filter chromosom started

NAMESPACE

@@ -3,30 +3,39 @@
 export(above_percentage)
 export(add_assay_classification)
 export(add_assay_norm)
+export(add_coldata_classification)
 export(calc_auc)
 export(calc_prediction)
 export(calculate_fit)
+export(correct_output)
 export(factorize_columns)
 export(filter_duplicates)
 export(filter_for_factor)
 export(filter_rare)
 export(filter_single_level)
 export(filter_zero)
 export(get_no_cores)
+export(is_correct)
 export(make_coeff_frame)
 export(make_explicit_na)
 export(make_model_matrix)
 export(merge_duplicated)
 export(methyl_frame)
 export(not_zero)
+export(plot_heatmap)
 export(preprocess_assay)
 export(preprocess_coldata)
 export(reorder_all)
+export(transform_assay)
 export(use_cvglmnet)
 export(use_glmnet)
+export(use_p_adjust)
+export(use_siber)
 export(use_silvermantest)
 import(AUC)
+import(ComplexHeatmap)
 import(GenomicRanges)
+import(Heatmap)
 import(SIBER)
 import(SummarizedExperiment)
 import(doParallel)
@@ -41,10 +50,12 @@ import(rlang)
 import(snow)
 import(survival)
 import(survminer)
-importFrom(ComplexHeatmap,Heatmap)
+importFrom(Biobase,isUnique)
 importFrom(ComplexHeatmap,HeatmapAnnotation)
 importFrom(FactoMineR,catdes)
 importFrom(RColorBrewer,brewer.pal)
+importFrom(biomaRt,getBM)
+importFrom(biomaRt,useMart)
 importFrom(dplyr,"%>%")
 importFrom(dplyr,group_by_)
 importFrom(dplyr,mutate)

R/preprocess_assay.R

@@ -116,7 +116,19 @@ above_percentage <- function(vector, min_counts, min_percentage) {
   return(isAbove)
 }
 
-filter_chromosome <- function(summarized_experiment, chromosomes = NULL) {
+#' Filter Specific Genes on Chromosome
+#' @description This function filters genes that are located on specific chromosomes.
+#' @param summarized_experiment SummarizedExperiment or matrix like object.
+#' @param chromosome Name of chromosomes to be filtered out | 1-24, X, Y
+#' @param df_biomart Data frame containing biomaRt results
+#' @details Gene names have to be in rows and as ensembl gene IDs. e.g ENSG00000091831. If there is already a data frame with genes and their location on the chromosome, you can use that df instead.
+#' That data frame needs to have the columns "ensembl_gene_id" and "chromosome_name".
+#' @return
+#' @import SummarizedExperiment
+#' @importFrom biomaRt
+#' @export
+
+filter_chromosome <- function(summarized_experiment, chromosome = NULL, df_biomart) {
 
 }
 

R/use_biomaRt.R

@@ -1,13 +1,10 @@
 #' Use biomaRt
 #' @description Short function get information from biomaRt.
-#' @param summarized_experiment SE
-#' @param gene_names Names of genes to be mapped
-#'
-#' @importFrom magrittr %>%
+#' @param gene_names Vector of gene names as ensembl_gene_id
 #' @examples
 #
 #' @return
-
+#' @importFrom biomaRt useMart getBM
 use_biomaRt <- function(gene_names, ... ) {
   ensembl = useMart("ensembl", dataset="hsapiens_gene_ensembl")
   ensg_de_lung <- getBM(mart=ensembl,

example/DAVID_280819.R

@@ -0,0 +1,34 @@
+#DAVID Web Service 28.08
+#http://bioconductor.org/packages/release/bioc/vignettes/RDAVIDWebService/inst/doc/RDavidWS-vignette.pdf
+library("RDAVIDWebService")
+
+david <-DAVIDWebService$new(email="sebastian.lieske@bioinfsys.uni-giessen.de") # does not work
+david <- DAVIDWebService(email="sebastian.lieske@bioinfsys.uni-giessen.de", url="https://david.ncifcrf.gov/webservice/services/DAVIDWebService.DAVIDWebServiceHttpSoap12Endpoint/")
+geneNames <- rownames(subBRCA)
+result <- addList(david, geneNames, idType = "ENSEMBL_GENE_ID", listName = "bimodal", listType = "Gene")
+#getIdTypes(david)
+#do it again with ENTREZ
+ensemblgene <- use_biomaRt(geneNames)
+hgnc <- ensemblgene$hgnc_symbol %>% na.omit() %>% .[.!= ""]
+entrez <- ensemblgene$entrezgene_id %>% na.omit() 
+
+result2 <- addList(david, entrez, idType = "ENTREZ_GENE_ID", listName = "bimodal1", listType = "Gene")
+#adds directly to david
+
+termCluster <- getClusterReport(david, type="Term") #terms or genes
+getClusterReportFile(david, type="Term",
+                      fileName="termClusterBimodalBRCA.tab")
+plot2D(termCluster, 1)
+
+davidGODag<-DAVIDGODag(members(termCluster)[[3]],
+                          pvalueCutoff=0.1, "MF")
+plotGOTermGraph(g=goDag(davidGODag),
+               r=davidGODag, max.nchar=40, node.shape="ellipse")
+
+clust1 <- termCluster@cluster[1]
+clust1df <- data.frame(clust1[[1]]$Members)
+
+silver <- use_silvermantest(subBRCA, 2, filter = F, alpha = 0.05, k = 1)
+silver2 <- use_silvermantest(subBRCA, 2, filter = F, alpha = 0.05, k = 2)
+
+a <- filter_all(ensemblgene, "X" %in% chromosome_name)

example/preprocessing.R

@@ -102,14 +102,5 @@ a <- t(assay(subBRCA, 2)[1:100,1:100])
 b <- dist(scale(a), method = "euclidean")
 c <- hclust(b, method = "ward.D2")
 
-a <- apply(X = assay(subBRCA, 3), MARGIN = 2, FUN = function(x) table(x)) %>% t() #count occurence of each class in patient
-colnames(a) <-  paste("class", colnames(a), sep = "_") #change colnames
-b <- data.frame(a)
-b <- data.frame(a)
-a <- add_coldata_classification(a, 3, to_coldata = F)
-b <- apply(X = a, MARGIN = 2, FUN = function(x) table(x)) %>% t() 
-colnames(b) <- paste("class", colnames(b), sep = "_")
-colnames(countMatrix) <-  paste("class", colnames(b), sep = "_") #change colnames
-
-c <- unnest_dataframe(colData(subBRCA))
-c <- cbind(c, b)
+a <- add_coldata_classification(subBRCA, 3, T)
+b <- unnest_dataframe(colData(a)) %>% group_by()