Lot's of small changes to satisfy BiocCheck

Author: const-ae

DESCRIPTION

@@ -1,7 +1,7 @@
 Package: proDA
 Type: Package
 Title: Differential Abundance Analysis of Label-Free Mass Spectrometry Data
-Version: 0.0.0.9000
+Version: 0.99.0
 Authors@R: c(person("Constantin", "Ahlmann-Eltze", email = "[email protected]", role = c("aut", "cre")),
              person("Simon", "Anders", email="[email protected]", role="ths"))
 Description: Account for missing values in label-free mass spectrometry data 
@@ -37,4 +37,6 @@ Imports:
     extraDistr
 URL: https://github.com/const-ae/proDA
 BugReports: https://github.com/const-ae/proDA/issues
+biocViews: Proteomics, MassSpectrometry, DifferentialExpression,
+    Bayesian, Regression, Software, Normalization, QualityControl
 VignetteBuilder: knitr

R/AllGenerics.R

@@ -6,6 +6,13 @@
 #' @param object the object to get from
 #' @param ... additional arguments used by the concrete implementation
 #'
+#' @examples
+#'   syn_data <- generate_synthetic_data(n_proteins = 10)
+#'   fit <- proDA(syn_data$Y, design = syn_data$groups)
+#'   abundances(fit)
+#'
+#' @return the original matrix that was fitted
+#'
 #' @seealso \link{accessor_methods} for the implementation for a 'proDAFit' object
 #' @export
 setGeneric("abundances", function(object, ...) standardGeneric("abundances"))
@@ -15,6 +22,13 @@ setGeneric("abundances", function(object, ...) standardGeneric("abundances"))
 #' @param object the object to get from
 #' @param ... additional arguments used by the concrete implementation
 #'
+#' @return a list with the values for each fitted hyper-parameter
+#'
+#' @examples
+#'   syn_data <- generate_synthetic_data(n_proteins = 10)
+#'   fit <- proDA(syn_data$Y, design = syn_data$groups)
+#'   hyper_parameters(fit)
+#'
 #' @seealso \link{accessor_methods} for the implementation for a 'proDAFit' object
 #' @export
 setGeneric("hyper_parameters", function(object, ...) standardGeneric("hyper_parameters"))
@@ -24,6 +38,13 @@ setGeneric("hyper_parameters", function(object, ...) standardGeneric("hyper_para
 #' @param object the object to get from
 #' @param ... additional arguments used by the concrete implementation
 #'
+#' @return a data.frame with information on each fit
+#'
+#' @examples
+#'   syn_data <- generate_synthetic_data(n_proteins = 10)
+#'   fit <- proDA(syn_data$Y, design = syn_data$groups)
+#'   feature_parameters(fit)
+#'
 #' @seealso \link{accessor_methods} for the implementation for a 'proDAFit' object
 #' @export
 setGeneric("feature_parameters", function(object, ...) standardGeneric("feature_parameters"))
@@ -33,6 +54,13 @@ setGeneric("feature_parameters", function(object, ...) standardGeneric("feature_
 #' @param object the object to get from
 #' @param ... additional arguments used by the concrete implementation
 #'
+#' @return a numeric matrix of size `nrow(fit) * p`
+#'
+#' @examples
+#'   syn_data <- generate_synthetic_data(n_proteins = 10)
+#'   fit <- proDA(syn_data$Y, design = syn_data$groups)
+#'   coefficients(fit)
+#'
 #' @seealso \link{accessor_methods} for the implementation for a 'proDAFit' object
 #' @export
 setGeneric("coefficients", function(object, ...) standardGeneric("coefficients"))
@@ -42,6 +70,14 @@ setGeneric("coefficients", function(object, ...) standardGeneric("coefficients")
 #' @param object the object to get from
 #' @param ... additional arguments used by the concrete implementation
 #'
+#' @return a list with as many entries as rows in the data. Each element is
+#'   a p*p matrix
+#'
+#' @examples
+#'   syn_data <- generate_synthetic_data(n_proteins = 10)
+#'   fit <- proDA(syn_data$Y, design = syn_data$groups)
+#'   coefficient_variance_matrices(fit)
+#'
 #' @seealso \link{accessor_methods} for the implementation for a 'proDAFit' object
 #' @export
 setGeneric("coefficient_variance_matrices", function(object, ...) standardGeneric("coefficient_variance_matrices"))
@@ -51,6 +87,14 @@ setGeneric("coefficient_variance_matrices", function(object, ...) standardGeneri
 #'
 #' @param object the object to get from
 #' @param ... additional arguments used by the concrete implementation
+#'
+#' @return a string
+#'
+#' @examples
+#'   syn_data <- generate_synthetic_data(n_proteins = 10)
+#'   fit <- proDA(syn_data$Y, design = syn_data$groups)
+#'   reference_level(fit)
+#'
 #' @seealso \link{accessor_methods} for the implementation for a 'proDAFit' object
 #' @export
 setGeneric("reference_level", function(object, ...) standardGeneric("reference_level"))
@@ -61,6 +105,13 @@ setGeneric("reference_level", function(object, ...) standardGeneric("reference_l
 #' @param object the object to get from
 #' @param ... additional arguments used by the concrete implementation
 #'
+#' @return a list with information on the convergence
+#'
+#' @examples
+#'   syn_data <- generate_synthetic_data(n_proteins = 10)
+#'   fit <- proDA(syn_data$Y, design = syn_data$groups)
+#'   convergence(fit)
+#'
 #' @seealso \link{accessor_methods} for the implementation for a 'proDAFit' object
 #' @export
 setGeneric("convergence", function(object, ...) standardGeneric("convergence"))
@@ -71,6 +122,14 @@ setGeneric("convergence", function(object, ...) standardGeneric("convergence"))
 #' @param object the object for which the distance is approximated
 #' @param ... additional arguments used by the concrete implementation
 #'
+#' @return a list with two elements: `mean` and `sd` both are formally
+#'   of class "dist"
+#'
+#' @examples
+#'   syn_data <- generate_synthetic_data(n_proteins = 10)
+#'   fit <- proDA(syn_data$Y, design = syn_data$groups)
+#'   dist_approx(fit)
+#'
 #' @seealso \code{\link[stats]{dist}} for the base R function and
 #'   \code{\link[proDA:dist_approx,proDAFit-method]{dist_approx()}} concrete implementation
 #'   for 'proDAFit' objects
@@ -83,6 +142,13 @@ setGeneric("dist_approx", function(object, ...) standardGeneric("dist_approx"))
 #' @param fit the fit to get the result_names from
 #' @param ... additional arguments used by the concrete implementation
 #'
+#' @return a character vector
+#'
+#' @examples
+#'   syn_data <- generate_synthetic_data(n_proteins = 10)
+#'   fit <- proDA(syn_data$Y, design = syn_data$groups)
+#'   result_names(fit)
+#'
 #' @export
 setGeneric("result_names", function(fit, ...) standardGeneric("result_names"))
 

R/calculate_distance.R

@@ -76,7 +76,8 @@ dist_approx_mean_var <- function(Y, Pred, X, coef_var, by_sample = FALSE){
                Pred)
 
   Pred_var <- mply_dbl(seq_len(nrow(Y)), function(i){
-    sapply(seq_len(nrow(X)), function(j) t(X[j,])  %zero_dom_mat_mult% coef_var[[i]]  %zero_dom_mat_mult% X[j,])
+    vapply(seq_len(nrow(X)), function(j) t(X[j,])  %zero_dom_mat_mult% coef_var[[i]]  %zero_dom_mat_mult% X[j,],
+           FUN.VALUE = 0.0)
   }, ncol=ncol(Y))
   Mu_var <- ifelse(! is.na(Y),
                    0,

R/generate_synthetic_data.R

@@ -49,6 +49,12 @@
 #'   if \code{return_summarized_experiment} is \code{FALSE}. Otherwise
 #'   returns a \code{SummarizedExperiment} with the same information.
 #'
+#' @examples
+#'   syn_data <- generate_synthetic_data(n_proteins = 10)
+#'   names(syn_data)
+#'   se <- generate_synthetic_data(n_proteins = 10, return_summarized_experiment = TRUE)
+#'   se
+#'
 #' @export
 generate_synthetic_data <- function(n_proteins, n_conditions = 2,
                                     n_replicates = 3,

R/methods.R

@@ -12,6 +12,8 @@
 #'   to get the formula that was used to create the linear model.
 #'   If no formula was used \code{NULL} is returned.
 #'
+#' @return See the documentation of the generics to find out what each method returns
+#'
 #' @name accessor_methods
 NULL
 

R/normalization.R

@@ -13,11 +13,17 @@
 #'
 #' @param X a matrix of proteins and samples
 #' @return the normalized matrix
+#'
+#' @examples
+#'   syn_data <- generate_synthetic_data(n_proteins = 10)
+#'   normalized_data <- median_normalization(syn_data$Y)
+#'   normalized_data
+#'
 #' @export
 median_normalization <- function(X){
   stopifnot(length(dim(X)) == 2)
   Xnorm <- X
-  for(idx in 1:ncol(X)){
+  for(idx in seq_len(ncol(X))){
     Xnorm[, idx] <- X[, idx, drop=FALSE] -
       median(X[, idx, drop=FALSE] - rowMeans(X, na.rm=TRUE), na.rm=TRUE )
   }

R/proDA.R

@@ -108,8 +108,10 @@ if(getRversion() >= "2.15.1")  utils::globalVariables(c("Condition1", "Condition
 #' @examples
 #'
 #' # Quick start
+#'
+#' # Import the proDA package if you haven't already done so
+#' # library(proDA)
 #' set.seed(1)
-#' library(proDA)
 #' syn_data <- generate_synthetic_data(n_proteins = 10)
 #' fit <- proDA(syn_data$Y, design = syn_data$groups)
 #' fit
@@ -295,7 +297,9 @@ fit_parameters_loop <- function(Y, model_matrix, location_prior_df,
   })
   Pred_init <- msply_dbl(res_init, function(x) x$coefficients) %*% t(model_matrix)
   Pred_init_var <- mply_dbl(seq_len(nrow(Y)), function(i){
-    sapply(seq_len(nrow(model_matrix)), function(j) t(model_matrix[j,]) %*% res_init[[i]]$coef_variance_matrix %*% model_matrix[j,])
+    vapply(seq_len(nrow(model_matrix)), function(j)
+      t(model_matrix[j,]) %*% res_init[[i]]$coef_variance_matrix %*% model_matrix[j,],
+      FUN.VALUE = 0.0)
   }, ncol=ncol(Y))
   s2_init <-  vapply(res_init, function(x) x[["s2"]], 0.0)
   df_init <- vapply(res_init, function(x) x[["df"]], 0.0)
@@ -355,10 +359,14 @@ fit_parameters_loop <- function(Y, model_matrix, location_prior_df,
     Pred_unreg <- msply_dbl(res_unreg, function(x) x$coefficients) %zero_dom_mat_mult% t(model_matrix)
     Pred_reg <- msply_dbl(res_reg, function(x) x$coefficients) %zero_dom_mat_mult% t(model_matrix)
     Pred_var_unreg <- mply_dbl(seq_len(nrow(Y)), function(i){
-      sapply(seq_len(nrow(model_matrix)), function(j) t(model_matrix[j,]) %zero_dom_mat_mult% res_unreg[[i]]$coef_variance_matrix  %zero_dom_mat_mult% model_matrix[j,])
+      vapply(seq_len(nrow(model_matrix)), function(j)
+        t(model_matrix[j,]) %zero_dom_mat_mult% res_unreg[[i]]$coef_variance_matrix  %zero_dom_mat_mult% model_matrix[j,],
+        FUN.VALUE = 0.0)
     }, ncol=ncol(Y))
     Pred_var_reg <- mply_dbl(seq_len(nrow(Y)), function(i){
-      sapply(seq_len(nrow(model_matrix)), function(j) t(model_matrix[j,]) %zero_dom_mat_mult% res_reg[[i]]$coef_variance_matrix  %zero_dom_mat_mult% model_matrix[j,])
+      vapply(seq_len(nrow(model_matrix)), function(j)
+        t(model_matrix[j,]) %zero_dom_mat_mult% res_reg[[i]]$coef_variance_matrix  %zero_dom_mat_mult% model_matrix[j,],
+        FUN.VALUE = 0.0)
     }, ncol=ncol(Y))
     s2_unreg <-  vapply(res_unreg, function(x) x[["s2"]], 0.0)
     df_unreg <-vapply(res_unreg, function(x) x[["df"]], 0.0)

R/stats_functions.R

@@ -54,6 +54,9 @@ invprobit <- function(x, rho, zeta, log=FALSE, oneminus=FALSE){
 }
 
 #' Same thing as invprobit, but without the parameter validation
+#'
+#' @return a numeric vector of \code{length(x)}
+#'
 #' @keywords internal
 invprobit_fast <- function(x, rho, zeta, log=FALSE, oneminus=FALSE){
   sign_sum <- sum(sign(zeta), na.rm=TRUE)

R/util.R

@@ -79,6 +79,11 @@ msply_dbl <- function(x, FUN, ...){
 
 #' Helper function that makes sure that NA * 0 = 0 in matrix multiply
 #'
+#' @param X a matrix of size `n*m`
+#' @param Y a matrix of size `m*p`
+#'
+#' @return a matrix of size `n*p`
+#'
 #' @keywords internal
 `%zero_dom_mat_mult%` <- function(X, Y){
   X[is.infinite(X)] <- NA

README.Rmd

@@ -339,8 +339,14 @@ test_res
 This walkthrough ends with the identification which proteins are differentially abundant. But for
 a real dataset, now the actual analysis only just begins. A list of significant proteins is hardly
 ever a publishable result, we need to make sense what the underlying biological mechanisms are. The
-precise question that should be asked very much dependent on the biological problem, but some
-helpful tool are gene ontology (GO) term analysis, set enrichment tests, and 
+precise question that should be asked very much dependent on the biological problem.
+
+
+# Session Info
+
+```{r}
+sessionInfo()
+```
 
 
 

README.md

@@ -433,5 +433,48 @@ differentially abundant. But for a real dataset, now the actual analysis
 only just begins. A list of significant proteins is hardly ever a
 publishable result, we need to make sense what the underlying biological
 mechanisms are. The precise question that should be asked very much
-dependent on the biological problem, but some helpful tool are gene
-ontology (GO) term analysis, set enrichment tests, and
+dependent on the biological problem.
+
+# Session Info
+
+``` r
+sessionInfo()
+#> R version 3.6.0 (2019-04-26)
+#> Platform: x86_64-pc-linux-gnu (64-bit)
+#> Running under: Ubuntu 18.04.2 LTS
+#> 
+#> Matrix products: default
+#> BLAS:   /usr/lib/x86_64-linux-gnu/openblas/libblas.so.3
+#> LAPACK: /usr/lib/x86_64-linux-gnu/libopenblasp-r0.2.20.so
+#> 
+#> locale:
+#>  [1] LC_CTYPE=en_US.UTF-8       LC_NUMERIC=C               LC_TIME=de_DE.UTF-8       
+#>  [4] LC_COLLATE=en_US.UTF-8     LC_MONETARY=de_DE.UTF-8    LC_MESSAGES=en_US.UTF-8   
+#>  [7] LC_PAPER=de_DE.UTF-8       LC_NAME=C                  LC_ADDRESS=C              
+#> [10] LC_TELEPHONE=C             LC_MEASUREMENT=de_DE.UTF-8 LC_IDENTIFICATION=C       
+#> 
+#> attached base packages:
+#> [1] stats     graphics  grDevices utils     datasets  methods   base     
+#> 
+#> other attached packages:
+#> [1] proDA_0.0.0.9000
+#> 
+#> loaded via a namespace (and not attached):
+#>  [1] Rcpp_1.0.1                  RColorBrewer_1.1-2          pillar_1.3.1               
+#>  [4] compiler_3.6.0              GenomeInfoDb_1.20.0         XVector_0.24.0             
+#>  [7] bitops_1.0-6                tools_3.6.0                 zlibbioc_1.30.0            
+#> [10] digest_0.6.18               gtable_0.3.0                evaluate_0.13              
+#> [13] tibble_2.1.1                lattice_0.20-38             pkgconfig_2.0.2            
+#> [16] rlang_0.3.4                 Matrix_1.2-17               cli_1.1.0                  
+#> [19] DelayedArray_0.10.0         rstudioapi_0.10             yaml_2.2.0                 
+#> [22] parallel_3.6.0              xfun_0.6                    GenomeInfoDbData_1.2.1     
+#> [25] stringr_1.4.0               extraDistr_1.8.10           knitr_1.22                 
+#> [28] S4Vectors_0.22.0            IRanges_2.18.0              stats4_3.6.0               
+#> [31] grid_3.6.0                  Biobase_2.44.0              fansi_0.4.0                
+#> [34] BiocParallel_1.18.0         rmarkdown_1.12              pheatmap_1.0.12            
+#> [37] magrittr_1.5                scales_1.0.0                htmltools_0.3.6            
+#> [40] matrixStats_0.54.0          BiocGenerics_0.30.0         GenomicRanges_1.36.0       
+#> [43] assertthat_0.2.1            SummarizedExperiment_1.14.0 colorspace_1.4-1           
+#> [46] utf8_1.1.4                  stringi_1.4.3               munsell_0.5.0              
+#> [49] RCurl_1.95-4.12             crayon_1.3.4
+```