Zoia G. Zaitseva

Super-enhancers (SEs) are large clusters of highly active enhancers that play key regulatory roles in cell identity, development, and disease. While conventional methods classify SEs in a binary fashion—super-enhancer or not—this threshold-based approach can overlook significant intermediate states of enhancer activity. Here, we present a dataset and accompanying framework that facilitate a more nuanced, non-binary examination of SE activation across mouse tissue types (mammary gland, lung tissue, and NMuMG cells) and various experimental conditions (normal, tumor, and drug-treated samples). By consolidating overlapping SE intervals and capturing continuous enhancer activity metrics (e.g., ChIP-seq signal intensities), our dataset reveals gradual transitions between moderate and high enhancer activity levels that are not captured by strictly binary classification. Additionally, the data include extensive functional annotations, linking SE loci to nearby genes and enabling immediate downstream analyses such as clustering and gene ontology enrichment. The flexible approach supports broader investigations of enhancer landscapes, offering a comprehensive platform for understanding how SE activation underpins disease mechanisms, therapeutic response, and developmental processes.

Super-enhancers (SEs) orchestrate high-level transcription by integrating multiple regulatory elements and signals. Although chromatin accessibility and transcription factor binding within SEs are extensively studied, the role of non-canonical DNA structures, particularly Z-DNA, remains underexplored. In this study, genome-wide predictions of Z-DNA-forming sequences (generated by the Z-DNA-BERT model) were applied to systematically investigate their distribution within typical enhancers and SEs across multiple human cancer cell lines. Statistically significant enrichment of Z-DNA sequences within SE regions, compared to random genomic controls, was observed. Furthermore, genetic variants overlapping these Z-DNA regions, identified using data from the 1000 Genomes Project, were found to alter binding motifs of the SP/KLF transcription factor family. These mutations exhibited population-specific clustering and overlapped previously reported pathogenic copy-number variations (CNVs) associated with neurodevelopmental disorders, potentially affecting transcription factor binding motifs related to neuronal growth and differentiation pathways. Population-level phylogenetic analysis revealed distinct clustering patterns of these variants, suggesting frequency-specific genetic architecture. Overall, the computational findings indicate that Z-DNA structures within super-enhancers might play regulatory roles and potentially influence population-specific genetic variation, highlighting specific genomic targets and providing new avenues for future experimental research.

The study discloses the application of transformer-based deep learning models for the task of super-enhancers prediction in human tumor cell lines with a specific focus on sequence-only features within studied entities of super-enhancer and enhancer elements in the human genome. The proposed SE-prediction method included the GENA-LM application at handling long DNA sequences with the classification task, distinguishing super-enhancers from enhancers using H3K36me, H3K4me1, H3K4me3 and H3K27ac landscape datasets from HeLa, HEK293, H2171, Jurkat, K562, MM1S and U87 cell lines. The model was fine-tuned on relevant sequence data, allowing for the analysis of extended genomic sequences without the need for epigenetic markers as proposed in early approaches. The study achieved balanced accuracy metrics, surpassing previous models like SENet, particularly in HEK293 and K562 cell lines. Also, it was shown that super-enhancers frequently co-localize with epigenetic marks such as H3K4me3 and H3K27ac. Therefore, the attention mechanism of the model provided insights into the sequence features contributing to SE classification, indicating a correlation between sequence-only features and mentioned epigenetic landscapes. These findings support the potential transformer models use in further genomic sequence analysis for bioinformatics applications in enhancer/super-enhancer characterization and gene regulation studies.

In this study, we present a novel and ultrasensitive magnetic lateral flow immunoassay (LFIA) tailored for the precise detection of zearalenone, a mycotoxin with significant implications for human and animal health. A versatile and straightforward method for creating non-covalent magnetic labels is proposed and comprehensively compared with a covalent immobilization strategy. We employ the magnetic particle quantification (MPQ) technique for precise detection of the labels and characterization of their functionality, including measuring the antibody sorption density on the particle surface. Through kinetic studies using the label-free spectral phase interferometry, the rate and equilibrium constants for the binding of monoclonal antibodies with free (not bound with carrier protein) zearalenone were determined to be k_on = 3.42 × 10⁵ M⁻¹s⁻¹, k_off = 7.05 × 10⁻⁴ s⁻¹, and K_D = 2.06 × 10⁻⁹ M. The proposed MPQ-LFIA method exhibits detection limits of 2.3 pg/mL and 7.6 pg/mL when employing magnetic labels based on covalent immobilization and non-covalent sorption, with dynamic ranges of 5.5 and 5 orders, correspondingly. We have successfully demonstrated the effective determination of zearalenone in barley flour samples contaminated with Fusarium graminearum. The ease of use and effectiveness of developed test systems further enhances their value as practical tools for addressing mycotoxin contamination challenges.