How to interprete the positive strand(+) and negative strand(-) in the result

[Mason]

Dear JASPAR team, I am a loyal user of JASPAR website which solves plenty of problems in the prediction of TF. However, there is a confusion in my mind is after I predict a possible binding site with TF and an input sequence, the result shows binding sites both in  positive strand(+) and negative strand(-), how to interprete them in the result? What's the biological effect of the binding between TF and negative strand since they are contrast with the direction of transcription? Looking forward to your answer and thanks a lot! 

[Jaime Castro]

Hi Mason

The strand corresponds to the DNA orientation.

An example with PITX1: http://jaspar.genereg.net/matrix/MA0682.2/?revcomp=1 

We clearly see that it recognizes the consensus sequence GGATTA, when we scan this motif in a sequence, for example: gctatcgactCCTAATgtcatctactacg

Here there are no hits of the motif in the example sequence.

However, we can reverse-complement the DNA sequence or the motif, and re-scan the sequence,

Original sequence: 5'-gctatcgactTAATCCgtcatctactacg-3' 

                      |||||||||||||||||||||||||||||
Reverse:           3'-cgatagctgaATTAGGcagtagatgatgc-5' ---> Reverse complement: 5'-cgtagtagatgacGGATTAagtcgatagc-3'

As you can see in the example above,now we found a hit of the motif  (GGATTA is the reverse-complement of TAATCC).

In this case, since we reverse-complemented the DNA sequence to find the hit, we will annotate the strand of GGATTA as (-).

If we would find a motif hit without reverse-complementing the DNA, then the Strand would be (+).

When we scan a TF motif, generally we scan both DNA strands, only if you want to scan RNA motifs, here you can limit to scan only one strand. 

When you reverse complement a sequence you are sure that you maintain the direction of transcription.

Jaime