Skip to main content
. Author manuscript; available in PMC: 2009 Nov 20.
Published in final edited form as: Diabetes Obes Metab. 2009 Feb;11(Suppl 1):31–45. doi: 10.1111/j.1463-1326.2008.01001.x

Fig. 3.

Fig. 3

The Bayesian information criteria (BIC) tree with the relative risks and corresponding 95% confidence intervals for each of the terminal leaves (HLA groups) using a neutral group as reference. The horizontal axis can be thought of as an axis of type 1 diabetes risk, and the vertical spacing is proportional to the error in the trees fit. NA is the group that is pure because it consists of pseudo-controls only.