Merge pull request #490 from GoekeLab/fix_issue

Fix issue related to gene id with transcript id information and code example

Author: cying111

R/prepareAnnotations_utilityFunctions.R

@@ -25,7 +25,7 @@ prepareAnnotationsFromGTF <- function(file) {
             "score", "strand", "frame", "attribute")
         data <- data[data$type == "exon", ]
         data$strand[data$strand == "."] <- "*"
-        data$GENEID <- gsub("gene_id (.*?);.*", "\\1", data$attribute)
+        data$GENEID <- gsub(".*gene_id (.*?);.*", "\\1", data$attribute)
         data$TXNAME <- gsub(".*transcript_id (.*?);.*", "\\1", data$attribute)
         multiTxCheck <- as_tibble(data) %>% select(seqname, GENEID) %>% distinct() %>% group_by(GENEID) %>% 
             mutate(n=n(), id=paste0('-',row_number()))

README.md

@@ -188,9 +188,9 @@ If you are only interested in the novel transcripts, one can filter this 'se' ob
 se.novel = se[mcols(se)$novelTranscript,]
 writeBambuOutput(se.novel, path = "./bambu/")
 ```
-If you are only interested in full-length transcripts that were detected by Bambu.
+If you are only interested in full-length transcripts that were detected by Bambu in at least 1 sample.
 ```rscript
-se.novel = se[assays(se)$fullLengthCounts >= 1,]
+se.novel = se[mcols(se)$novelTranscript&(apply(assays(se)$fullLengthCounts >= 1,1,sum)>=1),]
 writeBambuOutput(se.novel, path = "./bambu/")
 ```