To quantify the amount of variation in DNA methylation explained by genomic context, we considered the correlation between genomic context and principal components (PCs) of methylation levels across all 100 samples (Figure 4). We found that many of the features derived from a CpG site’s genomic context appear to be correlated with the first principal component (PC1). The methylation status of upstream and downstream neighboring CpG sites and a co-localized DNAse I hypersensitive (DHS) site are the most highly correlated features, with Pearson’s correlation r=[0.58,0.59] (P<2.2?10 ?16 ). Ten genomic features have correlation r>0.5 (P<2.2?10 ?16 ) with PC1, including co-localized active TFBSs ELF1 (ETS-related transcription factor 1), MAZ (Myc-associated zinc finger protein), MXI1 (MAX-interacting protein 1) and RUNX3 (Runt-related transcription factor 3), and co-localized histone modification trimethylation of histone H3 at lysine 4 (H3K4me3), suggesting that they may be useful in predicting DNA methylation status (Additional file 1: Figure S3). 67,P<2.2?10 ?16 ) [53,54].
Correlation matrix from forecast have with basic ten Personal computers off methylation account. The new x-axis represents one of several 122 has; the fresh y-axis means Pcs step 1 through 10. Color match Pearson’s relationship, as revealed about legend. Desktop computer, principal role.
Digital methylation reputation prediction
These observations about patterns of DNA methylation suggest that correlation in DNA methylation is local and dependent on genomic context. Using prediction features, including neighboring CpG site methylation levels and features characterizing genomic context, we built a classifier to predict binary DNA methylation status. Status, which we denote using ? we,j ? <0,1>for i ? <1,...,n> samples and j ? <1,...,p> CpG sites, indicates no methylation (0) or complete methylation (1) at CpG site j in sample i. We computed the status of each site from the ? i,j variables: \(\tau _ = \mathbb <1>[\beta _ > 0.5]\) . For each sample, there were 378,677 CpG sites with neighboring CpG sites on the same chromosome, which we used in these analyses.
Therefore, prediction from DNA methylation reputation based only to the methylation membership at neighboring CpG web sites might not work well, particularly in sparsely assayed aspects of the newest genome
New 124 have that individuals employed for DNA methylation condition prediction belong to four more classes (pick Extra document 1: Desk S2 to own a complete listing). For every single CpG webpages, i through the following the ability set:
neighbors: genomic ranges, binary methylation condition ? and you can accounts ? of one upstream and you to definitely downstream nearby CpG webpages (CpG sites assayed on number and you may adjoining about genome)
genomic standing: digital thinking appearing co-localization of the CpG website with DNA series annotations, and additionally promoters, gene human body, intergenic area, CGIs, CGI coastlines and shelves, and you will regional SNPs
DNA sequence properties: carried on values representing nearby recombination rates off HapMap , GC articles of ENCODE , included haplotype score (iHSs) , and you may genomic evolutionary rates profiling (GERP) phone calls
cis-regulatory issue: digital values indicating CpG website co-localization which have cis-regulating aspects (CREs), plus DHS web sites, 79 specific TFBSs, 10 histone amendment scratching and you will fifteen chromatin claims, all of the assayed in the GM12878 mobile line, new nearest meets so you’re able to entire blood
We used a RF classifier, which is an ensemble classifier that builds a collection of bagged decision trees and combines the predictions across all of the trees to produce a single prediction. The output from the RF classifier is the proportion of trees in the fitted forest that classify the test sample as a 1, \(\hat <\beta>_\in [0,1]\) for i=<1,...,n> samples and j=<1,...,p> CpG sites assayed. We thresholded this output to predict the binary methylation status of each CpG site, \(\hat <\tau>_ \in \<0,1\>\) , using a cutoff of 0.5. We quantified the generalization error for each feature set using a modified version of repeated random subsampling (see Materials and methods). In particular, we randomly selected 10,000 CpG sites genome-wide for the training set, and we tested the fitted classifier on all held-out sites in the same sample. We repeated this ten times. We quantified prediction accuracy, specificity, sensitivity (recall), precision (1? false discovery rate), area under the receiver operating characteristic (ROC) curve (AUC), mobilnÃ web ethiopianpersonals and area under the precision–recall curve (AUPR) to evaluate our predictions (see Materials and methods).