port all fixes (#509)

Author: danielnachun

NAMESPACE

@@ -192,7 +192,7 @@ export(penalized_rss)
 export(postprocess_finemapping_fits)
 export(prs_cs)
 export(prs_cs_weights)
-export(qc_individual_data)
+export(qc_regional_data)
 export(raiss)
 export(readAnnotations)
 export(readGenotypes)
@@ -201,7 +201,9 @@ export(read_afreq)
 export(read_sldsc_trait)
 export(region_data_to_colocboost_input)
 export(region_data_to_ind_input)
+export(region_data_to_mvsusie_rss_input)
 export(region_data_to_rss_input)
+export(region_data_to_susie_rss_input)
 export(region_to_df)
 export(regions_overlap)
 export(robust_mahalanobis)

R/AllClasses.R

@@ -2,6 +2,8 @@
 #' @description All S4 class definitions for pecotmr: genotype handles,
 #'   summary statistics, LD references, annotations, fine-mapping results,
 #'   TWAS weights, regional data, and heritability estimates.
+#' @name pecotmr-classes
+#' @keywords internal
 #' @importFrom methods setClass setMethod new is validObject
 #' @importFrom S4Vectors mcols mcols<-
 NULL

R/AllGenerics.R

@@ -1,5 +1,7 @@
 #' @title S4 Generic Function Definitions
 #' @description All S4 generic function definitions for pecotmr.
+#' @name pecotmr-generics
+#' @keywords internal
 #' @include AllClasses.R
 #' @importFrom methods setGeneric
 NULL

R/AllMethods.R

@@ -1,6 +1,8 @@
 #' @title S4 Method Implementations
 #' @description Constructors and accessor method implementations for all
 #'   S4 classes: LDData, RegionalData, FineMappingResult, TWASWeights.
+#' @name pecotmr-methods
+#' @keywords internal
 #' @include AllGenerics.R
 #' @importFrom SummarizedExperiment assay
 #' @importFrom S4Vectors DataFrame mcols mcols<-

R/colocboost_pipeline.R

@@ -179,7 +179,7 @@ region_data_to_colocboost_input <- function(region_data) {
 #' # Use richer LD metadata from load_LD_matrix() for QC, while still passing
 #' # ColocBoost's native LD input.
 #' ld_data <- load_LD_matrix(ld_meta_file, region)
-#' fit <- colocboost_analysis(sumstat = sumstat, LD = ld_data$LD_matrix,
+#' fit <- colocboost_analysis(sumstat = sumstat, LD = getCorrelation(ld_data),
 #'                            LD_reference_info = ld_data, qc_method = "none")
 #'
 #' # Individual-level input with explicit genotype QC thresholds.
@@ -678,14 +678,16 @@ colocboost_pipeline <- function(
 #' @param region_data A region data loaded from \code{load_regional_data}.
 #' @param maf_cutoff A scalar to remove variants with maf < maf_cutoff.
 #' @param pip_cutoff_to_skip_ind A vector of cutoff values for skipping individual contexts.
+#' @param keep_indel Logical; if \code{FALSE}, remove indel variants during
+#'   summary-statistic allele harmonization.
 #' @param pip_cutoff_to_skip_sumstat A vector of cutoff values for skipping summary-stat studies.
 #' @param qc_method Quality control method to use. Options are "none",
 #'   "slalom", or "dentist". \code{NULL} is treated as \code{"none"} for
 #'   basic-only summary-stat preprocessing.
 #' @param impute Logical; if TRUE, performs imputation when required metadata are available.
 #' @param impute_opts A list of imputation options.
 #' @return A list containing post-QC \code{individual_data} and \code{sumstat_data}.
-#' @noRd
+#' @export
 qc_regional_data <- function(region_data,
                              # - individual
                              maf_cutoff = 0.0005,
@@ -694,7 +696,7 @@ qc_regional_data <- function(region_data,
                              keep_indel = TRUE,
                              pip_cutoff_to_skip_sumstat = 0,
                              qc_method = NULL,
-                             impute = TRUE,
+                             impute = FALSE,
                              impute_opts = list(rcond = 0.01, R2_threshold = 0.6, minimum_ld = 5, lamb = 0.01)) {
   qc_method <- .resolve_summary_qc_method(qc_method)
   qced_individual_to_region_data <- function(ind_qc) {
@@ -793,7 +795,7 @@ qc_regional_data <- function(region_data,
 #' @param pip_cutoff_to_skip Optional single-effect PIP cutoff.
 #' @return A named list of cleaned context-level \code{X}/\code{Y} records, or
 #'   one cleaned record for matrix inputs.
-#' @export
+#' @noRd
 qc_individual_data <- function(X, Y, maf = NULL, X_variance = NULL,
                                missing_rate_thresh = NULL,
                                maf_cutoff = 0.0005,

R/file_utils.R

@@ -1376,6 +1376,9 @@ standardise_sumstats_columns <- function(sumstats, column_file_path = NULL, comm
 #' @export
 load_rss_data <- function(sumstat_path, column_file_path = NULL, n_sample = 0, n_case = 0, n_control = 0, region = NULL,
                           extract_region_name = NULL, region_name_col = NULL, comment_string = "#") {
+  n_sample <- if (length(n_sample) == 1L && is.na(n_sample)) 0 else n_sample
+  n_case <- if (length(n_case) == 1L && is.na(n_case)) 0 else n_case
+  n_control <- if (length(n_control) == 1L && is.na(n_control)) 0 else n_control
   # Validate input files exist
   if (!file.exists(sumstat_path)) {
     stop("Summary statistics file not found: ", sumstat_path)

R/genotype_io.R

@@ -2,6 +2,8 @@
 #' @description Read genotype data from various formats (VCF, plink1,
 #'   plink2, GDS) and provide block-level genotype extraction without
 #'   requiring format conversion.
+#' @name pecotmr-genotype-io
+#' @keywords internal
 #' @importFrom SummarizedExperiment SummarizedExperiment rowRanges
 #' @importFrom GenomicRanges GRanges seqnames
 #' @importFrom S4Vectors DataFrame mcols mcols<-

R/gwas_sumstats.R

@@ -1,6 +1,8 @@
 #' @title Summary Statistics Handling
 #' @description Functions for reading, validating, and constructing
 #'   \code{GWASSumStats} objects from various input formats.
+#' @name pecotmr-gwas-sumstats
+#' @keywords internal
 #' @importFrom GenomicRanges GRanges seqnames start
 #' @importFrom S4Vectors DataFrame mcols mcols<-
 #' @importFrom MungeSumstats format_sumstats

R/h2_annotations.R

@@ -1,6 +1,8 @@
 #' @title Annotation Handling for Stratified Heritability
 #' @description Read and manage genomic annotations for stratified
 #'   heritability analysis. Supports BED, BigWig, and LDSC .annot formats.
+#' @name pecotmr-h2-annotations
+#' @keywords internal
 #' @importFrom tools file_ext
 #' @importFrom GenomicRanges GRanges
 #' @include AllGenerics.R

R/h2_estimate_wrappers.R

@@ -2,6 +2,8 @@
 #' @description Top-level entry point for heritability estimation,
 #'   LD score computation methods, H2Estimate accessors, and a converter
 #'   to bridge H2Estimate into the sldsc_wrapper.R postprocessing pipeline.
+#' @name pecotmr-h2-wrappers
+#' @keywords internal
 #' @include AllGenerics.R
 #' @importFrom stats median
 NULL