Biology subjects

Oluwadare, O.

Publications and source records attributed to Oluwadare, O..

12 recordsLinked to original sources

bioRxiv2026

HiC-LEGO: Biologically Guided High-Resolution 3D Genome Reconstruction Preserves Chromatin Organization at Kilobase Resolution

Three-dimensional (3D) chromosome reconstruction from Hi-C contact maps remains challenging because genome organization is hierarchical and fine-resolution models must reconcile local structure with chromosome-scale constraints. Here we present HiC-LEGO, a domain-aware hierarchical framework integrating ensemble chromatin domains with graph-based structural learning and progressive chromosome assembly. By combining ensemble domain selection with hierarchical reconstruction, HiC-LEGO reduces dependence on individual domain definitions while maintaining local organization during chromosome-scale assembly. Across five human cell lines at 5-kb resolution, HiC-LEGO achieves higher reconstruction concordance than evaluated state-of-the-art methods while better preserving domain organization. At 1-kb resolution, HiC-LEGO reconstructs complete GM12878 chromosome 8 and recovers close spatial proximity between an epigenomically supported distal MYC enhancer and its promoter. Reconstructions from 5-kb Micro-C data show that 249 experimentally defined RCMC microcompartment interactions at the Ppm1g locus occupy compact 3D configurations. In the breast cancer dataset, HiC-LEGO reconstructs structures that maintain stable TAD organization across healthy breast, primary tumors, and liver metastases, while revealing greater inter-patient structural heterogeneity in malignant pleural effusion samples. Pore-C validation shows that experimentally observed multi-way contacts spanning 1-5 Mb are enriched in compact reconstructed configurations across all 23 chromosomes. Thus, HiC-LEGO preserves regulatory interactions, disease-associated chromatin organization and higher-order spatial relationships beyond pairwise contact-map concordance.

bioinformatics↗

bioRxiv2026

HiCP2GAN: A Plug and Play Foundation Model-based GAN for Hi-C Enhancement

The three-dimensional organization of chromatin shapes gene regulation and cellular function. Hi-C has emerged as the primary technique for mapping chromatin interactions genome-wide, yet high-resolution data remain costly and scarce, leaving many studies with sparse contact maps that limit downstream analysis. Deep learning methods, especially generative adversarial networks (GANs), have shown promise for enhancing low-resolution Hi-C data. Most existing GAN-based approaches, however, rely on custom discriminators trained from scratch, which can yield unstable training and limited generalization. Hi-C foundation models pretrained on large-scale data capture rich, transferable representations of chromatin structure; their use as discriminators within adversarial enhancement frameworks has not been explored. In this work, we introduce HiCP2GAN, a plug-and-play GAN that employs a pretrained Vision Transformer-based Hi-C foundation model as its discriminator. The discriminator was pretrained on 118 million Hi-C patches across diverse species and cell types, providing biologically meaningful gradients for adversarial supervision. The HiCP2GAN framework is generator-agnostic: any compatible Hi-C resolution enhancement architecture can serve as the generator, enabling fair comparison across methods. The encoder phase of the foundation model was adapted as a discriminator backbone and experimented with finetuning different numbers of layers from the input while freezing the deeper transformer layers. Finetuning the first few layers while freezing the rest preserved pretrained knowledge while allowing task-specific adaptation. Experiments on human cell lines show that HiCP2GAN consistently improves resolution over standalone generators and conventional GAN-based models, while serving as a plug-and-play framework for most non-GAN generator models. HiCP2GAN is publicly available at https://github.com/OluwadareLab/HiCP2GAN.

bioinformatics↗

bioRxiv2026

SpliceRead: Improving Canonical and Non-Canonical Splice Site Prediction with Residual Blocks and Synthetic Data Augmentation

Accurate splice site prediction is fundamental to understanding gene expression and its associated disorders. However, most existing models are biased toward frequent canonical sites, limiting their ability to detect rare but biologically important non-canonical variants. These models often rely heavily on large, imbalanced datasets that fail to capture the sequence diversity of non-canonical sites, leading to high false-negative rates. Here, we present SpliceRead, a novel deep learning model designed to improve the classification of both canonical and non-canonical splice sites using a combination of residual convolutional blocks and synthetic data augmentation. SpliceRead employs a data augmentation method to generate diverse non-canonical sequences and uses residual connections to enhance gradient flow and capture subtle genomic features. Trained and tested on a multi-species dataset of 400- and 600-nucleotide sequences, SpliceRead consistently outperforms state-of-the-art models across all key metrics, including F1-score, accuracy, precision, and recall. Notably, it achieves a substantially lower non-canonical misclassification rate than baseline methods. Extensive evaluations, including cross-validation, cross-species testing, and input-length generalization, confirm its robustness and adaptability. SpliceRead offers a powerful, generalizable framework for splice site prediction, particularly in challenging, low-frequency sequence scenarios, and paves the way for more accurate gene annotation in both model and non-model organisms.The open sourced code of SpliceRead and a detailed documentation is available at The open-sourced code of SpliceRead and detailed documentation are available at https://github.com/OluwadareLab/SpliceRead.

bioinformatics↗

bioRxiv2026

HiCInterpolate: 4D Spatiotemporal Interpolation of Hi-C Data for Genome Architecture Analysis.

MotivationStudying the three-dimensional (3D) structure of a genome, including chromatin loops and Topologically Associating Domains (TADs), is essential for understanding how the genome is organized, such as gene activation, cell development, protein-protein interaction, etc. Hi-C protocol enables us to study 3D genome structure and organization. Chromatin 3D structure changes dynamically over time, and modeling these continuous changes is crucial for downstream analysis in various domains such as disease diagnosis, vaccine development, etc. The high expense and impracticality of continuous genome sequencing, particularly what evolves between two timestamps, limit the most effective genomic analysis. It is crucial to develop a straightforward and cost-efficient method for constantly generating genomic data between two timestamps in order to address these constraints. ResultsIn this study, we developed HiCInterpolate, a 4D spatiotemporal interpolation architecture that accepts two timestamp Hi-C contact matrices to interpolate intermediate Hi-C contact matrices at high resolution. HiCInterpolate predicts the intermediate Hi-C contact map using a deep learning-based flow predictor, and a feature encoder and decoder architecture similar to U-Net. In addition, HiCInterpolate supports downstream analysis of multiple 3D genomic features, including A/B compartments, chromatin loops, TADs, and 3D genome structure, through an integrated analysis pipeline. Across multiple evaluation metrics, including PSNR, SSIM, GenomeDISCO, HiCRep, and LPIPS, HiCInterpolate achieved consistently strong performance. Biological validation further demonstrated preservation of key chromatin organization features, such as chromatin loops, A/B compartments, and TADs. Together, these results indicate that HiCInterpolate provides a robust computer vision-based framework for high-resolution interpolation of intermediate Hi-C contact matrices and facilitates biologically meaningful downstream analyses. AvailabilityHiCInterpolate is publicly available at https://github.com/OluwadareLab/HiCInterpolate.

bioinformatics↗

bioRxiv2026

ASPECT: Alternative Splicing Event Classification with Transformers

MotivationAlternative splicing (AS) is a fundamental regulatory mechanism that expands transcriptomic and proteomic diversity by generating multiple mRNA isoforms from a single gene. Aberrant AS has been implicated in numerous diseases through the production of dysfunctional or pathogenic protein variants. However, much of the existing AS research has focused predominantly on exon skipping and constitutive splicing, with comparatively limited attention to other biologically relevant AS event types. Moreover, many current computational approaches rely on short genomic sequence windows and conventional deep learning architectures, which may limit their ability to capture broader regulatory context associated with complex splicing decisions. Bridging these methodological and conceptual gaps is essential for advancing comprehensive AS characterization and improving our understanding of its role in human health and disease. ResultsWe present ASPECT, an alternative splicing event classification framework built upon DNABERT-2 with Byte Pair Encoding (BPE) tokenization. Across multiple binary alternative splicing event pair classification tasks, ASPECT achieves consistently strong performance as measured by AUC, F1-score,and accuracy, demonstrating reliable discrimination between closely related splicing event types. Importantly, ASPECT demonstrates consistent performance when applied to TCGA BRCA cancer-associated splicing events reconstructed from SpliceSeq annotations, supporting its applicability beyond the canonical splicing events used for training. AvailabilityThe open-source code, data, and detailed documentation used in this study are available at https://github.com/OluwadareLab/ASPECT. ContactOluwatosin.Oluwadare@unt.edu Supplementary informationN/A

bioinformatics↗

bioRxiv2024

ScHiCAtt: Enhancing Single-Cell Hi-C Resolution Using Attention-Based Models

The spatial organization of chromatin is fundamental to gene regulation and essential for proper cellular function. The Hi-C technique remains the leading method for unraveling 3D genome structures; however, limited resolution, data sparsity, and incomplete coverage in single-cell Hi-C data pose significant challenges for comprehensive analysis. Traditional CNN-based models often suffer from blurring and loss of fine details, while GAN-based methods encounter difficulties in maintaining diversity and generalization. Moreover, existing algorithms perform poorly in cross-cell line generalization, where a model trained on one cell type is used to enhance high-resolution data in another cell type. To address these limitations, we propose ScHiCAtt (Single-cell Hi-C Attention-Based Model), which leverages attention mechanisms to capture both long-range and local dependencies in Hi-C data, significantly enhancing resolution while preserving biologically meaningful interactions. We implement this mechanism and check its validity on data from different cells of the same organisms and data of different organisms. By dynamically focusing on regions of interest, attention mechanisms effectively mitigate data sparsity and enhance model performance in low-resolution contexts. Extensive experiments on Human and Drosophila single-cell Hi-C data demonstrate that ScHiCAtt consistently outperforms existing methods in terms of computational and biological reproducibility metrics across different downsampling ratios, especially under extreme downsampling conditions. The model is publicly available at https://github.com/OluwadareLab/ScHiCAtt.

bioinformatics↗

bioRxiv2023

Comparative study on chromatin loop callers using Hi-C data reveals their effectiveness

The chromosome is a fundamental component of cell biology, housing DNA that encapsulates hierarchical genetic information. DNA compresses its size by forming loops, and these loop regions contain numerous protein particles, including CTCF, SMC3, H3 histone, and Topologically Associating Domains (TADs). In this study, we conducted a comprehensive study of 22 loop calling methods. Additionally, we have provided detailed insights into the methodologies underlying these algorithms for loop detection, categorizing them into five distinct groups based on their fundamental approaches. Furthermore, we have included critical information such as resolution, input and output formats, and parameters. For this analysis, we utilized the primary and replicate GM12878 Hi-C datasets at 5KB and 10KB resolutions. Our evaluation criteria encompassed various factors, including loop count, reproducibility, overlap, running time, Aggregated Peak Analysis (APA), and recovery of protein-specific sites such as CTCF, H3K27ac, and RNAPII. This analysis offers insights into the loop detection processes of each method, along with the strengths and weaknesses of each, enabling readers to effectively choose suitable methods for their datasets. We evaluate the capabilities of these tools and introduce a novel Biological, Consistency, and Computational robustness score (BCCscore) to measure their overall robustness ensuring a comprehensive evaluation of their performance.

bioinformatics↗

bioRxiv2021

HiC-GNN: A Generalizable Model for 3D Chromosome Reconstruction Using Graph Convolutional Neural Networks

Chromosome conformation capture (3C) is a method of measuring chromosome topology in terms of loci interaction. The Hi-C method is a derivative of 3C that allows for genome wide quantification of chromosome interaction. From such interaction data, it is possible to infer the three-dimensional (3D) structure of the underlying chromosome. In this paper, we use a node embedding algorithm and a graph neural network to predict the 3D coordinates of each genomic loci from the corresponding Hi-C contact data. Unlike other chromosome structure prediction methods, our method can generalize a single model across Hi-C resolutions, multiple restriction enzymes, and multiple cell populations while maintaining reconstruction accuracy. We derive these results using three separate Hi-C data sets from the GM12878, GM06990, and K562 cell lines. We also compare the reconstruction accuracy of our method to four other existing methods and show that our method yields superior performance. Our algorithm outperforms the state-of-the-art methods in the accuracy of prediction and introduces a novel method for 3D structure prediction from Hi-C data. Author SummaryWe developed a novel method, HiC-GNN, for predicting the three-dimensional structures of chromosomes from Hi-C data. HiC-GNN is unique from other methods for chromosome structure prediction in that it learns in an eager setting rather than a lazy setting. Thus, the models learned by HiC-GNN can be generalized to unseen data. To the authors knowledge, this generalizing capability is not present in any existing methods. We show that this generalization is robust to input resolution, restriction enzyme, and contact sparsity. We also show that our method outperforms existing methods using both generalized and non-generalized models. Moreover, we also show that our method is more robust to contact variance than the compared methods. AvailabilityAll our source codes and data are available at https://github.com/OluwadareLab/HiC-GNN, and is made available as a containerized application that can be run on any platform.

bioinformatics↗

bioRxiv2021

ChromeBat: A Bio-Inspired Approach to 3D Genome Reconstruction

BackgroundWith the advent of Next Generation Sequencing and the Hi-C experiment, high quality genome-wide contact data is becoming increasingly available. This data represents an empirical measure of how a genome interacts inside the nucleus. Genome conformation is of particular interest as it has been experimentally shown to be a driving force for many genomic functions from regulation to transcription. Thus, the Three Dimensional Genome Reconstruction Problem seeks to take Hi-C data and produce the complete physical genome structure as it appears in the nucleus for genomic analysis. ResultsWe propose and develop a novel method to solve the Chromosome and Genome Reconstruction problem based on the Bat Algorithm which we called ChromeBat. We demonstrate on real Hi-C data that ChromeBat is capable of state of the art performance. Additionally, the domain of Genome Reconstruction has been criticized for lacking algorithmic diversity, and the bio-inspired nature of ChromeBat contributes algorithmic diversity to the problem domain. ConclusionsChromeBat is an effective approach at solving the Genome Reconstruction Problem. The source code and usage guide can be found here: https://github.com/OluwadareLab/ChromeBat.

bioinformatics↗

bioRxiv2021

ParticleChromo3D: A Particle Swarm Optimization Algorithm for Chromosome and Genome 3D Structure Prediction from Hi-C Data

The three-dimensional (3D) structure of chromatin has a massive effect on its function. Because of this, it is desirable to have an understanding of the 3D structural organization of chromatin. To gain greater insight into the spatial organization of chromosomes and genomes and the functions they perform, chromosome conformation capture techniques, particularly Hi-C, have been developed. The Hi-C technology is widely used and well-known because of its ability to profile interactions for all read pairs in an entire genome. The advent of Hi-C has greatly expanded our understanding of the 3D genome, genome folding, gene regulation and has enabled the development of many 3D chromosome structure reconstruction methods. Here, we propose a novel approach for 3D chromosome and genome structure reconstruction from Hi-C data using Particle Swarm Optimization approach called ParticleChromo3D. This algorithm begins with a grouping of candidate solution locations for each chromosome bin, according to the particle swarm algorithm, and then iterates its position towards a global best candidate solution. While moving towards the optimal global solution, each candidate solution or particle uses its own local best information and a randomizer to choose its path. Using several metrics to validate our results, we show that ParticleChromo3D produces a robust and rigorous representation of the 3D structure for input Hi-C data. We evaluated our algorithm on simulated and real Hi-C data in this work. Our results show that ParticleChromo3D is more accurate than most of the existing algorithms for 3D structure reconstruction. Our results also show that constructed ParticleChromo3D structures are very consistent, hence indicating that it will always arrive at the global solution at every iteration. The source code for ParticleChromo3D, the simulated and real Hi-C datasets, and the models generated for these datasets are available here: https://github.com/OluwadareLab/ParticleChromo3D

bioinformatics↗

bioRxiv2019

Deep Learning For Denoising Hi-C Chromosomal Contact Data

MotivationThe three-dimensional (3D) organization of an organisms genome and chromosomes plays a significant role in many biological processes. Currently, methods exist for modeling chromosomal 3D structure using contact matrices generated via chromosome conformation capture (3C) techniques such as Hi-C. However, the effectiveness of these methods is inherently bottlenecked by the quality of the Hi-C data, which may be corrupted by experimental noise. Consequently, it is valuable to develop methods for eliminating the impact of noise on the quality of reconstructed structures.\n\nResultsWe develop unsupervised and semi-supervised deep learning algorithms (i.e. deep convolutional autoencoders) to denoise Hi-C contact matrix data and improve the quality of chromosome structure predictions. When applied to noisy synthetic contact matrices of the yeast genome, our network demonstrates consistent improvement across metrics for contact matrix similarity including: Pearson Correlation, Spearman Correlation and Signal-to-Noise Ratio. Positive improvement across these metrics is seen consistently across a wide space of parameters to both gaussian and poisson noise functions.\n\nContactmrh8x5@mail.missouri.edu and chengji@missouri.edu

genomics↗

bioRxiv2018

Hierarchical Reconstruction of High-Resolution 3D Models of Human Chromosomes

Eukaryotic chromosomes are often composed of components organized into multiple scales, such as nucleosomes, chromatin fibers, topologically associated domains (TAD), chromosome compartments, and chromosome territories. Therefore, reconstructing detailed 3D models of chromosomes in high resolution is useful for advancing genome research. However, the task of constructing quality highresolution 3D models is still challenging with existing methods. Hence, we designed a hierarchical algorithm, called Hierarchical3DGenome, to reconstruct 3D chromosome models at high resolution (<=5 Kilobase (KB)). The algorithm first reconstructs high-resolution 3D models at TAD level. The TAD models are then assembled to form complete high-resolution chromosomal models. The assembly of TAD models is guided by a complete low-resolution chromosome model. The algorithm is successfully used to reconstruct 3D chromosome models at 5KB resolution for the human B-cell (GM12878). These high-resolution models satisfy Hi-C chromosomal contacts well and are consistent with models built at lower (i.e. 1MB) resolution, and with the data of fluorescent in situ hybridization experiments. The Java source code of Hierarchical3DGenome and its user manual are available here https://github.com/BDM-Lab/Hierarchical3DGenome.

bioinformatics↗