#37 · Epigenomics

Computational metrics quantify resistance, foreign-genotype cells and CpG methylation

The dominant thread across these items is deriving defined quantitative readouts from sequencing-based cell or read data. BARMIX pairs barcoded GIST cell mixtures with a probabilistic model to quantify genotype-specific drug resistance in vitro and in vivo. Cellector detects rare foreign-genotype cells in scRNAseq data, applied to leukemia post-transplant residual disease and microchimerism. MPCI introduces a metric quantifying consistent methylated and unmethylated CpG blocks across WGBS reads, reported to outperform MHL and dMHL on cell-type and tissue comparisons. What is new in these items is explicit metric definition at the level of individual cells, genotypes, or sequencing reads. The set is heterogeneous: only MPCI addresses DNA methylation directly, and no item integrates multiple modalities within a single framework.