recombination_predictions.gff regions are overlapping and duplicated ?

[tseemann]

cut -f4,5 core.full.aln.recombination_predictions.gff

##gff-version 3
##sequence-region SEQUENCE 1 3078339
3076391 3077947
2965179 2966076
2965516 2965695
2388306 2388384
258711  258716
1017165 1017176
2388306 2388384
1017165 1017176
2834125 2848235
821016  821123
1658438 1658554
2750811 2751010
686111  744010
2282185 2300157
2120097 2125052
686433  743972
2282206 2290657
2120211 2126098
618063  618587

If I sort them:

cut -f4,5 core.full.aln.recombination_predictions.gff | sort -k1,1n

##gff-version 3
##sequence-region SEQUENCE 1 3078339
258711  258716  # 5 bp
618063  618587
686111  744010
686433  743972 # inside previous one
821016  821123
1017165 1017176   # 11 bp
1017165 1017176  # DUPE
1658438 1658554
2120097 2125052
2120211 2126098 # overlap
2282185 2300157
2282206 2290657
2388306 2388384
2388306 2388384
2750811 2751010
2834125 2848235
2965179 2966076
2965516 2965695
3076391 3077947

And then merge them:

bedtools sort -i core.full.aln.recombination_predictions.gff | bedtools merge

SEQUENCE        258710  258716
SEQUENCE        618062  618587
SEQUENCE        686110  744010
SEQUENCE        821015  821123
SEQUENCE        1017164 1017176
SEQUENCE        1658437 1658554
SEQUENCE        2120096 2126098
SEQUENCE        2282184 2300157
SEQUENCE        2388305 2388384
SEQUENCE        2750810 2751010
SEQUENCE        2834124 2848235
SEQUENCE        2965178 2966076
SEQUENCE        3076390 3077947

[andrewjpage]

Recombinations can be detected at different internal nodes in the tree, so can be overlapping.

[tseemann]

I just got home and realised I'm an idiot 😁
Still should sort it though.