BWA: improper orientation but listed as properly paired

Question

[richard.deborja]

I have a set of BAM files for a targeted sequencing experiment. After reviewing things in IGV, I noticed that a pair of reads listed as orientation F2R1 are labelled as properly paired. I also noticed another pair listed as F1R2 that were also labelled as properly paired. From my understanding of paired-end sequencing, F1R2 is the correct orientation and F2R1 would be more consistent for mate-pair. The version of BWA used was 0.6.2 and Picard 1.107 for post-processing (conversion to BAM, sorting, indexing, read group addition). I confirmed alignment to the reference genome using BLAT and identified the reads in the original FASTQ files for read 1 and read 2. Any idea why it would list both orientations as properly paired?

I'm including the SAM entries below with modified read IDS:

17091    163    chrX    48649459    60    151M    =    48649634    326    AAGGATTTCTGTGTCTGAGGACCCCTTCTGTCCTCGCAGGTTAATCCCCAGAGGCTCCAGGGAGTTCCCTGGCCTGGGGTCCCTGGGGACCTCAGAGCCCCTCCCCCAGTTTGTGGATCCTGCTCTGGTGTCCTCCACACCAGAATCAGGG    @CCFFFFFHHHHHJJJJJJJJJJJJJJJJJJJJJJJJJJJJJJJJJJJJJIJJJJJJJJJJHHHFFFFFECCEEDBBDDDDDDDDDDDDDDCDDDC@CDDDDDDDDDDDDDED?CBDDDDDCCCDDDDCDDDDDDDDDDDDDB?CADCCCD    X0:i:1    X1:i:0    MD:Z:59T91    RG:Z:8932118E-CCA7-11E3-9394-35788F63E160    XG:i:0    AM:i:37    NM:i:1    SM:i:37    XM:i:1    XO:i:0    XT:A:U
49588    99    chrX    48649459    60    151M    =    48649532    224    AAGGATTTCTGTGTCTGAGGACCCCTTCTGTCCTCGCAGGTTAATCCCCAGAGGCTCCAGGGAGTTCCCTGGCCTGGGGTCCCTGGGGACCTCAGAGCCCCTCCCCCAGTTTGTGGATCCTGCTCTGGTGTCCTCCACACCAGAATCAGGG    CCCFFFFFHHHHHJJJJJJJJJJJJJJJJJJJJJJJJJJJJJJJJJJJJJHJJJHHJJJJJHHHFFFFFFEDEEDDDDDBDDDDDDDDDDDDDDDDCCDDDDDDDDDBDDDEDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDBDDDDDDD8    X0:i:1    X1:i:0    MD:Z:59T91    RG:Z:8932118E-CCA7-11E3-9394-35788F63E160    XG:i:0    AM:i:37    NM:i:1    SM:i:37    XM:i:1    XO:i:0    XT:A:U
49588    147    chrX    48649532    60    151M    =    48649459    -224    CTGGGGTCCCTGGGGACCTCAGAGCCCCTCCCCCAGTTTGTGGATCCTGCTCTGGTGTCCTCCACACCAGAATCAGGGGTTTTCTTCCCCTCTGGGCCTGAGGGCTTGGATGCAGCAGCTTCCTCCACTGCCCCGAGCACAGCCACCGCTG    DDDDDDDDDDDDDDCCDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDBBADDDDDDDDDDDDCADEDEDDDDDDDDDDDDDDDDDDDDDDDDFFFFHHHHJJIJJJJJJJJJJJJIJJJJJJJIIJJJJJJJJJJJHHHHHFFFFFCCC    X0:i:1    X1:i:0    MD:Z:151    RG:Z:8932118E-CCA7-11E3-9394-35788F63E160    XG:i:0    AM:i:37    NM:i:0    SM:i:37    XM:i:0    XO:i:0    XT:A:U
17091    83    chrX    48649634    60    151M    =    48649459    -326    GGCTTGGATGCAGCAGCTTCCTCCACTGCCCCGAGCACAGCCACCGCTGCAGCTGCGGCACTGGCCTACTACAGGGACGCTGAGGCCTACAGACACTCCCCAGGTAACTCCATTGAGTGGCTGTCTTGGCATTGGCTGAGTGCTGTTGGGG    8DDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDBBDDDDDDDDDDDDDDDDDDDDDDEFEDDDDDDDDDDEFFFFFHHHHHHJIIHHJJIJJJJJJJIJJJJJJJJJJJJJJJJJJJJJJJJJJJJJJJJHHHHHFFFFFCCC    X0:i:1    X1:i:0    MD:Z:151    RG:Z:8932118E-CCA7-11E3-9394-35788F63E160    XG:i:0    AM:i:37    NM:i:0    SM:i:37    XM:i:0    XO:i:0    XT:A:U

[Pierre Lindenbaum]

Hello richard.deborja!

It appears that your post has been cross-posted to another site: http://seqanswers.com/forums/showthread.php?t=44081

This is typically not recommended as it runs the risk of annoying people in both communities.

[richard.deborja]

Hello richard.deborja!

It appears that your post has been cross-posted to another site: Apparently cross posting is a bad idea, since I'm unable to delete/close the seqanswers post, closing this thread.

This is typically not recommended as it runs the risk of annoying people in both communities.

[Pierre Lindenbaum]

don't close the thread. That's ok. But in the future avoid to post on both sites within a short period or without linking to the cross-posts.

Answer 1

[Devon Ryan]

The reads are indeed properly paired. In one case you have:

read1---->
             <---read2

and in the other

read2---->
             <---read1

In both cases, the reads are pointing toward each other, meaning that they're in the appropriate orientation.

[richard.deborja]

If sequencing is done 5' to 3', wouldn't the orientation be:

<--- read1
                  read2--->

[Devon Ryan]

In the cases where read1 is reverse complemented, it's 3' of read2.