gives the expected frequency of draws of two blue and two clear tubes when sampling four tubes. This expected value is very close to what was observed in 10 draws of four tubes in Table 3.1.

The binomial formula can be used to calculate the expected probability of observing each of the possible outcomes when drawing samples of 2N = 4 and 2N = 20 micro-centrifuge tubes from beaker populations. These probability distributions (Fig. 3.5) summarize what we would expect to find if we drew many independent samples and then tabulated the results. The probability for each bar in the histograms of Fig. 3.5 was determined using the binomial formula. For example, the expected frequency of sampling 12 blue tubes in a total sample of 20 tubes is


These probability distributions explain why a fixation/ loss event was observed when 2N = 4 but not for 2N = 20, since the former outcome is expected in one out of 16 draws but the latter only once in 1,048,576 draws. With knowledge of the binomial probability distribution, the Wright-Fisher model of genetic drift (Fig. 3.2) makes a lot of sense. It was constructed, in fact, to articulate the assumptions that underlie the use of the binomial formula and binomial probability distributions to model genetic drift.

0 0

Post a comment