genomic positions on both directions

[carolW]

Hi,
Can a genomic position be on the forward and reverse strand at the same time? If so, which category of positions has this property? How is it possible to view such positions?

Carol

[dpryan]

A position is an abstract concept, so it can have any internally consistent properties you like. If you were to take a restriction enzyme cut site, then a genomic position would normally describe both strandless position.

[carolW]

In the output of a genomic mapper, I have got some positions both on + and - strand. How could they be explained?

[dpryan]

Reads will necessarily have an orientation, where aligning in a reverse complement manner is the same as the - strand. Note that whether this actually means that the original fragment sequenced originally came from that strand depends on how the library was made.

[carolW]

OK but if I get different mapped reads for the same position with differert strand directions, does it mean that the position is on a fragment that coming from both strand directions?

[dpryan]

The reads are from different fragments (unless they're paired-end or duplicates of some sort). Why don't you just say the context of the experiment (RNAseq, DNAseq, something else? What's the goal? What have you done? etc.) so I don't have to guess about what you're doing and write overly vague replies due to that.

[carolW]

every thing has got clear. thanks for your help