In summary, Bismark, BSMAP, and BS-seeker2 performed differently in different read conditions.
Furthermore, the integration reduced the dependency of detection accuracy on read conditions (i.e. error rate and length), proving that our method can facilitate the comprehensive analyses of multiple WGBS samples of which read conditions are heterogeneous.
The tendency was clearly revealed within short reads containing low error, so the d-accuracy of pwAve was the highest among the integration methods in those read conditions.
In summary, our integrative approach improved both quality and quantity of methylation results from WGBS data, and facilitated the comprehensive analyses of DNA methylation among various read conditions.
Thus, our method succeeded in facilitating comprehensive analyses of multiple WGBS datasets from various experiments by reducing the dependency of methylation results on read conditions.
The Wilcoxon single-rank test over 500 bps blocks revealed significantly low P value (≤5.0E-2) in most of read conditions (Additional file 3: Tables S1-S4).
To investigate the performances of the three mappers under diverse read conditions, we analyzed the mapping results of the three mappers in the context of varying read quality, read length, and methylation levels.
Head tilt and angle of gaze were altered by reading condition and font size.
The study design was an efficacy trial with two thirds of families randomly assigned to the dialogic reading condition and one third to a comparison condition.
A repeated-measures design was used to compare data obtained during a control reading condition and during rhythmic reading in 10 persons who stutter (PWS) and 10 normally fluent controls.
In a first analysis, we compared pathways for each reading condition thus revealing which pathways are more 'engaged' than others.
