Compare Bowtie2 and BWA summary

[Navy1991]

How to compare bowtie and bwa summary?

Here is the output of bowtie2:
28115453 reads; of these:
28115453 (100.00%) were paired; of these:
9177458 (32.64%) aligned concordantly 0 times
15185809 (54.01%) aligned concordantly exactly 1 time
3752186 (13.35%) aligned concordantly >1 times
----
9177458 pairs aligned concordantly 0 times; of these:
2358270 (25.70%) aligned discordantly 1 time
----
6819188 pairs aligned 0 times concordantly or discordantly; of these:
13638376 mates make up the pairs; of these:
9978644 (73.17%) aligned 0 times
1757628 (12.89%) aligned exactly 1 time
1902104 (13.95%) aligned >1 times
82.25% overall alignment rate

How to compare the information between BWA and bowtie2 for the same file.
I know that

For multiples hits with BWA, you have XT:A:R flag (R for multiple, you can have XT:A:U for unique reads. For paired-end reads, you might also consider XT:A:M (one-mate recovered) which means that one of the pairs is uniquely mapped and the other isn't)
For Bowtie2, XS:i flag determined if alignment is unique or not. Only present if the SAM record is for an aligned read and more than one alignment was found for the read.

I am using paired end sequencing. How to compare and verify if the summary from both bowtie2 and bwa is same

[dpryan]

Please forgot the entire concept of a "unique alignment"; it's misleading.

A proper comparison would be to look at concordance/discordance as a function of MAPQ. See this thread, starting around comment 8 for an example. You could also look at how often paired reads were aligned as singletons ("mate-recovered" in bwa parlance) in one aligner and not the other.