View Single Post
Old 04-26-2014, 02:41 PM   #34
Location: Münster, Germany

Join Date: Mar 2013
Posts: 42

Thanks Brian for your response almost at warp speed.

Unfortunately am still a bit lost - probably also due to my ignorance regarding the detail SAM Format Specification. This is, how I assumed the setting takes effect:
  • ambig=all:
    All best alignments are reported. All but one are flagged with the secondary alignment bit (0x100) and the xmtag (an integer ?) additionally reports the number of best alignments for the tuxedo package.
  • ambig=random:
    From the best alignments, one is randomly chosen. This alignment is reported as if there would not have been any secondary alignments. Only the xmtag > 1 indicates, that there would have been other best alignments as well, but the coordinates of those alignments are not present in the output.

I admit, I could easily have checked, whether ambig=random really does not report secondary alignments, but I was pretty confident to have understood the logic - until I read your post this afternoon.

Thanks also for emphasizing the xstag's importance. However the data is indeed unstranded unpaired phred64-age low-coverage dusty archaeological heritage from a previous PhD student .
Thias is offline   Reply With Quote