Entering edit mode
Hi Hervé Pagès and other maintainers in Bioconductor,
Thanks for your package instruction (
http://140.107.3.20/packages/2.14/bioc/manuals/GenomicAlignments/man/G
enomicAlignments.pdf)
so that I could understand the how does it work. Here you mentioned
the
dumped alignments but I was wondering whether they should be dumped?
seqnames strand cigar qwidth start end
> <rle> <rle> <character> <integer> <integer> <integer>
> SRR031714.2658602 chr2R + 21M384N16M 37 6983850 6984270 ## 2nd
mate/read2
> ...
> SRR031714.2658602 chr2R + 21M384N16M 37 6983850 6984270 ## 2nd
mate/read2
> ...
> SRR031714.2658602 chr2R - 13M372N24M 37 6983858 6984266 ## 1st
mate/read1
> ...
> SRR031714.2658602 chr2R - 13M378N24M 37 6983858 6984272 ## 1st
mate/read1
> ...
For them you mentioned:
pairs could be rec(1) <-> rec(3) and rec(2) <->
> rec(4), or they could be rec(1) <-> rec(4) and rec(2) <-> rec(3).
There is
> no way to disambiguate!
I agree the two pairing strategies above are ambiguous in light of
codes/computation. But it is fact that rec(1) and rec(2) are exactly
same,
thus these two alignments are actually one type of multiple alignment,
though wired but potentially useful. The factor that leads to the
multiple
mappable positions of one mate of the fragment and fixed position of
the
other mate could probably:
a> the reference sequence similarity between these two neighboring
regions
together with different splicing isoforms. Here first mate could have
2
mapped positions. # is exon while = is intron.
2nd mate: ---- ----
1st mate: ---- ----
1st mate: ---- ----
isoform1: ==####======#####==========####==####===
isoform2: ==######====#####==========####==####===
b> or just aligner artifact, etc. Therefore I cannot agree your
package
dumps those alignments. It needs better function name, at least, to
retrieve them.
Regards,
Yun
[[alternative HTML version deleted]]