Genome Biology

Accurate transcript abundance estimation is central to many transcriptomic studies. Many current quantification methods rely on reads mapped directly to the transcriptome, but transcriptome alignment can misassign reads from unannotated transcripts to annotated isoforms, leading to biased abundance estimates. We introduce Bramble, a method that projects spliced genomic alignments into transcriptomic coordinates to produce alignments compatible with downstream transcript quantification tools. Across simulated short- and long-read RNA-seq datasets and multiple levels of reference annotation completeness, incorporating Bramble into quantification pipelines consistently improved accuracy and reduced error. These results suggest that genome-derived transcriptomic alignments can improve transcript quantification by preserving compatible alignments to annotated transcripts while filtering alignments likely originating from unannotated transcripts.

Sequence-to-function models learn regulatory features from genomic sequence, but they remain limited in their ability to predict gene-expression differences among individuals. Cell-type-specific regulatory effects may be obscured in bulk RNA sequencing, whereas paired genotype and single-cell expression cohorts remain small. We evaluated whether deconvolution of bulk RNA-seq could provide scalable cell-type-specific targets for personal-genome expression prediction. GTEx v8 bulk RNA-seq from six tissues was deconvolved with BayesPrism using single-nucleus reference profiles, producing targets across 83 tissue-cell-type contexts. Deconvolved expression agreed with matched pseudobulked GTEx single-nucleus RNA-seq, with median donor-level Pearson correlations across genes ranging from 0.53 to 0.73 by tissue. We compared genotype-feature models, regressors trained on frozen Enformer representations, and fine-tuned Enformer and Borzoi models. Across random and nonlinear-enriched gene sets, sequence-derived approaches generally outperformed genotype-feature baselines, while frozen Enformer features were competitive with end-to-end fine-tuning. For the random gene set, Fisher-averaged Pearson correlations were 0.122-0.142 for sequence-derived approaches and 0.081-0.086 for genotype-feature baselines in a coverage-aware sensitivity analysis. Model performance was positively associated with deconvolution-pseudobulk agreement for sequence-derived models (r = 0.35-0.43 across tissue-cell-type contexts), suggesting that target reliability may constrain downstream prediction. Context-specific Enformer fine-tuning did not materially out-perform a shared, combined-context strategy. These results support deconvolution as a feasible approach for generating cell-type-resolved training targets, while showing that target quality and limited cohort size remain important constraints. Frozen pretrained representations provide a computationally efficient and competitive baseline for personal sequence-to-expression modeling.

With improvements in sequencing and assembly have come many high-quality telomere-to-telomere assemblies and reference pangenomes. However, the long-read sequencing recipes needed for high quality assemblies are expensive, and out of reach for many research groups. Here we propose ImpuT2T, a method that takes an assembly produced via inexpensive HiFi sequencing reads, and uses a panel of T2T (or near-T2T) assemblies to scaffold and fill ("patch") the gaps between the HiFi contigs. Benchmarking against reference assemblies demonstrates that ImpuT2T is highly effective at patching human HiFi assemblies, consistently outperforming existing patching approaches. Moreover, we show that including more haplotypes in the pangenome improves the quality of the patched assemblies, with the greatest gains achieved using the full HPRC Release 2 pangenome.

Sequence-to-function models predict molecular readouts directly from DNA, but recognizing a functional regulatory element is not equivalent to assigning the gene it regulates. We evaluated whether AlphaGenome deletion responses identify experimentally supported enhancer-gene relations, using a frozen K562 analysis and a primary-human-astrocyte CRISPR interference (CRISPRi) resource external to our analysis. The K562 mean contrast was positive but heavy-tailed, and exact joins showed direct collision with released Gasperini and ENCODE-rE2G resources; we therefore treated K562 as supporting evidence. In a frozen evaluation of 2,307 AstroREG relations, AlphaGenome deletion strength discriminated 133 functional relations from 2,174 well-powered nonfunctional relations (average precision 0.479, enhancer-cluster 95% confidence interval 0.394-0.561, prevalence 0.058; area under the receiver-operating-characteristic curve 0.726, 0.659-0.786). Adding AlphaGenome to distance, ABC score, enhancer length, measured expression and assay-depth context increased enhancer-grouped out-of-fold average precision from 0.396 to 0.534 and improved log loss from 0.169 to 0.150. The authors cross-fitted EGrf score was stronger alone (average precision 0.559); in a post-hoc calibration that held out both gene and enhancer folds, adding AlphaGenome increased average precision from 0.550 to 0.619 (paired enhancer-cluster increment 0.068, interval 0.023-0.115) and improved log loss from 0.143 to 0.132. This comparison had asymmetric inputs: EGrf was supervised on AstroREG labels and used local epigenomic and context features, whereas the AlphaGenome score was not fitted in this study to those labels or that feature panel but was read from a pre-existing primary-astrocyte RNA-seq output track. A post-hoc same-enhancer analysis gave conditional AUC 0.741 (0.663-0.814); a smaller same-gene analysis (34 genes, 155 relations) gave 0.701 (0.571-0.823). AstroREG labels and EGrf outputs were public before AlphaGenomes public release, so this evaluation is external to our study but not a post-release or proven-unseen benchmark. The results support complementary relation-level utility, not EGrf superiority, sequence-only deployment, causal assignment at arbitrary loci or equivalence between sequence deletion and CRISPRi.

Adenosine-to-inosine (A-to-I) RNA editing is a widespread post-transcriptional mechanism that diversifies the transcriptome. While ADAR enzymes catalyze this reaction, the upstream mechanisms that determine why individual adenosines are edited at markedly different efficiencies remain poorly understood. Here, we integrated multi-omics datasets from human cell lines and mouse embryonic tissues and developed machine learning models that distinguish high-and low-frequency editing sites based on local epigenetic features. Across species, tissues, and developmental stages, H3K36me3 consistently emerges as the strongest negative predictor of editing frequency, whereas the histone variant H2A.Z.1 shows a positive association. Functional validations in H2AFZ-knockdown cells reveal that H2A.Z.1 preferentially facilitates editing at high-frequency sites (76.3% of sites decreased, mean {Delta} =-0.07), whereas SETD2-knockout-mediated loss of H3K36me3 selectively derepresses editing at low-frequency sites (86.9% upregulated, [~]2.7-fold increase). These findings establish chromatin state as an upstream regulatory layer that modulates RNA editing independently of editing enzyme abundance and reveal opposing roles for H3K36me3 and H2A.Z.1 in shaping RNA editing landscapes. Together, our study provides a conceptual framework linking epigenetic regulation to post-transcriptional RNA modification, suggesting that chromatin-mediated regulation contributes to the establishment of site-specific RNA editing programs across mammalian genomes.

Fusion transcripts contribute to cancer, inherited diseases, developmental disorders, and evolution. Long-read RNA sequencing enables direct sequencing of full-length transcripts, creating new opportunities to detect complex fusion architectures, including previously inaccessible multi-segmented fusion transcripts. However, accurate transcriptome-wide fusion detection remains challenging because existing methods struggle to distinguish genuine fusion events from technical artefacts. Here, we present a comprehensive benchmark of transcriptome-wide fusion detection using simulated datasets and transcriptomes from three cancer cell lines across Oxford Nanopore Technologies (ONT) cDNA, PCR-cDNA, and direct RNA sequencing, Pacific Biosciences (PacBio) Kinnex sequencing, Illumina short-read RNA sequencing, six long-read fusion callers, and multiple analysis strategies. False-positive fusion calls remained the dominant limitation across sequencing platforms and algorithms. Increasing sequencing depth improved recall but also amplified spurious fusion calls, whereas higher read-support thresholds improved precision at the expense of sensitivity. ONT PCR-cDNA sequencing combined with CTAT-LR-Fusion achieved the best overall balance between precision and recall, whereas JAFFAL was the only caller to reliably identify simulated tri-gene fusions. Consensus calling reduced false positives but markedly reduced sensitivity, with only one of 400 simulated fusions detected by all six callers. Breakpoint localisation emerged as a major limitation across all methods. Long-read sequencing consistently recovered more validated fusion transcripts than short-read sequencing, enabled detection of complex tri-gene fusions, and produced more biologically plausible fusion landscapes with fewer promiscuous gene partners. Collectively, our results establish the first comprehensive benchmarking framework for transcriptome-wide fusion detection, using long-read RNA sequencing, and provide practical guidance for selecting sequencing workflows and computational strategies, while identifying key priorities for future algorithm development.

Alternative splicing is cell-type-specific and disease-relevant, but single-cell RNA sequencing is optimized for gene-level quantification, and how sequencing depth governs splice-junction recovery in the dominant 3'-biased chemistries has not been quantified, so experiments cannot be designed to a depth target and datasets cannot be interpreted against expected recovery. We generated deep single-cell RNA sequencing of human airway epithelial cells (12 samples, 102 to 803 million uniquely mapped reads per sample) and benchmarked junction recovery in five annotated samples against matched bulk RNA sequencing from a 190-donor cohort. Junction detection did not saturate with depth, and validity was governed by read support rather than annotation status. Recovering 90% of the observed full-depth junction complement required approximately 52,250 reads per cell, several-fold more than gene-level analysis. Quantifying splicing was more depth-limited than detecting it, and rare cell types were limited by cell number rather than depth. Against matched bulk, single-cell recall plateaued at a practical ceiling set by sample size and sparse sampling, and was independent of distance from the 3' end. These results establish a framework for designing splicing experiments in deep 3'-biased single-cell RNA sequencing, and provide an airway-epithelial dataset with matched bulk for benchmarking splicing methods.

Noncoding regulatory elements in the human genome can control gene expression through distinct mechanisms, including acting as enhancers, silencers, or topological elements. However, we lack tools to identify all such classes systematically and quantify their effects on gene expression in an endogenous genomic context. Here we develop Swap-seq, a method that combines pooled twin prime editing with flow-sorting to delete hundreds of endogenous regulatory sequences and measure their quantitative effects on the expression of a nearby gene. We apply Swap-seq to dissect the regulatory elements that control PPIF expression in THP-1 monocytes. We quantify the effects of distal and intronic enhancers, characterize CTCF insulators, and discover a silencer that regulates distal genes through histone deacetylase (HDAC)-mediated repression of neighboring enhancers. Hundreds of other genomic elements share its chromatin signatures and transcription factor binding sites, suggesting that this silencer represents a broader class of repressive elements. We leverage Swap-seq data to benchmark sequence-to-function models and show that they often capture the local activity of regulatory elements but fail to link them to distal target genes. Swap-seq thus provides a generalizable tool to discover regulatory elements across mechanistic classes, characterize their effects on gene expression, and generate large-scale datasets for evaluating sequence-to-function models.

Computational predictors of RNA splicing are increasingly used to interpret genetic variants and to design synthetic genes, yet they are almost always benchmarked on endogenous human sequences closely related to their training data. Whether their performance reflects genuine recognition of splicing signals, or instead exploits statistical features of natural genomes such as conservation and exon-intron composition, remains unclear. Here we benchmark eleven splicing predictors on thousands of synthetic GFP variants that are heavily recoded and dissimilar from any training data, using long-read sequencing to measure splicing directly at each position. Despite this distribution shift, modern deep-learning predictors retained strong performance, and the resulting ranking was largely stable across position-level and construct-level benchmarks. SpliceTransformer ranked highest, followed by AlphaGenome and SpliceAI. Tools that ignore long-range sequence context performed substantially worse, largely because they assign high scores to many non-spliced positions. This ranking broadly agrees with benchmarks on endogenous variants, indicating that the leading models capture transferable, sequence-intrinsic determinants of splicing. We further provide a unified calibration that maps each predictor's scores onto the measured fraction of spliced reads, allowing scores to be interpreted as splicing outcomes and compared directly between tools. Our results show that current deep-learning models generalise beyond natural genomes and provide a practical framework for splicing-aware sequence design.

AO_SCPLOWBSTRACTC_SCPLOWIdentifying spatially variable genes (SVGs), genes whose expression varies coherently across tissue space, is a central analytic goal in spatially resolved transcriptomics. Current methods rank spatially variable genes using either significance probabilities from parametric models or effect sizes such as the proportion of spatial variance, but parametric approaches impose distributional assumptions, such as Gaussian processes or negative binomial models, that may be violated for sparse or zero-inflated data. Furthermore, most detection tools cannot jointly model multiple biological replicates, and no existing framework provides both nonparametric significance probabilities and stabilized effect-size estimates with formal uncertainty quantification. Here, we introduce CytoKspace, a nonparametric framework that combines a sparse exponential kernel constructed from nearest-neighbor graphs with a quadratic-form test statistic and adaptive permutation testing. CytoKspace employs a multi-stage adaptive permutation schedule that yields substantial computational savings over fixed-permutation baselines, an adaptive shrinkage layer built on empirical Bayes estimation that stabilizes raw spatial effect sizes and provides posterior estimates with local false sign rates, and a scalable multi-sample extension via Fisher combination of significance probabilities and inverse-variance-weighted meta-analysis that accommodates studies with multiple biological replicates. In extensive simulations across a broad range of sample sizes, gene counts, spatially variable gene fractions, and effect sizes, as well as in applications to two real datasets from the Visium and seqFISH platforms, CytoKspace demonstrates competitive sensitivity, well-calibrated false positive rates, and practical computational requirements compared to existing methods. A software implementation of our method is freely available at https://github.com/Ghoshlab/CytoKspace.

Most genomic foundation models are pretrained on independent linear assemblies and therefore do not explicitly represent population-level segment sharing or local graph connectivity. We developed TomatoPGFM, a graph-conditioned model pretrained on 54.65 Gb of sequence from 66 tomato (Solanum spp.) accessions. Sequence tokens were conditioned on pangenome node attributes and local adjacency, and the model was optimised using masked language modelling and graph-feature reconstruction. To evaluate model responses to graph-conditioned input, we compared aligned, shuffled and disabled graph inputs in 25,000 windows from the training panel. Sequence-aligned graph input produced lower masked language modelling loss than graph-off at all five curriculum stages in both training-panel strata, while the shuffled perturbation generally yielded intermediate losses. We then assessed sequence-only transfer in Solanum sitiens LA1974 and S. lycopersicum MicroTom, neither of which was used for graph construction or pretraining. Frozen-probe AUROC values for gene-versus-intergenic and coding-sequence-versus-intergenic classification ranged from 0.8489 to 0.9593. TomatoPGFM produced higher AUROC point estimates than DNABERT-2 in all four comparisons. Enabling the zero-feature GraphAdapter pathway with adjacency messaging disabled changed throughput by less than 1% at 512-2,048 positions under the tested configuration. Together, these results show that TomatoPGFM responds consistently to sequence-aligned pangenome context in training-panel sequences and provides informative sequence representations for genic-region classification in accessions excluded from graph construction and pretraining.

Transcriptomics datasets generated using next-generation sequencing techniques such as single cell RNA-sequencing (scRNA-seq) and spatially-resolved transcriptomics (SRT) allow researchers to study patterns in gene expression across celltypes, temporal processes, and spatial organization at ever-higher resolutions and depths. scRNA-seq analyses produce gene expression profiles and celltype-specific gene sets that require annotation to provide biological meaning, a process that has traditionally relied on the manual interpretations of clinical scientists. Similarly, SRT experiments typically require subjective, time-consuming annotation of spatial domains. Recent advances in large language model (LLM) methods offer opportunities to assist in the interpretation of such datasets. Many current LLM-based approaches aim to annotate transcriptomics-derived gene sets by integrating information from publicly available and online biological resources. While these approaches can be effective, they often struggle when presented with weakly-related or fully uncorrelated genes, sometimes inferring and justifying biological relationships that are not supported by existing literature. Additionally, the quality of LLM-generated interpretations is dependent on the provision of appropriate biological context and careful prompt design, both of which can present significant barriers to effective use. To address these limitations we propose genesetGPT, an efficient, LLM-based framework that emphasizes both curated biological context and iterative prompt construction, thus enabling realistic summarization of heterogeneous gene sets at scale. genesetGPT is implemented as an open-source Python package available for download at https://github.com/jr-leary7/genesetGPT.

We benchmark four published splicing variant-effect predictors against a multiplexed experimental splicing assay. On 27,733 single-nucleotide variants in and around human exons from MFASS with measured exon-inclusion outcomes, Pangolin is the strongest predictor of splice-disrupting variants (AUROC 0.888, average precision 0.421), ahead of SpliceAI (0.819, 0.321) and SpliceTransformer (0.786, 0.317), with MMSplice fourth (0.758, 0.256); all four exceed the older SPANR model (0.748, 0.228). The ranking reproduces the relative performance reported by the Pangolin authors, a correctness check on the pipeline. A calibrated consensus of the three deep-learning sequence-window predictors, evaluated on an exon-grouped held-out split, does not meaningfully improve over Pangolin alone. Stratifying by distance to the splice site exposes a shared blind spot: all five tools detect disruptions within a few bases of the splice site well, but recall declines sharply in the exon interior, and 19% of disrupting variants are missed by every tool; these shared misses are enriched among variants away from splice sites, and are predominantly exon-interior. MMSplice, the one model built for modular exonic and intronic effects rather than splice-site recognition, shows the same distance-dependent decline, so the blind spot is not an artifact of splice-site-centric architectures. Every number is computed against fixed experimental ground truth and is reproducible from the public dataset and the released code.

Pooled CRISPR screens increasingly use longitudinal, donor-adjusted, and factorial designs, but beta-binomial screen methods have largely remained limited to pairwise comparisons. BARCS extends the library-total-conditional beta-binomial model to guide-level regression with an arbitrary design matrix, enabling direct estimation of time, covariate, and interaction effects. In four replicate-complete Cas13 screens, adding the intermediate time point modestly improved essential-gene recovery. Applying the same non-targeting-control scaling rule to BARCS, MAGeCK-MLE, edgeR-QL, DESeq2, and limma--voom produced similar calibration across all five methods, while the four alternatives ranked essential genes more strongly than BARCS. In an ordered-bin IL2RA screen, donor-adjusted BARCS recovered more validated regulators with fewer total calls than the matched four-bin MAGeCK-MLE fit, and cross-fitted controls exposed excess guide-level significance. Simulations showed gains from dispersion moderation and control-based denominators, but seed-specific results exposed denominator sensitivity and a null grid localized substantial gene-level error to correlated-guide aggregation rather than dispersion alone. Aggregation-matched control scaling reduced but did not eliminate this error. An external audit prompted by concerns about beta-binomial false discoveries showed that the reported CB2 null-discovery count disappeared when full-library totals were restored. This corrected one denominator-dependent result but did not refute the broader calibration concern; nominal-level calibration remained unresolved. BARCS therefore contributes a multivariable extension of the library-total-conditional beta-binomial model together with an explicit account of where its inference is valid: guide-level coefficients are supported by independent biological libraries, whereas gene-level summaries and partitioned-bin designs require correlation-aware aggregation or joint modelling that the present implementation provides diagnostically rather than generatively. We report this boundary because complex pooled designs make it consequential, not because it is unique to the beta-binomial model.

Reliable translation of microRNA sequencing data depends on effective harmonization to mitigate artifacts from variable experimental handling. Although many harmonization methods exist, prior evaluations have focused mainly on differential expression analysis, leaving the impact on subgroup discovery understudied. We present a framework for evaluating harmonization in the context of sample clustering, which integrates AI-augmented datasets, statistical evaluation pipelines, and accessible software tools, enabling systematic comparisons across diverse signal-to-artifact ratios and cluster composition settings. Using this framework, we show that harmonization can, often partially, restore clustering accuracy lost to artifacts, especially at moderate signal-to-artifact ratios, with the level of gains depending on the specific harmonization method, the paired clustering technique, and the cluster composition setting. We further confirm these findings by analyzing reconstructed cohorts from The Cancer Genome Atlas breast cancer microRNA sequencing data. Collectively, the results underscore the need for tailored harmonization to support reliable subgroup discovery and highlight the broader importance of context-specific workflows in translational genomics.

Ribosome profiling experiments capture a snapshot of translation by sequencing ribosome-protected fragments (RPFs). Alignment to a well-curated transcriptome captures most RPF signal, but can miss ribosome activity when expressed exons are absent from the reference. Genome alignment of RPFs expands the detectable sequence space, but can expose ambiguity between homologous loci. Existing workflows often couple reference choice with the handling of ambiguously mapped reads, making it difficult to isolate the effects of either decision on downstream analyses. Here, we developed RiboFlow v2 by adding genome alignment and configurable handling of multimapping reads to the existing transcriptome workflow. Using 24 human ribosome profiling libraries and their matched RNA-seq data, we found that nucleotide-level coverage at matched coding-sequence positions and translation efficiency (TE) were broadly concordant between alignment strategies under stringent alignment filters. Even under these conditions, individual genes showed substantial differences in coverage and TE estimates. These differences included coverage in alternative exons absent from selected transcripts and ambiguous alignments between protein-coding genes and processed pseudogenes. Clustering genes by read-assignment patterns identified groups associated with pseudogene prevalence and highlighted shared exonic sequence among reference transcripts as a source of alignment discrepancies. RiboFlow v2 extends the analysis beyond selected reference transcripts and enables identification of genes whose ribosome coverage and abundance estimates depend on reference composition and how ambiguously mapped reads are handled.

Correcting batch effects and integrating single-cell sequencing datasets has been a crucial step in large-scale biological studies. Many methods have been published for this task, and excel in various scenarios. Our previous work, LIGER, leveraging integrative non-negative matrix factorization (iNMF), stands out in providing an interpretable low-dimensional representation. To adapt to the modern need for integrating millions of cells, we developed a highly-optimized parallel factorization solution with on-demand loading from disk. The upgraded LIGER algorithm shows significant improvements in time and memory efficiency for single-cell data integration. We also developed a new downstream embedding alignment method significantly improved performance in conserving biological variation while still aligning corresponding cell types across datasets.

Single-cell perturbation experiments measure how interventions alter cellular phenotypes. However, the number of possible perturbations and biological contexts far exceeds what can be tested experimentally. Motivated by this constraint, predictive models aim to extrapolate cellular responses to unseen conditions. Despite substantial efforts in model development, benchmark studies have reached inconsistent conclusions about the capabilities of current perturbation-response models. A major challenge is that evaluation protocols vary widely across studies, making results difficult to compare. Furthermore, the lack of consensus on evaluation hampers progress because it is unclear which predictive capabilities new models should prioritize. To help build consensus on evaluation principles, we develop a taxonomy that decomposes evaluation protocols into their representation, metric, score transformation, and reporting strategies. We characterize how these choices determine which aspects of prediction quality a benchmark measures and discuss criteria for selecting and assessing protocols in relation to specific benchmarking goals. We additionally provide scPertEval, a Python package with reference implementations of selected evaluation protocols, and use it to assess protocol behavior across seven publicly available single-cell perturbation datasets. By making evaluation choices and their underlying trade-offs explicit, we aim to stimulate a community discussion about developing more comparable and task-aligned evaluation protocols. Graphical Abstract O_FIG O_LINKSMALLFIG WIDTH=200 HEIGHT=129 SRC="FIGDIR/small/740433v1_ufig1.gif" ALT="Figure 1"> View larger version (32K): org.highwire.dtl.DTLVardef@bd1e3corg.highwire.dtl.DTLVardef@c26c2org.highwire.dtl.DTLVardef@1c49afdorg.highwire.dtl.DTLVardef@9b988c_HPS_FORMAT_FIGEXP M_FIG C_FIG

Accurate predictions of transcriptomic responses to genetic perturbations could unlock our understanding of gene functions and regulatory networks. While a growing number of methods and benchmarks target this task, existing evaluations focus on mean expression accuracy alone. This overlooks differential expression (DE), which captures both mean and variance and forms the basis for biological interpretation and experimental follow-up. Here, we systematically evaluate a diverse set of deep learning and non-deep-learning methods for their ability to predict DE outcomes under two generalization regimes: unseen perturbations within the same cell line, and unseen cellular contexts across cell lines. We find that simple baselines, such as embedding-based nearest neighbors, are competitive and often outperform specialized deep learning models for DE classification across datasets and evaluation metrics. We further show that sparsity calibration, motivated by the structure of single-cell data, substantially improves DE classification for deep learning models that do not explicitly account for sparsity. Together, our findings establish practical baselines and evaluation principles for benchmarking perturbation models on DE prediction.

Single-cell RNA sequencing (scRNA-seq) studies often profile only one or a few biological samples, limiting assessment of discovery stability under between-sample variation. We present scStable, a framework for assessing scRNA-seq analysis stability by generating synthetic samples that preserve cell-level heterogeneity from a single-sample reference while injecting bulk-derived between-sample variation. In gastric cancer data comprising 21 matched scRNA-seq and bulk RNA-seq profiles, scStable better recapitulates observed between-sample variation than ZINB-WaVE, subsampling, and Gaussian noise injection. Using synthetic sam-ples, scStable enables stability-aware analysis, including prioritization of stable differentially expressed genes and selection of clustering methods and hyperparameters.