Biology subjects

Chikina, M.

Publications and source records attributed to Chikina, M..

At least 19 recordsLinked to original sources

bioRxiv2026

HyphAeon: Attention on Evolution Across Deep Time Transforms Comparative Genomics

Detecting Darwinian natural selection is fundamental to evolutionary biology and functional genomics, yet standard methods based on phylogenetic models that estimate the ratio of non-synonymous to synonymous substitution rates (dN/dS) fail to scale with modern genomic volumes. Fitting continuous-time Markov substitution matrices across dense trees with hundreds of species requires extensive compute, forcing comparative genomics to rely on aggressive taxon subsampling or static whole-tree summaries that dilute transient adaptive bursts. Here we present HyphAeon, a lightweight (~1.91M parameter backbone, 2.46M across the full multi-task suite) phylogeny-informed foundation transformer trained to amortize the detection of episodic diversifying selection across 742-species mammalian coding alignments (17,186 genes, 9.77 x 10^6 codons). HyphAeon approaches the discriminative accuracy of numerical maximum-likelihood selection tests (MEME) across episodic burst regimes (ROC-AUC up to 0.942, mean 0.659; empirical Precision-Recall lift up to 25.8x, mean 7.7x; rank concordance up to rho = 0.983) while executing >1,000x faster per locus (averaging <1 ms per site across genome-wide scans) and >10,000x faster at proteome scale, generalizing outside its mammalian training distribution without retraining. Beyond accelerating classical tests, embedding molecular evolution into a differentiable geometric latent space enables analytical capabilities inaccessible to static dN/dS models: (1) targeted alignment artifact correction via counterfactual attribution; (2) macromolecular contact recovery and multi-site epistatic sectors (CESI); (3) directional phenotype-to-genotype attribution in lineage space (PARS); and (4) continuous temporal surveillance regression that tracks positive sweep velocities across longitudinal cohorts (evaluated across 12,167 timestamped genomes and benchmarked against external frequencies from >9.34 million genomes), rescuing adaptive substitutions obscured by post-fixation dilution. By bridging statistical phylogenetics with geometric representation learning, HyphAeon establishes comparative genomics as an interactive, high-throughput computational framework for evolutionary discovery.

bioinformatics↗

bioRxiv2026

RERconverge Update: Runtime Reduction and Analysis Function Overhaul

MotivationConvergent evolution, or the independent acquisition of similar phenotypes in distinct lineages, provides a powerful framework for investigating genomic changes associated with a phenotype. This paper details an update to RERconverge, a powerful R package that tests for associations between gene relative evolutionary rates (RERs) and convergent phenotypes to infer genomic regions associated with traits or selective pressures. We introduce new customizable analysis choices and scalable and efficient algorithms that can process larger genomic datasets, a critical improvement as genomic data become available for more species. ResultsModifications to core functions in the RERconverge pipeline resulted in an immense speedup (by a factor of up to 28.6). The function that tests for associations between phenotypes and RERs has been expanded to include two new analytical methods for outlier control; we also provide here a summary of the statistical tests users can perform, along with their use cases. Availability and implementationThe code and walkthrough vignettes for the package are available at https://github.com/nclark-lab/RERconverge. ContactNathan L. Clark nclark@pitt.edu; Maria Chikina mchikina@pitt.edu

genomics↗

bioRxiv2026

A modality gap in personal-genome prediction by sequence-to-function models

Sequence-to-function (S2F) models trained on reference genomes have achieved strong performance on regulatory prediction and variant-effect benchmarks, yet they still struggle to predict inter-individual variation in gene expression from personal genomes. We evaluated AlphaGenome on personal genome prediction in two molecular modalities--gene expression and chromatin accessibility--and observed a striking dichotomy: AlphaGenome approaches the heritability ceiling for chromatin accessibility variation, but remains far below baseline for gene-expression variation, despite improving over Borzoi. Context truncation and fine-mapped QTL analyses indicate that accessibility is governed by local regulatory grammar captured by current architectures, whereas gene-expression variation requires long-range regulatory integration that remains challenging.

bioinformatics↗

bioRxiv2026

BUSTED-PH: Isolating the genomic signatures of convergent phenotypes

Convergent evolution offers a natural test for adaptive predictability, yet pinpointing its molecular basis remains difficult. Current methods often grapple with distinguishing adaptive convergence from background noise, either by demanding overly stringent identical substitutions or relying on low-resolution evolutionary rate shifts. Here we introduce BUSTED-PH (Branch-site Unrestricted Statistical Test for Episodic Diversification - Phenotype), a branch-site codon model that detects phenotype-associated episodic diversifying selection. Unlike standard approaches, BUSTED-PH explicitly contrasts selective regimes between phenotype-positive (foreground) and phenotype-negative (background) lineages, effectively winnowing out spurious associations driven by pervasive background adaptation. BUSTED-PH has already been applied in numerous independent studies, and here we rigorously validate it using canonical positive controls and simulations to confirm high power and strict false positive control. A genome-wide scan of 120 mammalian species using strict statistical criteria (FDR [≤] 0.01) identified 72 genes associated with echolocation; while recovering paradigmatic auditory drivers (e.g., Prestin, TMC1 ), we also uncover novel candidates in physiological support systems ranging from lipid homeostasis to neural development. A parallel analysis of mammalian gigantism identifies 91 genes linked to musculoskeletal reinforcement, organ size governance, and genomic integrity, characterizing the molecular adaptations required to support massive body size. As projects like Zoonomia and the Vertebrate Genomes Project expand comparative datasets, BUSTED-PH provides a robust framework for dissecting the genetic architecture of complex convergent traits.

evolutionary biology↗

bioRxiv2026

When Less Is Enough: Low-Rank Structure in DNA Sequence-to-Function Models

MotivationThe rapid success of deep learning sequence-to-function (S2F) models has driven a trend toward ever larger architectures for regulatory genomics. While these models achieve meaningful predictive performance, their growing size and computational cost have made them increasingly opaque, difficult to deploy, and impractical for widespread use. As S2F models mature, an open question is how much model capacity is truly necessary to represent regulatory sequence function. ResultsWe show that, within linear layers, much of the predictive power of state-of-the-art S2F models is concentrated in low-rank structure. Using post hoc singular value decomposition, we construct low-rank approximations of the Sei model that reduce model size by up to 90% while preserving over 90% correlation with full-model predictions. Strikingly, extremely low-rank models--including rank 1--can match or exceed full-model performance on several variant-effect benchmarks. This indicates dominant regulatory signals may lie in a surprisingly low-dimensional subspace. Combining low-rank approximation with static model quantization enables practical CPU inference, yielding over 5x speedups relative to the full Sei model and more than 100x faster inference than larger S2F models on comparable tasks. Applying the same approach to Borzoi, Enformer, and AlphaGenome reveals consistent behavior, demonstrating that low-rank structure is a general property of modern S2F architectures, not a model-specific artifact. Availability and ImplementationLow-rank Sei is available at https://github.com/kostkalab/seillra. Contactmchikina@pitt.edu, kostka@pitt.edu

bioinformatics↗

bioRxiv2025

InfluenceNet: Encoding Motif Influence for Interpretable Modeling of Cis-Regulatory Syntax

Cis-regulatory elements shape gene expression by recruiting transcription factors (TFs) to DNA motifs, yet how motifs cooperate or compete across distances remains poorly understood. Existing deep learning models predict TF binding with high accuracy but rely on computationally intensive post-hoc analyses to infer regulatory syntax, limiting interpretability and scalability. Here we present InfluenceNet, a transparent convolutional architecture that directly encodes motif influence-a position- and motif-specific profile quantifying how one motif modulates binding at another. Trained on ChIP-nexus data for pluripotency TFs, InfluenceNet achieves performance comparable to state-of-the-art models while providing immediate, global interpretability. The learned influence profiles recapitulate known principles of motif syntax, including Oct4-Sox2 pioneering activity and Nanogs striking 10.5-bp helical periodicity, and reveal a cooperative Nanog-Nanog grammar supported by de novo motif discovery. Integrating these results with structural and molecular-dynamics modeling, we propose that Nanog forms chain-like oligomers on DNA through asymmetric contacts between WRD (tryptophan repeat domain) and DBD (DNA-binding domain) that maintain helical phase alignment. Together, these findings establish InfluenceNet as a general framework for accurate prediction and mechanistic interpretation of cis-regulatory grammar, bridging deep learning and biophysical insight.

bioinformatics↗

bioRxiv2025

Mechanistic Genome Folding at Scale through the Differentiable Loop Extrusion Model

The spatial folding of the genome shapes gene regulation by controlling which loci interact, yet inferring the mechanisms behind these 3D structures from contact maps remains difficult. Cohesin-mediated loop extrusion is a key organizer of domains and loops, but existing methods either predict contacts without mechanistic insight or simulate extrusion with limited scalability. We present the differentiable loop extrusion model (dLEM), a scalable framework that reformulates extrusion as a smooth, trainable process. dLEM represents extrusion through position-specific velocity profiles for leftward and rightward cohesin movement. Fitting dLEM to chromosome conformation capture data yields a one-dimensional, interpretable description of extrusion dynamics that aligns with genomic and epigenomic features. dLEM parameters also capture architectural changes under CTCF and WAPL perturbations, enabling genome-wide prediction of extrusion disruptions. Extending our observations, we demonstrate that dLEM can be seamlessly incorporated into deep learning models to infer extrusion parameters directly from sequence and chromatin features, reducing model complexity by nearly three orders of magnitude while preserving predictive accuracy. Indeed, when incorporated into deep dLEM, dLEM acts as a biophysically-motivated layer for long range genomic communication, and together they provide a predictive, interpretable framework linking 1D genomic features to 3D chromatin folding and its response to sequence and chromatin state, with dLEMs mechanistic modeling enabling prediction of trans factor perturbation effects.

genomics↗

bioRxiv2025

DALE-Eval: A comprehensive cell type-specific expression deconvolution benchmark for transcriptomics data

Deconvolution of bulk transcriptomic data unlocks rich, cell type-specific insights from complex tissue samples. While cell type fraction deconvolution has been extensively developed and bench-marked, the next frontier--cell type-specific expression (CTSE) deconvolution--remains largely underexplored. Here, we introduce DALE-Eval, a comprehensive benchmark of CTSE methods, featuring novel evaluation metrics, identification of key drivers of performance, and practical recommendations for maximizing the value of CTSE in downstream applications.

bioinformatics↗

bioRxiv2025

PLIERv2: bigger, better and faster

Gene expression analysis has long been fundamental for elucidating molecular pathways and gene-disease relationships, but traditional single-gene approaches cannot capture the coordinated regulatory networks underlying complex phenotypes; although unsupervised matrix factorization methods (e.g., PCA, NMF) reveal coexpression patterns, they lack the ability to incorporate prior biological knowledge and often struggle with interpretability and technical noise correction. Semi-supervised strategies such as PLIER have improved interpretability by integrating pathway annotations during latent variable extraction, yet the original PLIER implementation is prohibitively slow and memory-intensive, making it impractical for modern large-scale resources like ARCHS4 or recount3. Here, we introduce CLAMP, which overcomes these constraints through a two-phase algorithmic design (an unsupervised "CLAMPbase" initialization followed by a "CLAMPfull" regression that incorporates priors via glmnet), rigorous internal cross-validation to tune regularization parameters for each latent variable, and efficient on-disk data handling using memory-mapped matrices from the bigstatsr package. Benchmarking on GTEx, recount2, and ARCHS4 demonstrates that CLAMP achieves 7x-41x speedups over PLIER, succeeds in modeling hundreds of thousands of samples that PLIER cannot handle, and maintains or improves biological specificity of latent variables as shown by tissue-alignment and pathway enrichment analyses. By filling the gap in scalable, biologically informed latent variable extraction, CLAMP enables comprehensive analysis of modern transcriptomic compendia and paves the way for deeper insights into gene regulatory networks and downstream applications in translational genomics.

bioinformatics↗

bioRxiv2025

Frustratingly easy domain adaptation for cross-speciestranscription factor binding prediction

MotivationUnderstanding how DNA sequence encodes gene regulation remains a central challenge in genomics. While deep learning models can predict regulatory activity from sequence with high accuracy, their generalizability across species--and thus their ability to capture fundamental biological principles--remains limited. Cross-species prediction provides a powerful test of model robustness and offers a window into conserved regulatory logic, but effectively bridging species-specific genomic differences remains a major barrier. ResultsWe present MORALE, a novel and scalable domain adaptation framework that significantly advances cross-species prediction of transcription factor (TF) binding. By aligning statistical moments of sequence embeddings across species, MORALE enables deep learning models to learn species-invariant regulatory features without requiring adversarial training or complex architectures. Applied to multi-species TF ChIP-seq datasets, MORALE achieves state-of-the-art performance--outperforming both baseline and adversarial approaches across all TFs--while preserving model interpretability and recovering canonical motifs with greater precision. In the five-species transfer setting, MORALE not only improves human prediction accuracy beyond human-only training but also reveals regulatory features conserved across mammals. These results highlight the potential of simple yet powerful domain adaptation techniques to drive generalization and discovery in regulatory genomics. Crucially, MORALE is architecture-agnostic and can be seamlessly integrated into any embedding-based sequence model. AvailabilityCode is available at https://github.com/loudrxiv/frustrating.

genomics↗

bioRxiv2025

Deep genomic models of allele-specific measurements

Allele-specific quantification of sequencing data allows for a causal investigation of how DNA sequence variations influence cis gene regulation. Current methods for analyzing allele-specific measurements for causal analysis rely on statistical associations between genetic variation across individuals and allelic imbalance. Instead, we propose DeepAllele, a novel deep learning sequence-to-function model using paired allele-specific input, designed to learn sequence features that predict subtle changes in gene regulation between alleles. Our approach is suited for datasets with unambiguous phasing, such as F1 hybrids and other controlled genetic crosses, or long-read sequencing technologies used in Fiber-seq, in which reads can be assigned to complete allele sequences. We apply our framework to allele-specific measurements in immune cells from F1 hybrid mice, and show that the models additionally learned cis-regulatory grammar aligns with known biological mechanisms across a significantly larger number of genomic regions compared to baseline models. In summary, our work presents a computational framework to leverage genetic variation to uncover functionally-relevant regulatory motifs, enhancing causal discovery in genomics.

bioinformatics↗

bioRxiv2025

Context-dependent Interactors Regulate TDP-43 Dysfunction in ALS/FTLD

TDP-43 mislocalization, aggregation, and loss of splicing function are neuropathological hallmarks in over 97% of Amyotrophic Lateral Sclerosis (ALS), 45% of Frontotemporal Lobar Degeneration (FTLD), and 60% of Alzheimers Disease, which has been reclassified as LATE-NC. However, the mechanisms underlying TDP-43 dysfunction remain elusive. Here, we utilize APEX2-driven proximity labeling and mass spectrometry to characterize the context-dependent TDP-43 interactome in conditions of cytoplasmic mislocalization, impaired RNA-binding contributing to aggregation, and oxidative stress. We describe context-dependent interactors, including disrupted interactions with splicing-related proteins and altered biomolecular condensate (BMC) associations. By integrating ALS and FTLD snRNA-seq data, we uncover disease-relevant molecular alterations and validate our dataset through a functional screen that identifies key TDP- 43 regulators. We demonstrate that disrupting nuclear speckle integrity, particularly through the downregulation of the splicing factor SRRM2, promotes TDP-43 mislocalization and loss of function. Additionally, we identify NUFIP2 as an interactor associated with mislocalization that sequesters TDP-43 into cytoplasmic aggregates and co-localizes with TDP-43 pathology in patient tissue. We also highlight HNRNPC as a potent TDP-43 splicing regulator, where precise modulation of TDP-43 or HNRNPC can rescue cryptic exon splicing. These findings provide mechanistic insights and potential therapeutic targets for TDP-43 dysfunction.

molecular biology↗

bioRxiv2025

ERC 2.0 - evolutionary rate covariation update provides more powerful inference of functional interactions across large phylogenies

Evolutionary Rate Covariation (ERC) is an established comparative genomics method that identifies sets of genes sharing patterns of sequence evolution, which suggests shared function. Whereas many functional predictions of ERC have been empirically validated, its predictive power has hitherto been limited by its inability to tackle the large numbers of species in contemporary comparative genomics datasets. This study introduces ERC2.0, an enhanced methodology for studying ERC across phylogenies with hundreds of species and tens of thousands of genes. ERC2.0 improves upon previous iterations of ERC in algorithm speed, normalizing for heteroskedasticity, and normalizing correlations via Fisher transformations. These improvements have resulted in greater statistical power to predict biological function. In exemplar yeast and mammalian datasets, we demonstrate that the predictive power of ERC2.0 is improved relative to the previous method, ERC1.0, and that further improvements are obtained by using larger yeast and mammalian phylogenies. We attribute the improvements to both the larger datasets and improved rate normalization. We demonstrate that ERC2.0 has high predictive accuracy for known annotations and can predict the functions of genes in non-model systems. Our findings underscore the potential for ERC2.0 to be used as a single-pass computational tool in candidate gene screening and functional predictions.

evolutionary biology↗

bioRxiv2025

A scalable approach to investigating sequence-to-expression prediction from personal genomes

Sequence-to-function (S2F) models hold the promise of evaluating arbitrary DNA sequences, providing a powerful framework for linking genotype to phenotype. Yet, despite strong performance across genomic loci, these models often struggle to capture inter-individual variation in gene expression. To address this, we propose personal genome training--training models to make genotype-specific predictions at a single locus. We introduce SAGE-net, a scalable framework and software package for training and evaluating S2F models using personal genomes. Using SAGE-net, we systematically explore model architectures and training regimes, showing that personal genome training improves gene expression prediction accuracy for held-out individuals. However, performance gains arise primarily from identifying predictive variants, rather than learning a cis-regulatory grammar that generalizes across loci. This lack of generalization persists across a wide range of hyperparameters. In contrast, when applied to DNA methylation (DNAm), personal genome training enables improved generalization to unseen individuals in unseen genomic regions. This suggests that S2F models may more readily capture the sequence-level determinants of inter-individual variation in epigenomic traits. These findings highlight the need for further exploration to unlock the full potential of S2F models in decoding the regulatory grammar of personal genomes. Scalable software and infrastructure development will be critical to this progress.

genomics↗

bioRxiv2024

Minus the Error: Estimating dN/dS and Testing for Natural Selection in the Presence of Residual Alignment Errors

Positive selection is an evolutionary process which increases the frequency of advantageous mutations because they confer a fitness benefit. Inferring the past action of positive selection on protein-coding sequences is fundamental for deciphering phenotypic diversity and the emergence of novel traits. With the advent of genome-wide comparative genomic datasets, researchers can analyze selection not only at the level of individual genes but also globally, delivering systems-level insights into evolutionary dynamics. However, genome-scale datasets are generated with automated pipelines and imperfect curation that does not eliminate all sequencing, annotation, and alignment errors. Positive selection inference methods are highly sensitive to such errors. We present BUSTED-E: a method designed to detect positive selection for amino acid diversification while concurrently identifying some alignment errors. This method builds on the flexible branch-site random effects model (BUSTED) for fitting distributions of dN/dS, with a critical modification: it incorporates an "error-sink" component to represent an abiological evolutionary regime. Using several genome-scale biological datasets that were extensively filtered using state-of-the art automated alignment tools, we show that BUSTED-E identifies pervasive residual alignment errors, produces more realistic estimates of positive selection, reduces bias, and improves biological interpretation. The BUSTED-E model promises to be a more stringent filter to identify positive selection in genome-wide contexts, thus enabling further characterization and validation of the most biologically relevant cases.

evolutionary biology↗

bioRxiv2024

Ultra-fast variant effect prediction using biophysical transcription factor binding models

Sequence variation within TF binding sites can significantly affect TF-DNA interactions, influencing gene expression and contributing to disease susceptibility or phenotypic traits. Despite recent progress in deep sequence-to-function models that predict functional output from sequence data, these methods perform inadequately on some variant effect prediction tasks, especially with common genetic variants. This limitation underscores the importance of leveraging biophysical models of TF binding to enhance interpretability of variant effect scores and facilitate mechanistic insights. We introduce motifDiff, a novel computational tool designed to quantify variant effects using mono and dinucleotide position weight matrices. motifDiff offers several key advantages, including scalability to score millions of variants within minutes, implementation of statistically rigorous normalization strategy critical for optimal performance, and support for both dinucleotide and mononucleotide models. We demonstrate motifDiff s efficacy by evaluating it across diverse ground truth datasets that quantify the effects of common variants in vivo, thereby establishing robust benchmarks for the predictive value of variant effect calculations. Finally, we show that our tool provides unique insights when interpreting human accelerated regions. motifDiff is available as a standalone Python application at https://github.com/rezwanhosseini/MotifDiff.

bioinformatics↗

bioRxiv2023

RERconverge Expansion: Using Relative Evolutionary Rates to Study Complex Categorical Trait Evolution

Comparative genomics approaches seek to associate evolutionary genetic changes with the evolution of phenotypes across a phylogeny. Many of these methods, including our evolutionary rates based method, RERconverge, lack the capability of analyzing non-ordinal, multicategorical traits. To address this limitation, we introduce an expansion to RERconverge that associates shifts in evolutionary rates with the convergent evolution of multi-categorical traits. The categorical RERconverge expansion includes methods for performing categorical ancestral state reconstruction, statistical tests for associating relative evolutionary rates with categorical variables, and a new method for performing phylogenetic permulations on multi-categorical traits. In addition to demonstrating our new method on a three-category diet phenotype, we compare its performance to naive pairwise binary RERconverge analyses and two existing methods for comparative genomic analyses of categorical traits: phylogenetic simulations and a phylogenetic signal based method. We also present a diagnostic analysis of the new permulations approach demonstrating how the method scales with the number of species and the number of categories included in the analysis. Our results show that our new categorical method outperforms phylogenetic simulations at identifying genes and enriched pathways significantly associated with the diet phenotype and that the new ancestral reconstruction drives an improvement in our ability to capture diet-related enriched pathways. Our categorical permulations were able to account for non-uniform null distributions and correct for non-independence in gene rank during pathway enrichment analysis. The categorical expansion to RERconverge will provide a strong foundation for applying the comparative method to categorical traits on larger data sets with more species and more complex trait evolution.

bioinformatics↗

bioRxiv2023

Quick and effective approximation of in silico saturation mutagenesis experiments with first-order Taylor expansion

To understand the decision process of genomic sequence-to-function models, various explainable AI algorithms have been proposed. These methods determine the importance of each nucleotide in a given input sequence to the models predictions, and enable discovery of cis regulatory motif grammar for gene regulation. The most commonly applied method is in silico saturation mutagenesis (ISM) because its per-nucleotide importance scores can be intuitively understood as the computational counterpart to in vivo saturation mutagenesis experiments. While ISM is highly interpretable, it is computationally challenging to perform, because it requires computing three forward passes for every nucleotide in the given input sequence; these computations add up when analyzing a large number of sequences, and become prohibitive as the length of the input sequences and size of the model grows. Here, we show how to use the first-order Taylor approximation to compute ISM, which reduces its computation cost to a single forward pass for an input sequence. We use our theoretical derivation to connect ISM with the gradient of the model and show how this approximation is related to a recently suggested correction of the models gradients for genomic sequence analysis. We show that the Taylor ISM (TISM) approximation is robust across different model ablations, random initializations, training parameters, and data set sizes.

bioinformatics↗