Given the regularity with which this problem comes uparound here, I concluded that this problem is used as a sample question for those learning to use perl in the Bioinformatics field.

I have this mental picture of a lecturer somewhere (or several somewheres:), giving the question out as homework, running through the myriad variations of regex and substr solutions that result and then writing 3 lines of perl (including use Bioperl;) to demonstrate the "right way" to do it:).

As problems go, it is a nice, clean, self-contained, real-world problem with a million possible solutions just ideal for teaching purposes.