diff --git a/scripts/11_annotate_hg19_create_pseudo_epicv1_from_epicv2.Rmd b/scripts/11_annotate_hg19_create_pseudo_epicv1_from_epicv2.Rmd index 2c57f79..34b1e51 100644 --- a/scripts/11_annotate_hg19_create_pseudo_epicv1_from_epicv2.Rmd +++ b/scripts/11_annotate_hg19_create_pseudo_epicv1_from_epicv2.Rmd @@ -1,6 +1,6 @@ --- title: "Script 11: Create hg19 annotation for EPICv2 and pseudo-EPICv1 version from EPICv2 data" -date: '`r format(Sys.time(), %d %B, %Y")`' +date: '`r format(Sys.time(), "%d %B, %Y")`' output: html_document --- @@ -27,11 +27,10 @@ start_time <- Sys.time() ```{r setup, include=FALSE} user_choices <- readRDS("../data/user_choices.rds") -knitr::opts_knit$set(root.dir = paste0(user_choices$personal_path, "/")) knitr::opts_chunk$set(echo = FALSE) -if(user_choices$package_directory != "PLEASE FILL IN"){ - .libPaths(c(user_choices$package_directory, .libPaths())) +if(user_choices$package_path != "PLEASE FILL IN"){ + .libPaths(c(user_choices$package_path, .libPaths())) } needed_packages <- c("BiocManager", "dplyr", "knitr", "rmarkdown", "tibble", "minfi", "janitor") @@ -57,6 +56,7 @@ load(paste0(user_choices$project_name, "/reports/annotations_clean_unfiltered.Rd load(paste0(user_choices$project_name, "/processed_data/PhenoData_clean.Rdata")) annotation_hg19 <- read.table(paste0(user_choices$personal_path, "/epic_preprocessing_k2h/data/EPICv2.hg19.manifest.tsv")) annotation_hg19 <- row_to_names(annotation_hg19, row_number = 1, remove_row = TRUE, remove_rows_above = TRUE) +knitr::opts_knit$set(root.dir = paste0(user_choices$personal_path, "/")) ``` ```{r, check array type, include = TRUE} @@ -68,7 +68,7 @@ if (user_choices$array_type == "v1") { ```{r, hg19 annotation filtered data, include = FALSE} annations_clean_filtered_hg19 <- annotation_hg19 %>% - dplyr::filter(Probe_ID %in% rownames(Betas_clean_quantile_bmiq_filtered_combated)) + dplyr::filter(Probe_ID %in% rownames(Betas_clean_filtered_quantile_bmiq_combated)) save(annations_clean_filtered_hg19, file = paste0(user_choices$project_name, "/reports/annations_clean_filtered_hg19.Rdata")) ``` @@ -81,10 +81,10 @@ save(annotions_clean_unfiltered_hg19, file = paste0(user_choices$project_name, " ```{r, create pseudo-EPICv1 version for filtered data, include=FALSE} # create annotation only containing filtered CpGs that are also on v1 and exclude any duplicated loci annotations_clean_filtered_pseudo_v1 <- annotations_clean_filtered %>% - filter(Name %in% rownames(Betas_clean_filtered_quantile_bmiq_combated)) %>% + dplyr::filter(Name %in% rownames(Betas_clean_filtered_quantile_bmiq_combated)) %>% # only keep loci on EPICv1, filter wrongly annotated empty value - filter(EPICv1_Loci != "") %>% - filter(duplicated(EPICv1_Loci) == FALSE) + dplyr::filter(EPICv1_Loci != "") %>% + dplyr::filter(duplicated(EPICv1_Loci) == FALSE) # only keep betas that are on v1 and exclude any duplicated loci (keep first CpG) keep_betas <- (rownames(Betas_clean_filtered_quantile_bmiq_combated) %in% annotations_clean_filtered_pseudo_v1$Name) @@ -103,7 +103,7 @@ save(Betas_clean_filtered_quantile_bmiq_combated_pseudo_v1, file = paste0(user_c cat(paste0(keep_betas_df %>% filter(keep_betas == FALSE) %>% nrow(), " filtered CpGs were not found on EPICv1 and therefore excluded"), sep = "
\n") - dim_Betas_filtered <- dim(Betas_clean_quantile_bmiq_filtered_combated_pseudo_v1) + dim_Betas_filtered <- dim(Betas_clean_filtered_quantile_bmiq_combated_pseudo_v1) step_number <- c("14") step <- c("Create pseudo-EPICv1 version (filtered data)") data_class <- c("Betas") @@ -117,9 +117,9 @@ save(Betas_clean_filtered_quantile_bmiq_combated_pseudo_v1, file = paste0(user_c ```{r, create pseudo-EPICv1 version for unfiltered data, include=FALSE} # create annotation only containing unfiltered CpGs that are also on v1 and exclude any duplicated loci annotations_clean_unfiltered_pseudo_v1 <- annotations_clean_unfiltered %>% - filter(Name %in% rownames(Betas_clean_unfiltered_quantile_bmiq_combated)) %>% - filter(EPICv1_Loci != "") %>% - filter(duplicated(EPICv1_Loci) == FALSE) + dplyr::filter(Name %in% rownames(Betas_clean_unfiltered_quantile_bmiq_combated)) %>% + dplyr::filter(EPICv1_Loci != "") %>% + dplyr::filter(duplicated(EPICv1_Loci) == FALSE) # only keep betas that are on v1 and exclude any duplicated loci (keep first CpG) keep_betas <- (rownames(Betas_clean_unfiltered_quantile_bmiq_combated) %in% annotations_clean_unfiltered_pseudo_v1$Name) @@ -138,7 +138,7 @@ save(Betas_clean_unfiltered_quantile_bmiq_combated_pseudo_v1, file = paste0(user cat(paste0(keep_betas_df %>% filter(keep_betas == FALSE) %>% nrow(), " unfiltered CpGs were not found on EPICv1 and therefore excluded"), sep = "
\n") - dim_Betas_unfiltered <- dim(Betas_clean_quantile_bmiq_unfiltered_combated_pseudo_v1) + dim_Betas_unfiltered <- dim(Betas_clean_unfiltered_quantile_bmiq_combated_pseudo_v1) step_number <- c("15") step <- c("Create pseudo-EPICv1 version (unfiltered data)") data_class <- c("Betas")