How can DNA profiling be used to determine the number of organisms from a specific species that is within a certain area?

Question

Here is the tedious exam question that asks this: question 8(b). https://drive.google.com/drive/folders/0B23NLD5L6099VWJya3dfLVJpbmc

The question states:

In 2007, the brown bear was a threatened species in Alberta, Canada. A study was carried out to determine the number of bears in the area. Fur wascollected from trees that the bears had rubbed against and their DNA analysed.

(i) Explain how DNA analysis would help the scientists to determine the number of bears in the area.

The answer simply states:

1. idea of comparing bands

2. idea that each bear has unique {DNA / banding patterns}

3. idea that the number of different banding patterns would equate to the number of different bears

My question is: How does the entire process fit into together?

I am absolutely baffled as to how the entire investigation would fit into each other, and I've been trying to articulate it for hours. It seems like if I'm just going to be collecting fur samples off trees there are going to be cells belonging to different species, then the DNA pattern produced will include a different number of bands from all of the different species. To overcome this problem, I think 'would we somehow separate the different cells into different groups with each group belonging to a specific bear by some technique that isn't DNA profiling and I don't need to know about at high school level? I then reason that we would have accomplished our task of identifying the number of different number of bears present within the sample. How would one identify the different numbers of bears?

The second part of my question is that I don't understand how the different samples from trees and fur will be used in order to estimate the overall population size within the area? The only animal sampling technique that I know of is that capture-recapture method.

Answer 1

The general process of DNA profiling (I prefer calling it "DNA fingerprinting", but that's just me) generally involves developing PCR primers specific for a particular region surrounding some sort of highly polymorphic region. So in this case one would need primers surrounding VNTRs (or other polymorphic regions, have a look at the wikipedia article for an overview of some other methods). Presuming the PCR primers are specific for bears (one could blast them to get an idea of this), then regardless of what other animals were mixed in, the resulting bands would predominantly be from bears (after all, PCR results in an exponential amplification of favored species). This would be answer part (1).

For part 2, one needs to find a sufficient number of sufficiently variable regions for profiling to work. A single site might not be informative enough, but a couple of them together will be.

Part 3 is the logical outcome of parts 1 and 2. If you can amplify a bunch of mixed samples together and the repeats are of variable length (thus the use of VNTRs or similar repeats), then you'll see those as different sized bands on a gel.