Python Programming for Biology: Bioinformatics and Beyond

3

It will be a Voronoi diagram.

4

There is no point taking the square root given that we only want to find the closest.

5

 Note  that  unlike  the  range(a,  b)  function  the  randint(a,  b)  nomenclature  will  give

numbers including b.

6

A completely different k to the one used in these clustering algorithms.

---

7

Sugar, C.A., and James, G.M. (2003). Finding the number of clusters in a data set: an

information theoretic approach. Journal of the American Statistical Association 98:  750–

763.

8

 Covariance  is  discussed  more  fully  in

Chapter  22

.  Here  the  covariance  matrix  has

elements  that  show  the  covariance,  and  hence  scaled  correlation,  between  the  different

dimensions or features of a data set.

9

See

Chapter 9

for details about such matrix manipulations.