Table 6. Pairwise average misclassification error rate for the best glmnet model.

	22q	4p	5p	CDL	fraX	MPS2	MPS3	Noon	Pro	PWS	SLO	Sot	TCS
4p	.05
5p	.20	.14
CDL	.05	.00	.09
fraX	.03	.00	.04	.15
MPS2	.10	.11	.18	.04	.00
MPS3	.09	.11	.22	.00	.06	.38
Noon	.11	.28	.14	.07	.00	.11	.05
Pro	.03	.12	.05	.00	.00	.00	.00	.00
PWS	.16	.04	.24	.27	.14	.11	.10	.04	.00
SLO	.05	.11	.16	.06	.00	.10	.18	.04	.05	.11
Sot	.02	.19	.19	.00	.00	.10	.05	.14	.00	.04	.07
TCS	.06	.18	.12	.04	.00	.12	.00	.13	.00	.04	.04	.04
WBS	.06	.06	.09	.08	.04	.08	.08	.02	.00	.09	.12	.00	.02