Ma, Tao

Abstract Motivation Illumina DNA methylation arrays have evolved rapidly, expanding genomic coverage while introducing backward incompatibilities by removing many CpG sites present in earlier versions. These changes result in systematic missing values when integrating data across array generations and substantially limiting the reuse of legacy datasets. Results We developed a two-stage framework for imputing missing DNA methylation values. The procedure first imputes randomly missing values using standard imputation techniques and then addresses systematic missingness using multi-output machine learning models, including support vector regression, nearest-neighbor methods, random forest models, and deep neural networks. When evaluated on real datasets with up to fifty percent induced missingness, the proposed framework consistently outperformed conventional imputation approaches. It also accurately imputes the missing CpG sites between methylation arrays and reduced representation bisulfite sequencing data, enabling robust cross-platform data integration. Analyses of large brain tumor methylation datasets demonstrate that the method restores array-specific methylation patterns while preserving biological complexity. Importantly, imputing missing methylation sites significantly improves the performance of epigenetic age prediction models. Availability and implementation This tool is implemented in the Python package “ultra-impute,” freely available at https://github.com/liguowang/ultra-impute. A code snippet demonstrating the usage of the ultra-impute package is provided in a Jupyter Notebook (https://github.com/liguowang/ultra-impute/blob/master/doc/Tutorial.ipynb).

Obesity is an epidemic with myriad health effects, but little is understood regarding individual obese phenotypes and how they may respond to therapy. Epigenetic changes associated with obesity have been detected in blood, liver, pancreas, and adipose tissues. Previous work using human organoids found that dietary glucose hyperabsorption is a steadfast trait in cultures derived from some obese subjects, but detailed transcriptional or epigenomic features of the intestinal epithelia associated with this persistent phenotype are unknown. This study evaluated differentially expressed genes and relative chromatin accessibility in intestinal organoids established from donors classified as non-obese, obese, or obese hyperabsorptive by body mass index and glucose transport assays. Transcriptomic analysis indicated that obese hyperabsorptive subject organoids have significantly upregulated dietary nutrient absorption transcripts and downregulated type I interferon targets. Chromatin accessibility and transcription factor footprinting predicted that enhanced HNF4G binding may promote the obese hyperabsorption phenotype. Quantitative RT-PCR assessment in organoids representing a larger subject cohort suggested that intestinal epithelial expression of CUBN, GIP, SLC5A11, and SLC2A5 were highly correlated with hyperabsorption. Thus, the obese hyperabsorption phenotype was characterized by transcriptional changes that support increased nutrient uptake by intestinal epithelia, potentially driven by differentially accessible chromatin. Recognizing unique intestinal phenotypes in obesity provides a new perspective in considering therapeutic targets and options with which to manage the disease.