Fig. 3

DISMISS interrogation of MeDIP-Seq datasets can accurately predict the DNA strand where 5mC occurs. A comparative analysis of 5mC detected by DISMISS interrogation of MeDIP-Seq data with 5mC detected by BS-Seq is illustrated (see the Implementation subsection on data sets for SRA identifiers) a Despite originating from different developmental forms, 5mC detected by DISMISS analysis of MeDIP-Seq datasets and BS-Seq datasets are concordant. Each horizontal bar represents a DISMISS quantified peak class (plus stranded or minus stranded) of MeDIP-Seq data. From all of the DISMISS predicted peak regions in a class, the percentage of 5mCs on the Plus and Minus strands was calculated from the BS-Seq dataset. Each bar shows the percentage of 5mC identified by BS-seq on either plus strands (+, Black) or minus strands (−, Grey). The DISMISS predicted ‘Minus Stranded’ and ‘Plus Stranded’ peaks predominantly contain BS-Seq identified 5mCs on either Minus or Plus strands, respectively. b Loci-specific examples of 5mC detected by DISMISS analysis of MeDIP-Seq datasets, with the corresponding BS-Seq predictions, over three genomic regions. Three randomly selected regions from the A. mellifera genome (assembly NC_007070.3), corresponding to DISMISS predicted peak classes (i, Minus stranded 5mC; ii, Plus stranded 5mC; and iii, 5mC peaks found on both strands) are indicated. The x-axis shows the NC_007070.3 base pair coordinates of each selected region. The first mate read coverage for each base using DISMISS-assessed MeDIP-Seq alignment data is shown along the y-axis. The y-axis is divided into two parts by a central line where the dark-grey coverage above the line represents reads predicted to map to the Plus strand and the light-grey coverage shows reads predicted to map to the Minus strand (see key). The black dots above the line indicate BS-Seq identified 5mCs found on the Plus strand, while those below the line show BS-Seq identified 5mCs found on the Minus strand