pre-cran submission correction

DESCRIPTION

@@ -1,8 +1,8 @@
 Package: colocboost
 Type: Package
-Date: 2025-11-22
+Date: 2026-06-06
 Title: Multi-Context Colocalization Analysis for QTL and GWAS Studies
-Version: 1.0.7
+Version: 1.0.8
 Authors@R: c(
   person(given = "Xuewei", family = "Cao", email = "xc2270@cumc.columbia.edu", role = c("cre", "aut", "cph")),
   person(given = "Haochen", family = "Sun", email = "hs3393@cumc.columbia.edu", role = c("aut", "cph")),

R/colocboost.R

@@ -424,7 +424,7 @@ colocboost <- function(X = NULL, Y = NULL, # individual data
 #'   \item{sumstat}{Processed list of summary statistics data.frames}
 #'   \item{LD}{Processed list of LD matrices}
 #'   \item{X_ref}{Processed list of reference genotype matrices}
-#'   \item{ref_label}{Style of reference matrics}
+#'   \item{ref_label}{Style of reference matrices}
 #'   \item{sumstatLD_dict}{Dictionary mapping sumstat to LD}
 #'   \item{keep_variable_sumstat}{List of variant names for each sumstat}
 #'   \item{Z}{List of z-scores for each outcome}
@@ -514,7 +514,7 @@ colocboost_validate_input_data <- function(X = NULL, Y = NULL,
     })
     keep_variable_individual <- lapply(X, colnames)
     if (!is.list(X) & !is.list(Y)) {
-      warning("Error: Input X and Y must be the list containing genotype matrics and all phenotype vectors!")
+      warning("Error: Input X and Y must be the list containing genotype matrices and all phenotype vectors!")
       return(NULL)
     } else {
       if (length(X) == 1) {
@@ -968,4 +968,3 @@ colocboost_validate_input_data <- function(X = NULL, Y = NULL,
     npc_outcome_cutoff = npc_outcome_cutoff_updated
   ))
 }
-

R/colocboost_output.R

@@ -255,7 +255,7 @@ get_robust_colocalization <- function(cb_output,
   for (i in 1:length(cos_details$cos$cos_index)) {
     cos_npc_config <- cos_details$cos_outcomes_npc[[i]]
     npc_outcome <- cos_npc_config$npc_outcome
-    pos_pass <- which(npc_outcome >= npc_outcome_cutoff)
+    pos_pass <- which(npc_outcome >= npc_outcome_cutoff & npc_outcome > 0)
     if (!is.null(pvalue_cutoff)) {
       cos_tmp <- cos_details$cos$cos_index[[i]]
       cos_trait <- cos_details$cos_outcomes$outcome_index[[i]]
@@ -517,8 +517,7 @@ get_robust_ucos <- function(cb_output,
   }
 
   if (npc_outcome_cutoff == 0 && is.null(pvalue_cutoff)) {
-    message("All possible uncolocalized events are reported regardless of their relative evidence (npc_outcome_cutoff = 0).")
-    return(cb_output)
+    message("All possible uncolocalized events with positive relative evidence are reported (npc_outcome_cutoff = 0).")
   } else {
     if (is.null(pvalue_cutoff)) {
       message(paste0(
@@ -580,7 +579,7 @@ get_robust_ucos <- function(cb_output,
   ucolocset_names <- ucos_min_npc_outcome <- c()
   for (i in 1:length(ucos_details$ucos$ucos_index)) {
     npc_outcome <- ucos_details$ucos_outcomes_npc$npc_outcome[i]
-    pos_pass <- which(npc_outcome >= npc_outcome_cutoff)
+    pos_pass <- which(npc_outcome >= npc_outcome_cutoff & npc_outcome > 0)
     if (!is.null(pvalue_cutoff)) {
       ucos_tmp <- ucos_details$ucos$ucos_index[[i]]
       ucos_trait <- ucos_details$ucos_outcomes$outcome_index[[i]]
@@ -592,8 +591,6 @@ get_robust_ucos <- function(cb_output,
       pos_pass_pvalue <- which(minPV <= pvalue_cutoff)
       if (length(pos_pass_pvalue) == 0) {
         pos_pass <- NULL
-      } else {
-        pos_pass <- 1
       }
     }
     if (length(pos_pass) == 0) {
@@ -1472,4 +1469,4 @@ merge_ucos_details <- function(ucos_details, ucos_from_cos) {
     ),
     "ucos_outcomes_npc" = rbind(ucos_details$ucos_outcomes_npc, ucos_from_cos$ucos_outcomes_npc)
   )
-}
+}

cran-comments.md

@@ -1,10 +1,26 @@
+## colocboost 1.0.8 release comments
 
-## New release comments
+This is an update to colocboost 1.0.7.
 
-* This release includes three major improvements:
-  - Enhanced main function by adding post-filtering step.
-  - Enhanced plot function with flexible highlighting options.
-  - Optimized performance and computational efficiency.
+This release includes:
+
+* Added X_ref support as a memory-efficient alternative to precomputed LD matrices for summary-statistics workflows.
+* Added and refined robust post-filtering for colocalization and trait-specific uncolocalized events.
+* Improved computational efficiency for repeated matrix products in reference-panel workflows.
+* Improved plotting robustness for extreme association signals and coefficient or z-score displays.
+* Updated documentation and vignettes, including the bioinformatics pipeline vignette.
+
+## R CMD check results
+
+There is one NOTE about installed package size:
+
+* checking installed package size ... NOTE
+  installed size is 5.0 MB
+  sub-directories of 1Mb or more:
+    data 2.0 MB
+    doc  1.9 MB
+
+This NOTE is expected. The installed size is mainly due to reduced example datasets and rendered vignettes with figures. These files are kept to make the tutorials reproducible and self-contained for multi-trait colocalization workflows. No external data are downloaded during examples or vignette rendering.
 
 ## Previous comments
 
@@ -15,4 +31,4 @@
   - Fixed reset users' options issues
   - Added proper COPYRIGHT HOLDER and ORGANIZATION to LICENSE
   - Added explanation of acronyms used in this package to inst/WORDLIST
-* The examples and vignettes use small datasets to avoid long check times
+* The examples and vignettes use small datasets to avoid long check times

inst/WORDLIST

@@ -7,6 +7,7 @@ HyPrColoc       # Hypothesis Prioritization in multi-trait Colocalization method
 SuSiE           # Sum of Single Effects regression model
 
 # Statistical and Genetic Terms
+bim             # PLINK BIM variant information file
 eQTL            # Expression Quantitative Trait Loci
 GWAS            # Genome-Wide Association Study
 INDELs          # Insertions and Deletions
@@ -31,11 +32,14 @@ GTEx            # Genotype-Tissue Expression project
 Micromamba      # Lightweight Conda implementation
 Nealelab        # Lab developing genetics GWAS summary statistics
 PLINK           # Whole genome association analysis toolset
+RAISS           # Summary-statistics imputation method
 tabix           # Tool for indexing genomic data
 UKBB            # UK Biobank dataset
+YAML            # Human-readable configuration file format
 medRxiv         # A preprint resource
 pixi            # An environment manager
 conda           # Package and environment management system
+xz              # Compression format
 
 # Researcher Names
 Jager           # Philip L. de Jager
@@ -48,18 +52,23 @@ et              # and more
 al              # and more
 
 # Technical Terms
+changelog       # Release change log
 cis             # Referring to nearby location of a regulatory element
 chrom           # Chromosome
 decayrate       # Decay rate, an input parameter in our package
 doi             # Digital Object Identifier
+func            # Function abbreviation used in parameter names
 grey            # One color name in R
 iteratively     # Performed through iterations
+loglik          # Log-likelihood abbreviation used in parameter names
 lfsr            # Local False Sign Rate
 lth             # Lower threshold
 modularity      # Property of network structure
 omics           # Collective biological data fields
 phenotypes      # Observable traits
 pos             # Position in genome
+precomputed     # Computed in advance
+precomputes     # Computes in advance
 probabilistically # Based on probability theory
 qc              # Quality Control
 rcond           # Reciprocal condition number
@@ -89,4 +98,4 @@ npc             # Normalization probability of colocalization in our proposed Co
 Pre             # Before
 pre             # Before
 jk              # Index used in ColocBoost
-nd              # Second
+nd              # Second

tests/testthat/test_inference.R

@@ -107,6 +107,38 @@ generate_ucos_test_data <- function(n = 500, p = 60, L = 3, seed = 42, output_le
   return(result)
 }
 
+generate_cos_test_result <- function(n = 250, p = 30, L = 3, seed = 20260606) {
+  set.seed(seed)
+
+  sigma <- 0.9^abs(outer(seq_len(p), seq_len(p), "-"))
+  X <- MASS::mvrnorm(n, rep(0, p), sigma)
+  colnames(X) <- paste0("SNP", seq_len(p))
+
+  true_beta <- matrix(0, p, L)
+  true_beta[8, 1] <- 1.2
+  true_beta[8, 2] <- 1.1
+  true_beta[22, 3] <- 1.2
+
+  Y <- matrix(0, n, L)
+  for (l in seq_len(L)) {
+    Y[, l] <- X %*% true_beta[, l] + rnorm(n, 0, 0.6)
+  }
+
+  suppressWarnings({
+    result <- colocboost(
+      X = replicate(L, X, simplify = FALSE),
+      Y = lapply(seq_len(L), function(l) Y[, l]),
+      M = 80,
+      output_level = 3,
+      cos_npc_cutoff = 0,
+      npc_outcome_cutoff = 0,
+      pvalue_cutoff = NULL
+    )
+  })
+
+  return(result)
+}
+
 
 
 # Test for get_strong_colocalization
@@ -125,6 +157,61 @@ test_that("get_robust_colocalization filters results correctly", {
   expect_error(suppressWarnings(get_robust_colocalization(cb_res, pvalue_cutoff = 0.05)), NA)
 })
 
+test_that("get_robust_colocalization handles validation and early return branches", {
+
+  cb_res <- generate_cos_test_result()
+  expect_false(is.null(cb_res$cos_details))
+
+  expect_error(
+    get_robust_colocalization("not_a_colocboost_object"),
+    "colocboost object"
+  )
+
+  no_cos <- cb_res
+  no_cos$cos_details <- NULL
+  expect_message(
+    no_cos_result <- get_robust_colocalization(no_cos),
+    "No colocalization results"
+  )
+  expect_null(no_cos_result$cos_details)
+
+  expect_warning(
+    bad_pvalue_result <- get_robust_colocalization(cb_res, pvalue_cutoff = 1.5),
+    "pvalue cutoff"
+  )
+  expect_equal(bad_pvalue_result, cb_res)
+
+  expect_message(
+    all_events_result <- get_robust_colocalization(
+      cb_res,
+      cos_npc_cutoff = 0,
+      npc_outcome_cutoff = 0
+    ),
+    "All possible colocalization events"
+  )
+  expect_equal(all_events_result, cb_res)
+})
+
+test_that("get_robust_colocalization removes CoS with zero npc_outcome", {
+
+  cb_res <- generate_cos_test_result()
+  expect_false(is.null(cb_res$cos_details))
+
+  cb_res$cos_details$cos_outcomes_npc[[1]]$npc_outcome <- 0
+
+  expect_message(
+    filtered <- get_robust_colocalization(
+      cb_res,
+      cos_npc_cutoff = 0.2,
+      npc_outcome_cutoff = 0
+    ),
+    "Extracting colocalization results"
+  )
+
+  expect_s3_class(filtered, "colocboost")
+  expect_null(filtered$cos_details)
+})
+
 # Test for get_hierarchical_clusters
 test_that("get_hierarchical_clusters functions correctly", {
   # Test case 1: Simple 2x2 correlation matrix with high correlation
@@ -598,7 +685,7 @@ test_that("get_robust_ucos handles npc_outcome_cutoff = 0 correctly", {
   # With npc_outcome_cutoff = 0 and no pvalue_cutoff, should return unchanged
   expect_message(
     result <- get_robust_ucos(cb_res, npc_outcome_cutoff = 0),
-    "All possible uncolocalized events are reported"
+    "positive relative evidence"
   )
 
   # Should be essentially unchanged
@@ -608,6 +695,33 @@ test_that("get_robust_ucos handles npc_outcome_cutoff = 0 correctly", {
   )
 })
 
+test_that("get_robust_ucos removes zero npc_outcome even with zero cutoff", {
+
+  # Generate test data
+  cb_res <- generate_ucos_test_data(output_level = 2)
+
+  # Skip if no ucos were detected
+  skip_if(is.null(cb_res$ucos_details), "No ucos detected in test data")
+
+  n_ucos_original <- length(cb_res$ucos_details$ucos$ucos_index)
+  cb_res$ucos_details$ucos_outcomes_npc$npc_outcome[1] <- 0
+
+  expect_message(
+    result <- get_robust_ucos(cb_res, npc_outcome_cutoff = 0),
+    "positive relative evidence"
+  )
+
+  if (n_ucos_original == 1) {
+    expect_null(result$ucos_details)
+  } else {
+    expect_equal(
+      length(result$ucos_details$ucos$ucos_index),
+      n_ucos_original - 1
+    )
+    expect_false(any(result$ucos_details$ucos_outcomes_npc$npc_outcome == 0))
+  }
+})
+
 test_that("get_robust_ucos handles missing ucos_details", {
 
   # Generate test data