identifying strings within intervals, pt 2 - awk

Question

For each row, I would like to know if the numerical string in the 6th column resides within the start and end intervals of the 3rd and 4th column. The issue for me is that identical strings in the 1st and 5th column are not always in the same row (e.g., uce-6459 is on the same line as uce-432).
Input
locus match start end subset pos region
uce-3280 uce-3280_p1 269 388 uce-3280 222
uce-6620 uce-6620_p1 297 416 uce-6620 198
uce-6620 uce-6620_p1 297 416 uce-6620 300
uce-432 uce-432_p2 328 447 uce-432 205
uce-432 uce-432_p1 268 387 uce-6459 207
uce-6459 uce-6459_p1 210 329 uce-6459 275
uce-6459 uce-6459_p1 210 329 uce-6459 288
uce-6459 uce-6459_p1 210 329 uce-374 373
uce-374 uce-374_p2 509 628 uce-3393 327
uce-374 uce-374_p1 449 568 uce-3393 416
uce-3393 uce-3393_p1 439 558 uce-3393 712
uce-3393 uce-3393_p1 439 558 uce-1200 416
uce-3393 uce-3393_p1 439 558 uce-805 397
uce-1200 uce-1200_p3 341 460 uce-627 326
uce-805 uce-805_p1 333 452 uce-2299 340
uce-627 uce-627_p1 396 515 uce-2126 481
uce-2299 uce-2299_p1 388 507 uce-5427 562
uce-2126 uce-2126_p1 323 437 uce-5427 711
uce-5427 uce-5427_p1 509 628 uce-5893 242
uce-5427 uce-5427_p1 509 628 uce-5893 330
uce-5893 uce-5893_p1 477 582 uce-5893 398
Desired output
locus match start end subset pos region
uce-3280 uce-3280_p1 269 388 uce-3280 222 no
uce-6620 uce-6620_p1 297 416 uce-6620 198 no
uce-6620 uce-6620_p1 297 416 uce-6620 300 yes
uce-432 uce-432_p2 328 447 uce-432 205 no
uce-432 uce-432_p1 268 387 uce-6459 207 no
uce-6459 uce-6459_p1 210 329 uce-6459 275 yes
uce-6459 uce-6459_p1 210 329 uce-6459 288 yes
uce-6459 uce-6459_p1 210 329 uce-374 373 no
uce-374 uce-374_p2 509 628 uce-3393 327 no
uce-374 uce-374_p1 449 568 uce-3393 416 no
uce-3393 uce-3393_p1 439 558 uce-3393 712 no
uce-3393 uce-3393_p1 439 558 uce-1200 416 yes
uce-3393 uce-3393_p1 439 558 uce-805 397 yes
uce-1200 uce-1200_p3 341 460 uce-627 326 no
uce-805 uce-805_p1 333 452 uce-2299 340 no
uce-627 uce-627_p1 396 515 uce-2126 481 no
uce-2299 uce-2299_p1 388 507 uce-5427 562 yes
uce-2126 uce-2126_p1 323 437 uce-5427 711 no
uce-5427 uce-5427_p1 509 628 uce-5893 242 no
uce-5427 uce-5427_p1 509 628 uce-5893 330 no
uce-5893 uce-5893_p1 477 582 uce-5893 398 no
Any help would be appreciated.

Answer 1

Here is a full stand-alone awk-script:
#!/usr/bin/awk -f
# goes through the whole file, saves boundaries of the intervals
NR > 1 && NR == FNR {
starts[$1] = $3
ends[$1] = $4
#print "Scanned interval: "$1" = ["starts[$1]","ends[$1]"]"
}
# prints the header of the table
NR != FNR && FNR == 1 {
print $0
}
# annotates each line with "yes"/"no"
FNR > 1 && NR != FNR {
print $0" "(starts[$5] <= $6 && $6 <= ends[$5] ? "yes" : "no")
}
Depending on what OS you have and what awk you are using, you might need to adjust the first line (use which awk to find the right path).
In order to make it run, you have to save it in a separate file (e.g. analyzeSnp.awk), then make it executable (e.g. chmod u+x analyzeSnp.awk), and then run it like this:
./analyzeSnp.awk inputData inputData
Alternatively, you can enclose it in single quotes and paste it directly into the console like this:
$ awk ' #!/usr/bin/awk -f
# goes through the whole file, saves boundaries of the intervals
NR > 1 && NR == FNR {
starts[$1] = $3
ends[$1] = $4
#print "Scanned interval: "$1" = ["starts[$1]","ends[$1]"]"
}
# prints the header of the table
NR != FNR && FNR == 1 {
print $0
}
# annotates each line with "yes"/"no"
FNR > 1 && NR != FNR {
print $0" "(starts[$5] <= $6 && $6 <= ends[$5] ? "yes" : "no")
}' loci.txt loci.txt
Important: you have to specify your input file twice, because we need a first scan to build a table with intervals, and then a second scan for actual annotation.
Here is the produced output:
locus match start end subset pos region
uce-3280 uce-3280_p1 269 388 uce-3280 222 no
uce-6620 uce-6620_p1 297 416 uce-6620 198 no
uce-6620 uce-6620_p1 297 416 uce-6620 300 yes
uce-432 uce-432_p2 328 447 uce-432 205 no
uce-432 uce-432_p1 268 387 uce-6459 207 no
uce-6459 uce-6459_p1 210 329 uce-6459 275 yes
uce-6459 uce-6459_p1 210 329 uce-6459 288 yes
uce-6459 uce-6459_p1 210 329 uce-374 373 no
uce-374 uce-374_p2 509 628 uce-3393 327 no
uce-374 uce-374_p1 449 568 uce-3393 416 no
uce-3393 uce-3393_p1 439 558 uce-3393 712 no
uce-3393 uce-3393_p1 439 558 uce-1200 416 yes
uce-3393 uce-3393_p1 439 558 uce-805 397 yes
uce-1200 uce-1200_p3 341 460 uce-627 326 no
uce-805 uce-805_p1 333 452 uce-2299 340 no
uce-627 uce-627_p1 396 515 uce-2126 481 no
uce-2299 uce-2299_p1 388 507 uce-5427 562 yes
uce-2126 uce-2126_p1 323 437 uce-5427 711 no
uce-5427 uce-5427_p1 509 628 uce-5893 242 no
uce-5427 uce-5427_p1 509 628 uce-5893 330 no
uce-5893 uce-5893_p1 477 582 uce-5893 398 no
This looks pretty close to your example snippet.
A general remark: AWK is a surprisingly modern and powerful scripting language (at least for it's age: this thing is almost 40 years old!). If you want to use AWK, you are not constrained to cryptic single-liners.