Hello,

I have RNA-seq reads from Nanopore(kit RNA002, U-based fastq), I aligned them with minimap2 and tried to count with featureCounts. The reference was transcriptome from https://www.gencodegenes.org/human/. my command was

 fc<- Rsubread::featureCounts(files="aln.bam",annot.inbuilt="hg38",isLongRead="True")

 Process BAM file aln.bam...                                                ||
||    Single-end reads are included.                                          ||
||    Total alignments : 2081002                                              ||
||    **Successfully assigned alignments : 0 (0.0%)**                             ||
||    Running time : 0.26 minutes   
Warning message:
In paste("readSummary", ann, files_C, fout, as.numeric(isPairedEnd),  :
  NAs introduced by coercion

I checked my bam file and have such a statistics:

SN  raw total sequences:    632940
SN  filtered sequences: 0
SN  sequences:  632940
SN  is sorted:  1
SN  1st fragments:  632940
SN  last fragments: 0
SN  reads mapped:   609258
SN  reads mapped and paired:    0   # paired-end technology bit set + both mates mapped
SN  reads unmapped: 23682

What could it be?

A lot of thanks.

I don't have experience with minimap2, but a quick read of lh3's GitHub page makes me think it's designed for aligning to the genome. Yet you say you aligned to the transcriptome. If you did what you say you did, then featureCounts is probably not the tool for the job, as it is intended for counting reads that are aligned to a genome.

If you did in fact align to the genome, do note that the in-built annotation is based on NCBI Gene IDs, and GenCode is based on EBI/EMBL, so the chromosome in the first instance will be e.g., chr1, whereas in the second it will be just 1. You will probably either need to generate an SAF file, or get a GFF or GTF file from GenCode. See the Details section of ?featureCounts.