Skip to main content

Table 4 Impact of two marks in MARS model

From: Application of machine learning methods to histone methylation ChIP-Seq data reveals H4R3me2 globally represses gene expression

Bivalent MARS term low-low low-high high-low high-high
H3K27me2-H3K9me1 4.796 4.796 3.866 3.377
H3K27me2-H3K27me3 5.212 3.723 3.677 3.677
H2BK5me1-H3K27me2 5.009 3.677 3.807 3.677
H3K36me3-H4R3me2 3.530 3.468 5.618 4.556
H3K79me2-H4K20me1 3.709 5.211 3.709 4.879
H3K79me1-H3K79me3 3.009 5.938 4.721 5.228
H3K36me3-H4K20me1 3.217 4.163 4.599 7.218
  1. The mean predicted gene expression using high (95th percentile) and low (5th percentile) amplitude values of each mark described in the leftmost column. The permutations of high and low values in each column correspond to the mark order in the leftmost column. The rows are sorted by the values in the last column.