Skip to main content
Fig. 1 | BMC Bioinformatics

Fig. 1

From: Improving sensitivity of linear regression-based cell type-specific differential expression deconvolution with per-gene vs. global significance threshold

Fig. 1

Differential expression detection sensitivity is primarily affected by two factors: cell type-specific expression estimate (point estimate) variability and cell type-specific differential expression (a). A two-sample t-statistic is computed using the observed effect size (cell type-specific differential expression (Additional file 1: section 1.3). If the t-statistic does not exceed the t-critical value, which is based on the alpha significance threshold, then we cannot conclude that a significant difference has been observed between the two groups. Given an observed difference which is determined to be significant, then we may reject the null hypothesis of no difference between controls and cases and calculate sensitivity for this observed difference, based upon the distance from the case group expression estimate to the t-critical value (b). Bell curves represent distribution of cell type-specific expression estimates (point estimates - vertical dashed lines). The cell type-specific differential expression estimate (effect size) corresponds to the distance between vertical dashed lines for cases and controls (blue/purple bell curves, respectively)

Back to article page