A computational framework named scTELL has been developed to quantify chromatin accessibility at individual transposable element loci using single-cell assay for transposase-accessible chromatin sequencing data. Transposable elements account for approximately half of the human genome and contribute substantially to gene regulatory networks by providing promoters, enhancers, and other cis-regulatory sequences. However, because retrotransposon copies share high sequence similarity, most existing single-cell chromatin accessibility investigations have been restricted to aggregate or family-level summaries. Standard analytical workflows in single-cell epigenomics have remained predominantly gene-centric, leaving the regulatory activity of individual mobile element insertions unmeasured across individual cells. The new approach, termed single-cell Transposable Element Locus-Level analysis, establishes a distance-weighted scoring strategy that models the chromatin state of distinct transposable element insertions, enabling researchers to construct cell-by-locus accessibility matrices directly from processed single-cell sequencing files.
To capture chromatin accessibility at specific genomic coordinates, scTELL adapts distance-dependent weighting principles previously established for estimating gene activity scores. Rather than relying on simple overlap counts between transposable element sequences and accessibility peaks, the model integrates transposase insertion signal intensities within defined flanking windows using an exponential decay function. This mathematical weighting assigns higher value to accessibility signals located closest to the element boundary while accounting for local chromatin structure and technical variation. In cancer analyses examining long elements such as LINE-1, SVA, and HERVK insertions exceeding specific length thresholds, the authors employed an exponential decay distance of 1 kilobase across 50-base-pair tiles over a window extending 1 kilobase beyond the element boundaries. In contrast, for peripheral blood mononuclear cells focused on compact Long Terminal Repeat sequences with a median length of 321 base pairs, the framework applied a 500-base-pair decay distance across 5-base-pair tiles with a 500-base-pair upstream analysis window.
The methodological effectiveness of scTELL’s distance-weighted scoring was evaluated against an unweighted baseline method in healthy peripheral blood mononuclear cells across 12,414 transposable element loci. In this benchmarking comparison, the unweighted baseline, which simply summed peak scores across an identical 500-base-pair upstream window, achieved a cell-type clustering silhouette score of 0.085. In comparison, scTELL achieved a silhouette score of 0.226, representing a 2.66-fold improvement in clustering quality and demonstrating that distance-dependent weighting effectively distinguishes biological signal from background noise. Furthermore, the clustering performance of scTELL was comparable to the silhouette score of 0.227 obtained from standard gene activity matrices computed using default weighted genic models. These evaluations confirmed that locus-specific transposable element accessibility profiles alone possess sufficient biological fidelity to resolve distinct immune cell types with precision comparable to conventional gene-based chromatin analyses.
Because many transposable element insertions reside within or adjacent to protein-coding genes, the investigators conducted rigorous filtering analyses to verify that transposable element clustering signals were not mere reflections of genic open chromatin. When the researchers completely removed all sequencing reads mapping to annotated gene bodies prior to score computation, the resulting silhouette score remained at 0.179, and uniform manifold approximation and projection embeddings continued to display distinct immune cell groupings. In complementary locus-level filtering, excluding transposable elements that directly overlapped gene annotations yielded a silhouette score of 0.219 across 10,996 intergenic loci, while adding a conservative 5-kilobase buffer around gene boundaries yielded a score of 0.210 across 10,033 loci. These findings demonstrated that the moderate decrease in separation observed under read-level exclusion reflected the loss of bona fide cis-regulatory transposable elements located within introns or promoters rather than the elimination of technical artifacts.
Applying scTELL to single-cell chromatin profiles of healthy peripheral blood mononuclear cells enabled the comprehensive identification of lineage-specific Long Terminal Repeat elements. Among the 12,414 analyzed loci, elements belonging to the ERVL-MaLR family dominated the profile and exhibited pronounced cell-type preferences, with monocytes demonstrating the largest overall count of active loci. When tracking cell-type markers, the authors identified distinctive patterns such as specific LTR2B elements that showed predominant activity in B cells, while selected individual LTR2B loci displayed unique accessibility within CD8-positive T cells or monocytes. The validity of these single-cell predictions was corroborated using independent bulk chromatin accessibility data from fluorescence-activated cell sorting of sorted immune populations, where the top 20 marker loci identified per cell lineage demonstrated concordant, normalized accessibility signatures across corresponding bulk cell types.
Unbiased transcription factor motif enrichment analyses of accessible peaks surrounding cell-type-specific transposable elements revealed coordinated regulatory programs across immune lineages. For example, CD4-positive T-cell-specific transposable element peaks were significantly enriched for putative binding motifs of LEF1, KLF15, and SP4, consistent with lymphoid regulatory roles. Peaks specific to CD8-positive T cells demonstrated enrichment for RUNX1, TBX2, and EOMES, the latter known for guiding cytotoxic T-cell functionality, while natural killer cells showed high enrichment scores for EOMES, TBX20, and TBX2. B-cell-associated elements were enriched for ITF1, STAT1:STAT2, and SPIB motifs, and monocytes exhibited heightened representation of CEBPA, CEBPD, and SPIC motifs. Furthermore, family-wide assessments across 466 transposable element families showed that even loci belonging to the same element family recruited divergent transcription factor motifs depending on the cellular context, while comparison against gene promoters identified 192 motifs specifically enriched within transposable element regions relative to transcription start sites.
To explore transposable element activity in disease, the researchers applied scTELL to clear cell renal cell carcinoma, analyzing independent discovery and validation cohorts comprising 54,966 cells from 19 samples and 26,795 cells from 3 samples, respectively. By focusing on structurally intact, young transposable elements, the team identified 955 candidate loci, highlighting a specific L1PA2 retrotransposon located at chromosome 20:8,595,101–8,601,127 on the negative strand that displayed clear tumor-specific accessibility across both patient cohorts. Genomic inspection revealed that this L1PA2 element resides within an intron of the PLCB1 gene, where single-cell coverage plots revealed pronounced upstream chromatin opening in malignant cells. When evaluated against patient outcome data from the bulk TCGA-KIRC cohort, individuals exhibiting higher chromatin accessibility at this extended L1PA2 locus experienced a significantly shorter progression-free interval compared to patients with low accessibility, establishing an empirical correlation between retrotransposon accessibility and clinical disease trajectory.
The study also examined intratumoral and inter-patient molecular diversity in breast cancer by evaluating 147,183 single cells collected across 24 primary tumor specimens from the Human Tumor Atlas Network, focusing on 90,322 annotated malignant cells. Utilizing a cluster-free density estimation method adapted from the singleCellHaystack algorithm, the authors quantified non-random spatial divergence across embedding coordinates without imposing discrete cluster boundaries. Analysis of 1,388 young transposable element loci revealed marked inter-patient differences in accessibility among HERVK, L1PA, and SVA subfamilies. Connecting the most divergent breast cancer loci to bulk chromatin accessibility profiles from TCGA revealed six loci significantly associated with overall patient survival. Specifically, four loci exhibited poorer survival among patients with high accessibility, whereas high accessibility at two loci—an intronic L1PA4 element within SLC1A1 and an L1PA7 insertion within ST8SIA1—correlated with improved survival outcomes.
Despite its analytical capabilities, the authors noted several technical boundaries and methodological considerations inherent to the current scTELL implementation. The framework depends on uniquely aligned sequencing reads to maintain locus specificity, which necessarily limits the resolution of evolutionarily young transposable element subfamilies whose copies share nearly identical sequences. Additionally, the reported clinical associations in clear cell renal cell carcinoma and breast cancer remain strictly correlative, and the authors emphasize that prospective clinical cohorts and targeted functional experiments will be necessary to determine whether any individual retrotransposon locus exerts a direct mechanistic or causal influence on tumor progression. Future extensions that integrate single-cell chromatin accessibility directly with matched single-cell transcriptomic measurements will help clarify whether locus-specific accessibility directly dictates retrotransposon transcription or alters the expression of neighboring host genes.
In summary, scTELL establishes a dedicated computational method for quantifying chromatin accessibility at individual transposable element insertions from single-cell ATAC-seq data, bridging a major analytical gap in single-cell epigenomics. By providing direct compatibility with standard analytical workflows such as Signac and ArchR, the tool enables investigators to incorporate repetitive genomic elements into routine single-cell investigations without re-processing raw sequencing read files. The findings demonstrate that locus-level transposable element accessibility captures cell-type-specific chromatin states, reveals distinct transcription factor motif architectures, and identifies non-random accessibility patterns associated with clinical outcomes in human cancers, establishing a systematic approach for dissecting the regulatory contributions of repetitive DNA at single-cell resolution.
Subject of Research: Single-cell epigenomic analysis of locus-specific transposable element chromatin accessibility
Article Title: scTELL: a single-cell ATAC-seq tool for locus-specific transposable element identification in chromatin accessibility
Article References: Jeong, K., Ha, H., Xing, J., Choi, J., & Kim, K. (2026). scTELL: a single-cell ATAC-seq tool for locus-specific transposable element identification in chromatin accessibility. Mobile DNA, 17(1), Article 10. https://doi.org/10.1186/s13100-026-00395-y
Image Credits: AI Generated
DOI: 10.1186/s13100-026-00395-y
Keywords: transposable elements, scATAC-seq, chromatin accessibility, epigenomics, computational biology, cellular heterogeneity, retrotransposons, LINE-1, clear cell renal cell carcinoma, breast cancer, peripheral blood mononuclear cells, long terminal repeats, transcription factors, motif enrichment, cancer epigenetics, single-cell genomics
Cite Scienmag News
Nathaniel Bowman. (November 20, 2026). New Single-Cell Tool Maps Individual Transposable Element Accessibility Across Human Cells. Scienmag. https://scienmag.com/new-single-cell-tool-maps-individual-transposable-element-accessibility-across-human-cells/
Nathaniel Bowman. "New Single-Cell Tool Maps Individual Transposable Element Accessibility Across Human Cells." Scienmag, 20 November 2026, https://scienmag.com/new-single-cell-tool-maps-individual-transposable-element-accessibility-across-human-cells/. Accessed 20 November 2026.
Nathaniel Bowman. "New Single-Cell Tool Maps Individual Transposable Element Accessibility Across Human Cells." Scienmag. November 20, 2026. https://scienmag.com/new-single-cell-tool-maps-individual-transposable-element-accessibility-across-human-cells/








