why HISAT2 and BWA BAM files differ in number of entries?

Question

[marongiu.luigi]

Dear all,

I have aligned the same input fastq files with both BWA and HISAT2 with the commands, respectively:

bwa mem -t 10 <ref> <file_1.fq.gz> <file_2.fq.gz> -o <bwa_aln.sam>
hisat2 -p 10 -q -x <ref> -1 <file_1.fq.gz>  -2 <file_2.fq.gz> -S <hst_aln.sam>

then converted to BAM with:

samtools view -Sb <xxx_aln.sam> > <xxx_aln.bam>

I then compared the header and counted the number of reads in each BAM file with:

$ samtools view -H <bwa_aln.bam>
@SQ SN:21   LN:46709983
@PG ID:bwa  PN:bwa  VN:0.7.17-r1188 CL:bwa mem -t 10 ./ref/GRCh38-21.fa 501N-1_1.fq.gz 501N-1_2.fq.gz -o aln/501N_bwa.sam
$ samtools view -H <hst_aln.bam>
@HD VN:1.0  SO:unsorted
@SQ SN:21   LN:46709983
@PG ID:hisat2   PN:hisat2   VN:2.1.0    CL:"/home/gigiux/src/hisat2/hisat2-align-s --wrapper basic-0 -p 10 -q -x ./ref/GRCh38-21.fa -S aln/501N_hst.sam -1 /tmp/9196.inpipe1 -2 /tmp/9196.inpipe2"

and

$ samtools view -c 501N_bwa.bam 
379141213
$ samtools view -c 501N_hst.bam 
384279994

This is unexpected: shouldn't both files have the same number of reads? Any reasons why BWA shows 5 138 781 reads less than HISAT2?

PS: also BWA's header misses the HD flag...

Thank you

[lakhujanivijay]

In addition to what Wouter suggested, it's good to understand the HISAT alignment summary. Check my earlier post here

[marongiu.luigi]

thank you, this was also interesting

Answer 1

[WouterDeCoster]

Reads can get mapped to multiple locations and will then get counted more than once.

[marongiu.luigi]

fair enough, thank you