Now, the point is, how can you identify those subgroups? Some attempt has been made. For instance, you can use very sophisticated statistical analysis, like, for instance, what has been done in this paper here, where a topological analysis of patient similarity has been performed. Now, this is very complex, believe me. Please don’t ask me to give you an explanation of how this sophisticated statistical analysis works. But what’s important here is that it has been possible, first of all, to identify people with type 2 diabetes and then to group these people according to similarities among them. And when this has been done, what has also been identified were three major groups of people with type 2 diabetes that were very much distinct from one another. And they were distinct not only in terms of phenotype, but also in terms of a number of factors that could be very important in predicting the progression of the disease, the risk of complications and, potentially, the treatment that may be used for those individuals. And here is an example. This is a very busy table, but what I have highlighted for you here are specific characteristics, like differences in age, differences in body weight but, more important, differences in kidney function which suggest that this subgroup may differ in terms of their risk of one of the most dangerous complications of diabetes, which is diabetic nephropathy. But, as I mentioned earlier, this is a very sophisticated analysis. It is very unlikely that we can use it in our everyday clinics.