Help with summarizeOverlaps function in RNASeq analysis using R

Question

[FJCF]

Hi,

I am a new bioinformatics MSc student and I need help with the summarizeOverlaps function from the GenomicAlignments R package in order to perform a RNASeq analysis. I keep getting this error:

Error in array(x, c(length(x), 1L), if (!is.null(names(x))) list(names(x),  : 
  'data' must be of a vector type, was 'NULL'*.

Could someone help me to find where the problem is and how to fix it?

pacman::p_load(Rsamtools, GenomicFeatures, GenomicAlignments, org.Mm.eg.db)

gtfFile <- "genes.gtf"
txdb <- makeTxDbFromGFF(gtfFile, format = "gtf", organism = "Mus musculus")
genes <- exonsBy(txdb, by = "gene")
files <- list.files("/BAM", pattern = ".bam")
bamLst <- BamFileList(files, index=character(), obeyQname = TRUE)

PRJNA838600 <- summarizeOverlaps(features = genes,
                                 read = bamLst,
                                 mode = "Union",
                                 singleEnd = FALSE,
                                 ignore.strand = TRUE,
                                 fragments = FALSE)

save(PRJNA838600, file = "PRJNA838600.rda")

[BioinfGuru]

1) Which line of code produces the error?

2) Can you check the contents of your objects and post the results here?

head(c(txdb, genes, files, bamLst))

[Ram]

Why did you delete the post?

[BioinfGuru]

me? what post?

[GenoMax]

That comment from @Ram was directed at original poster. They probably had deleted this post after having received comment/answer.

[BioinfGuru]

"Thumbs up"

Answer 1

[Soheil]

I'm guessing you are trying to get the gene expression counts from your BAM files. In this case, I suggest using featureCounts from the subread toolkit to quantify gene expression from BAM files generated with STAR (ideally the reads should be QC'ed/Trimmed). This is pretty much the community standard to process raw RNAseq fastq files.

[BioinfGuru]

The summarizeOverlaps count modes are fashioned after the "Union", "IntersectionStrict", and "IntersectionNotEmpty" methods found in HTSeq