Fig. 1: Workflow for assessing the impact of sample size on discovery metrics. | Nature Communications

Fig. 1: Workflow for assessing the impact of sample size on discovery metrics.

From: Optimized murine sample sizes for RNA sequencing studies revealed from large scale comparative analysis

Fig. 1

We define the ‘gold standard’ signature as the set of DEGs in the full cohort of mice (30 Dchs1 Het vs 30 WT). From this full cohort, we generated a smaller “mini-experiment” (trial) by randomly selecting N Het and N WT mice (Down-sampling strategy, top middle). The DEG from this trial were compared to the gold standard signature to assess sensitivity, specificity, and effect size bias. The number of mice selected (N) range from 3 to 29, and 40 trials were performed for each N. A separate approach (disjoint sets) was used to assess consistency between pairs of mini-experiments. Here, two mini-experiments of size N are created in each trial, and the resulting two DEG sets are compared. Higher consistency between DEGs is interpreted as each DEG capturing more biological signal.

Back to article page