GENCODE.v38.Exons: GENCODE.v38.GTF <-...

In J0bbie/R2CPCT: Misc. functions to perform import and analysis of WGS and RNA-Seq samples

Description

# Subset on selected genes. data('GENCODE.v38', package = 'R2CPCT') tmp <- tmp[tmp$gene_id

Usage

data(GENCODE.v38.Exons)

Format

GRanges object containing the cleaned-up non-overlapping exonic regions per gene from GENCODE v38.

Details

# Only retain transcripts with at least one non-suspect transcript or are single exons. tmp <- tmp[!grepl('pseudogene|nonsense_mediated_decay|TEC|retained_intron|non_stop_decay', tmp$transcript_type) | tmp$transcript_type == 'polymorphic_pseudogene',]

# Only retain the exon-level elements. tmp <- tmp[tmp$type == 'exon',]

# Remove unused columns. S4Vectors::mcols(tmp) <- S4Vectors::mcols(tmp)[!grepl('remap_|havana_|phase|score|ont|hgnc_id|ccdsid|protein|gene_status|transcript_status', colnames(S4Vectors::mcols(tmp)))]

# Only retain unique exons per gene. tmp.UniqueExons <- GenomicRanges::GRangesList(pbapply::pblapply(base::unique(tmp$gene_id), function(g)

# Get the original (overlapping) exons of multiple transcripts of the gene. x <- tmp[tmp$gene_id == g,]

# Make unique non-overlapping exonic regions. uniqueExons <- GenomicRanges::reduce(x, with.revmap = TRUE)

# Retrieve the identifiers of the original exons. overlappingExons <- base::unlist( base::lapply(uniqueExons$revmap, function(y)

overlappingExons <- base::sprintf('

if(base::length(overlappingExons) > 2) return(base::paste(overlappingExons[1:2], collapse = ', ')) else return(base::paste(overlappingExons, collapse = ', ')) ) )

uniqueExons$overlappingExons <- overlappingExons uniqueExons$SYMBOL <- base::unique(x$gene_name) uniqueExons$ENSEMBL <- base::unique(x$gene_id) uniqueExons$source <- base::paste(base::unique(x$source), collapse = ', ') uniqueExons$type <- 'exon' uniqueExons$totalOverlappingExons <- base::unlist(base::lapply(uniqueExons$revmap, dplyr::n_distinct)) uniqueExons$revmap <- NULL

return(uniqueExons)

, cl = 60))

tmp.UniqueExons <- base::unlist(tmp.UniqueExons)

# Convert columns to best class. GenomeInfoDb::seqlevels(tmp.UniqueExons) <- base::paste0('chr', GenomeInfoDb::seqlevels(tmp.UniqueExons))

# Retrieve the common gene name. tmp.UniqueExons$SYMBOL <- as.character(tmp.UniqueExons$SYMBOL) commonSYMBOL <- limma::alias2SymbolTable(tmp.UniqueExons$SYMBOL) tmp.UniqueExons$SYMBOL <- base::ifelse(is.na(commonSYMBOL), tmp.UniqueExons$SYMBOL, commonSYMBOL)

# Save to package. GENCODE.v38.Exons <- tmp.UniqueExons

usethis::use_data(GENCODE.v38.Exons, overwrite = TRUE)

Source

http://ftp.ebi.ac.uk/pub/databases/gencode/Gencode_human/release_38/GRCh37_mapping/gencode.v38lift37.annotation.gtf.gz

J0bbie/R2CPCT documentation built on Feb. 24, 2022, 8:15 a.m.