Skip to main content
Fig. 6 | BMC Bioinformatics

Fig. 6

From: Leveraging permutation testing to assess confidence in positive-unlabeled learning applied to high-dimensional biological datasets

Fig. 6

P value from the z-score remains stable regardless of the number of permutations. Statistical comparisons between scores from actual and permuted labels in Lakhashe et al. dataset (A) and a dimension-matched synthetic dataset (B) with varied numbers of permutation repetitions. First row: sample mean and standard deviation (error bar) of the distribution of scores under the permuted labels. Second row: Cliff’s Delta estimates with a 95% confidence interval. Third row: p-value from one-sample z-score. Dashed line: p value = 0.05

Back to article page

BMC Bioinformatics

ISSN: 1471-2105

Contact us