Bed Coordinates For Negative/Reverse Strand

Question

[brianrepko]

I understand from the BED "standard" at UCSC that it is a residue, 0-based, half-open system. And all the examples make sense - on the positive strand.

For example - http://alternateallele.blogspot.de/2012/03/genome-coordinate-conventions.html - is a great post about this.

However for the negative strand - which end is the half-open end? Is it always the largest value or is it the end "not-used"?

I found this link - http://genomewiki.ucsc.edu/index.php/Coordinate_Transforms - to be a hot mess as the numbers are reversed - which is not what I want.

If I'm converting a negative strand gene annotation from GTF to BED - is it always GTF(start,end)--> BED(start-1,end) - even for negative strand?

Answer 1

[KCC]

For BED, It's the same regardless of the strand. The first base in the interval is at the start value and the last is end-1. The first base in the genome is 0.

For GTF, it's the same regardless of strand, the first base in the interval is at the start value and the last is end. The first base in the genome is 1.

So, as you say GTF:start,end = BED:start-1,end

Answer 2

[Matt LaFave]

I believe (start-1,end) would be the correct conversion for both strands, since the start and end position of the BED are just read left to right, regardless of the orientation of the feature.