Skip to main content
. Author manuscript; available in PMC: 2018 Sep 1.
Published in final edited form as: Hum Mutat. 2017 Jun 28;38(9):1225–1234. doi: 10.1002/humu.23256

Table 1.

Performance of the six CAGI4 submissions

Method description GWAS loci used AUC 95% CI
Naïve Bayes model (conditional probability based) 138 0.72 [0.62–0.82]
Consensus machine learning method from Naïve Bayes (odds ratio based), Logistic regression and Random Forest (1000 trees) 138 0.72 [0.63–0.81]
Logistic regression 138 0.71 [0.61–0.80]
Logistic regression 90 0.66 [0.54–0.75]
Consensus machine learning method from Naïve Bayes (odds ratio based), Logistic regression and Random Forest (1000 trees) 90 0.65 [0.55–0.75]
Rare predicted high impact missense variant contribution together with common variant contribution in a Naïve Bayes model (odds ratio based) 90 0.63 [0.54–0.73]