nucleotide extraction using coordinates

Question

[blooming.daisy333]

Dear all,

i have following coordinate file

chr1    329728  329839  -
chr1    330066  330757  -
chr1    581256  581357  +

from this file i want to derive specific nucleotides such that for coordinate 329728 i should get 2 nucleotides of 329728 and 329729 coordinate (i.e one nucleotide from this coordinate along with next nucleotide) and for 329839, 2 nucleotides 329838 and 329839 ( i.e one nucleotide from this coordinate along with one previous nucleotide)

can anyone kindly suggest command/script/tool for this purpose???

kind help will highly be appreciated. many thanks

[WouterDeCoster]

The answer to your question is probably bedtools getfasta

Answer 1

[Bastien Hervé]

I would create a new bed like this :

chr1    329728    329729
chr1    329838    329839
chr1    330066    330067
chr1    330756    330757
...

Then use bedtools getfasta on this new bed file

[finswimmer]

I would create a new bed like this :

$ awk -v OFS="\t" '{print $1,$2,$2+2,$4"\n"$1,$3-1,$3+1,$4}' input.bed > output.bed

Take care that the bed format uses 0-based, half-open intervals.

fin swimmer

[Bastien Hervé]

maybe this would work :

bedtools getfasta -fi genome.fa -bed <(echo awk -v OFS="\t" '{print $1,$2,$2+2,$4"\n"$1,$3-1,$3+1,$4}' input.bed)

[cpad0112]

is it $2,$2+2 or $2-1,$2+1 as user want start, start+1 for $2?

[finswimmer]

As the coordinates in bed files are half-open it should be $2,$2+2. Half-open means the start position is included in the interval but not the end position.

But I don't know whether the OP is aware about the common problem with 0-based vs. 1-based coordinates and half-open vs. closed intervalls. So it might be neccessary do adopt the awk command to get correct coordinates of interest.

fin swimmer

[cpad0112]

keep bed format in view, while creating new bed @OP.

[blooming.daisy333]

this simple method is working well. thanks for king suggestion.