Andrey A. Mironov

The paper presents a machine learning framework to predict RNA-protein interactions using eCLIP data and shows its limitations in extrapolating to proteins without such data.

The paper compares RNA-chromatin interactome data from two experimental approaches to assess contact completeness and specificity using an overrepresentation metric.

The paper jointly analyzes RNA-DNA and DNA-DNA interactomes across human and mouse cell lines, showing their close relationship is tissue-specific and conserved.

The paper introduces a triad interaction framework to reduce noise in RNA-chromatin interaction data and validate interactions using experimental controls.

The paper integrates HiMoRNA and RNA-Chrom resources to generate hypotheses about lncRNA epigenetic regulation mechanisms using experimentally supported RNA-chromatin interaction data.

BaRDIC improves peak calling in RNA-DNA interaction data by addressing chromatin heterogeneity and distance-dependent decay, outperforming existing methods.

The paper evaluates the completeness and accuracy of RNA-chromatin interactions data from human and mouse, finding that 'all-to-all' experiments are incomplete and require deeper sequencing.

The paper introduces a spike-in method to improve allele-specific expression analysis by reducing technical noise and cost.

ortho2align identifies orthologues of novel lncRNAs across species by leveraging synteny and sequence conservation, enabling prediction of unannotated orthologous sequences.

The paper introduces Red-C, a method to study RNA-DNA interactions, identifies ncRNAs associated with repressed chromatin, and traces mRNA production kinetics in K562 cells.

The paper assesses allelic imbalance variability in RNA sequencing and finds technical replicates show up to 7-fold higher variation than expected, which can be mitigated to improve differential analysis.

StereoGene estimates genomewide correlation of continuous genomic features, revealing spatial relationships between epigenomic features and splicing sites.