simulated genotypes and traits for 2 traits and 2 common causal variants with heterogeneous effects
 #'
 #' @format ## `Weaker_GWAS_Effect`
 #' A list with 3 elements
 #' \describe{
 #'   \item{X}{List of genotype matrices}
-#'   \item{Y}{List of outcomes}
-#'   \item{TrueCausalVariants}{List of causal variants}
+#'   \item{Y}{List of traits}
+#'   \item{variant}{Incides of two causal variants}
 #' }
-#' @source See Cao et. al. 2025 for details. TO-DO-LIST
+#' @source The Weaker_GWAS_Effect dataset contains 2 simulated phenotypes alongside corresponding genotype matrices.
+#' There are two causal variants, one of which has a weaker effect on the focal trait compared to the other trait.
+#' See Figure 2b in Cao et. al. 2025 for details. 
+#' 
 #' @family colocboost_data
 "Weaker_GWAS_Effect"
 
 
-#' Summary level data for 5 traits
+#' Individual level data for 2 traits and 2 causal variants, but the strongest margianl association is not causal
 #'
-#' An example dataset with simulated statistics and LD for 5 traits
+#' An example dataset with simulated genotypes and traits for 2 traits and 2 common causal variants, but the strongest marginal association is not causal variant.
 #'
 #' @format ## `Non_Causal_Strongest_Marginal`
 #' A list with 3 elements
 #' \describe{
 #'   \item{X}{List of genotype matrices}
-#'   \item{Y}{List of outcomes}
-#'   \item{TrueCausalVariants}{List of causal variants}
+#'   \item{Y}{List of traits}
+#'   \item{variant}{Incides of two causal variants}
 #' }
-#' @source See Cao et. al. 2025 for details. TO-DO-LIST
+#' @source The Non_Causal_Strongest_Marginal dataset contains 2 simulated phenotypes alongside corresponding genotype matrices.
+#' There are two causal variants, but the strongest marginal association is not a causal variant.
+#' See Figure 2b in Cao et. al. 2025 for details. 
+#' 
 #' @family colocboost_data
 "Non_Causal_Strongest_Marginal"

README.md

@@ -35,7 +35,7 @@ Learn how to perform colocalization analysis with step-by-step examples. For det
 
 If you use ColocBoost in your research, please cite:
 
-> Cao X, Sun H, Feng R, Mazumder R, Najar CFB, Li YI, de Jager PL, Bennett D, The Alzheimer's Disease Functional Genomics Consortium, Dey KK, Wang G. (2025+). Integrative multi-omics QTL colocalization maps regulatory architecture in aging human brain. bioRxiv. [https://doi.org/](https://doi.org/)
+> Cao X, Sun H, Feng R, Mazumder R, Najar CFB, Li YI, de Jager PL, Bennett D, The Alzheimer's Disease Functional Genomics Consortium, Dey KK, Wang G. (2025+). Integrative multi-omics QTL colocalization maps regulatory architecture in aging human brain. medRxiv. [https://doi.org/](https://doi.org/)
 
 
 ## License

inst/CITATION

@@ -0,0 +1,27 @@
+citHeader("To cite colocboost in publications use:")
+
+citEntry(
+  entry    = "Article",
+  title    = "Integrative multi-omics QTL colocalization maps regulatory architecture in aging human brain",
+  author   = c(person("Xuewei", "Cao"),
+               person("Haochen", "Sun"),
+               person("Ru", "Feng"),
+               person("Rajib", "Mazumder"),
+               person("Cristian F. B.", "Najar"),
+               person("Yang I.", "Li"),
+               person("Philip L.", "de Jager"),
+               person("David", "Bennett"),
+               person("The Alzheimer's Disease Functional Genomics Consortium"),
+               person("Kushal K.", "Dey"),
+               person("Guanghao", "Wang")),
+  journal  = "medRxiv",
+  year     = "2025",
+  note     = "Preprint",
+  url      = "https://doi.org/",
+  textVersion = paste(
+    "Cao X, Sun H, Feng R, Mazumder R, Najar CFB, Li YI, de Jager PL, Bennett D, The Alzheimer's Disease Functional Genomics Consortium, Dey KK, Wang G. (2025).",
+    "Integrative multi-omics QTL colocalization maps regulatory architecture in aging human brain.",
+    "medRxiv.",
+    "https://doi.org/[YOUR_DOI_HERE]"
+  )
+)