I want to obtain read counts at the exon level using featureCounts. I run Rsubread and use these options:
GTF.featureType="exon", GTF.attrType="exon_id", useMetaFeatures=FALSE
The resulting count matrix has 1,182,163 rows but only 678,276 unique row identifiers, so there are multiple rows for the same exon-id.
Of course when I set useMetaFeatures=TRUE, then the resulting count matrix has 678,276 rows all with unique identifiers.
Sorry for the ignorant question, but why do I have so many more "exons" than "exon-ids"?