Skip to main content
Fig. 4 | BMC Bioinformatics

Fig. 4

From: IntEREst: intron-exon retention estimator

Fig. 4

The effect of sample size and sequencing library size sensitivity. a The number of significantly higher retained introns in ZRSR2mut samples vs controls, relative to the various number of biological replicates. b Similar to the panel A but for the significantly less retained introns in ZRSR2mut samples. c The significantly higher retained introns in ZRSR2mut samples vs controls, relative to the number of reads mapped to the introns and exons. A p adj  < 0.01 threshold was used for all analyses. The data points on the far right in each panel (8 biological replicates in panels a and b; ~60M reads in panel c represent the complete MDS dataset used in the analysis. This leads to zero variance in panels a and b because the resampling size for the complete data is 1 (8 ZRSR2mut vs 8 controls), and less significantly differential IR events compared to the resampled in panel C due to variable size datasets. The conditions of the resampled data in panel C are idealistic, as in these analyses the overall mapped reads for all samples are assumed to be equal (as opposed to in the MDS data where it varies from 51-75 million); hence their number of retained introns are higher compared to the real MDS data

Back to article page