I am new in RNA-seq. I am using Tophat2 to map single end reads to mm9. I am using tophat2 in this way:
tophat -p 10 --max-multihits 1 -G genes_mm9.gff -o output genome_mm9 reads.fastq
With --max-multihits 1, I assume I will get 1 alignment per read. Assuming that, the number of total reads that tophat2 uses for the mapping (19196075) should be the number of alignments in accepted_hits.bam file (8797938) (because --max-multihits 1) plus the total reads in unmapped.bam file (7538885). But that is not the case, there are 2859252 reads missing. Am I correct?
Thank you for your help