Biology subjects

Murtaza, G.

Publications and source records attributed to Murtaza, G..

10 recordsLinked to original sources

bioRxiv2026

scWeave: A deep learning model that bidirectionally translates between gene expression and chromatin structure at single cell resolution

Chromatin structure and gene expression are intimately linked, yet characterizing how the two covary has proven challenging, primarily because the two modalities are rarely measured in the same cells. Recently, single-cell co-assay protocols have enabled simultaneous profiling of both modalities within the same cells, but these experiments remain costly and technically challenging. To better characterize the relationship between 3D chromatin architecture and gene expression and to enable cross-modality inference from single-modality measurements, we developed a model called scWeave that bidirectionally translates between gene expression (scRNA-seq) and 3D chromatin architecture (scHi-C) at single-cell resolution. The scWeave model employs dual autoencoders to extract separate cell-level latent representations and learns to translate between these representations using dedicated translation modules. We evaluate scWeave on six publicly available co-assay datasets spanning mouse embryonic development, mouse cortex, mouse olfactory epithelium, and human bone marrow. On held-out mouse cells, scWeave outperforms a nearest-neighbor baseline and existing methods adapted to single-cell resolution, achieving a 57% improvement in median Spearman correlation when predicting gene expression from chromatin structure and an 18.8% improvement in median HiCRep similarity in the reverse direction relative to the next-best baseline. We further show that scWeave learns cross-modally aligned latent representations at single-cell resolution, enabling cells profiled in one modality to be matched to their counterparts in the other. Finally, scWeave generalizes to entirely held-out developmental timepoints in mouse olfactory epithelium and performs well on held-out human bone marrow cells despite limited human training data. By predicting the unmeasured chromatin architecture or transcriptional state from a single measured modality, scWeave offers a route to extend the benefits of costly co-assays to the many cell types, developmental stages, and species that are currently profiled with only one modality.

genomics↗

bioRxiv2025

Puget predicts gene expression across cell types using sequence and 3D chromatin organization data

Gene expression is governed by both linear DNA sequence and three-dimensional (3D) chromatin architecture. Most gene expression prediction models rely on sequence alone, thereby failing to capture structural context and to generalize to unseen cell types. We present Puget, a deep learning model that predicts cell type-specific gene expression from sequence and Hi-C data, which captures 3D chromatin organization. Puget pairs pretrained sequence and Hi-C encoders with a lightweight transformer decoder. Using paired Hi-C/RNA-seq from 36 human and 4 mouse biosamples, we evaluate Pugets ability to generalize to held-out genes, held-out biosamples, and from human to mouse. Relative to a sequence-only baseline, Puget improves cross-biosample Pearson correlation by up to 25% on highly variable genes in training biosamples and, unlike the sequence-only model, generalizes to held-out biosamples and across species. In addition, in silico perturbation experiments show that Puget can prioritize experimentally validated enhancer-gene pairs. Together, these results highlight a generalizable approach for modeling gene expression from sequence and 3D chromatin organization.

bioinformatics↗

bioRxiv2025

Evo2HiC: a multimodal foundation model for integrative analysis of genome sequence and architecture

Understanding how genomic sequences shape three-dimensional (3D) genome architecture is funda-mental to interpreting diverse biological processes. Although previous studies have shown that sequence information can predict 3D genome architecture, they fall short in capturing cell type-specific structures because they are trained solely on sequence inputs. The widely available Hi-C data, which contain rich structural information across biosamples, can provide complementary features to sequence data for study-ing cell type-specific architectures. Recently, DNA foundation models have demonstrated encouraging performance in capturing long-range genomic dependencies, holding promise for modeling chromatin interactions. However, the extremely high computational cost of running these models limits their applicability to Hi-C analysis, which requires genome-wide sequence embeddings. Here, we present Evo2HiC, a multimodal foundation model that jointly models genomic sequences and structures to study cell type-specific chromatin structure. The key idea of Evo2HiC is to distill a large-scale DNA foundation model, Evo 2 (7B), into a compact encoder, while guiding the distillation with Hi-C data to preserve genomic features critical for 3D genome analysis. The model supports two types of encoders, one that operates directly on DNA sequences, and a second that additionally takes as input corresponding Hi-C data. Using the DNA-only encoder and predicting Hi-C contact matrices, Evo2HiC improved Spearman correlation by 10.9% over Orca. Moreover, by jointly embedding Hi-C and sequence information Evo2HiC achieved the best overall Pearson correlation when predicting five representative epigenomic assays. Interpretation analysis of Evo2HiC revealed its ability to identify cell type-specific sequence motifs that explain changes in epigenomic signals. Finally, we demonstrated the cross-species generalizability of Evo2HiC on 177 species from the DNA Zoo dataset for Hi-C resolution enhancement. In summary, Evo2HiC is a foundation model that integrates genome sequences and 3D chromatin structure information, substantially reduces computational cost while maintaining state-of-the-art accuracy on predicting various epigenomic signals and genome architecture, enables the identification of cell type-specific motifs, and demonstrates robust generalizability across species.

bioinformatics↗

bioRxiv2025

Upregulation of interleukin-1β and interleukin-18 in traumatic brain injury patients and their potential as biomarkers

BackgroundTraumatic brain injury (TBI) is diagnosed using radiological imaging and biomarker analysis, each with certain limitations. Accurate and accessible detection of TBI is essential for its treatment. Since interleukin (IL)-1{beta} and IL-18 are key neuroinflammatory cytokines, we aim to investigate the regulation of IL-1{beta} and IL-18 in TBI patients and assess their potential as biomarkers of TBI. MethodsForty TBI and 40 healthy subjects were recruited for this study. The Glasgow coma scale (GCS) was used to assess the clinical severity of TBI. Spearmans rank-order correlation was employed to find the association between GCS scores and IL levels. An enzyme-linked immunosorbent assay (ELISA) with human IL-1{beta} and IL-18 ELISA kits was conducted to measure the serum IL-1{beta} and IL-18 levels. The area under the curve (AUC), sensitivity, and specificity were measured to assess the potential of IL-1{beta} and IL-18 as biomarkers. ResultsThe patients had mild TBI (mTBI) with a mean GCS score of 11.05 {+/-} 1.67. A negative correlation was found between GCS scores and IL levels. IL-1{beta} levels in the TBI and healthy subjects were 1882.19 {+/-} 796.72 pg/mL and 1473.50 {+/-} 333.045 pg/mL, respectively, with a significant difference (p = 0.001). Moreover, the IL-18 levels in TBI subjects (25.92 {+/-} 14.81 ng/L) were also significantly higher (p = 0.000) compared to those of the control group subjects (19.14 {+/-} 2.99 ng/L). The IL-1{beta} exhibited an AUC of 0.72 with 72.5% sensitivity and 70% specificity, at a cut-off point of 1527.25 (pg/mL), while IL-18 had an AUC of 0.78 with 77.5% sensitivity and 72.7% specificity, at a cut-off point of 19.645 (ng/L). ConclusionsIL-1{beta} and IL-18 levels were increased remarkably in mTBI patients and exhibited fair potential as biomarkers for the diagnosis and prognosis of TBI.

neuroscience↗

bioRxiv2025

HypoAD: volumetric and single-cell analysis reveals changes in the human hypothalamus in aging and Alzheimer's disease

Alterations in metabolism, stress response, sleep, circadian rhythms, and neuroendocrine processes are key features of aging and neurodegeneration. These fundamental processes are regulated by the hypothalamus, yet how its functionally distinct subregions and cell types change during human aging and Alzheimers Disease (AD) remains largely unexplored. Here, we present HypoAD, a comprehensive atlas of the human hypothalamus in aging and AD, integrating high-resolution MRI from 202 individuals with single-nucleus RNA-seq (snRNA-seq) of 614,403 nuclei from young, AD, and age-matched non-dementia controls. Our analysis reveals that hypothalamic subregions governing metabolism, stress, and circadian rhythms are particularly vulnerable, exhibiting significant changes in both volumes and gene expression during aging and AD. At the molecular level, machine learning models identified the inflammatory response and regulators of circadian rhythms as key cellular predictors of AD. These signatures were reflected in specific cell types: microglia transitioned to a pro-inflammatory state, while inhibitory neurons within sleep-and circadian-regulating hypothalamic subregions showed the most profound transcriptional alterations, including disruptions in ligand-receptor interactions and G-protein-coupled receptor signaling. Together, HypoAD provides a high-resolution volumetric map and a comprehensive transcriptomic atlas of the human hypothalamus in aging and AD, linking lifestyle and behavioral changes to their underlying volumetric and molecular pathways. Additionally, HypoAD provides a framework to investigate hypothalamic dysfunction and establishes a roadmap for targeted interventions aimed at mitigating physiological disruptions to potentially slow disease progression.

neuroscience↗

bioRxiv2025

scContrast: A contrastive learning based approach for encoding single-cell gene expression data

Single-cell RNA sequencing (scRNA-seq) captures gene expression at a individual cell resolution, which reveals critical insights into cellular diversity, disease processes, and developmental biology. However, a key challenge in scRNA-seq analysis is clustering similar cells across multiple batches, particularly when distinct sequencing protocols are used. In this work, we present scContrast, a semi-supervised contrastive learning method tailored for embedding scRNA-seq data from both plate- and droplet-based protocols into a universal representation space. By leveraging five simple augmentations, scContrast extracts biologically relevant signals from gene expression data while filtering out batch effects and technical artifacts. We trained scContrast on a subset of Tabula Muris tissues and evaluated its zero-shot performance on unseen tissues. Our results demonstrate that scContrast generalizes effectively to new tissues and outperforms the leading UCE approach in integrating scRNA-seq data from droplet- and plate-based sequencing protocols.

genomics↗

bioRxiv2024

scGrapHiC: Deep learning-based graph deconvolution for Hi-C using single cell gene expression

Single-cell Hi-C (scHi-C) protocol helps identify cell-type-specific chromatin interactions and sheds light on cell differentiation and disease progression. Despite providing crucial insights, scHi-C data is often underutilized due the high cost and the complexity of the experimental protocol. We present a deep learning framework, scGrapHiC, that predicts pseudo-bulk scHi-C contact maps using pseudo-bulk scRNA-seq data. Specifically, scGrapHiC performs graph deconvolution to extract genome-wide single-cell interactions from a bulk Hi-C contact map using scRNA-seq as a guiding signal. Our evaluations show that scGrapHiC, trained on 7 cell-type co-assay datasets, outperforms typical sequence encoder approaches. For example, scGrapHiC achieves a substantial improvement of 23.2% in recovering cell-type-specific Topologically Associating Domains over the baselines. It also generalizes to unseen embryo and brain tissue samples. scGrapHiC is a novel method to generate cell-type-specific scHi-C contact maps using widely available genomic signals that enables the study of cell-type-specific chromatin interactions. Availabilityhttps://github.com/rsinghlab/scGrapHiC Contactritambhara@brown.edu

genomics↗

bioRxiv2022

GrapHiC: An integrative graph based approach for imputing missing Hi-C reads

Hi-C experiments allow researchers to study and understand the 3D genome organization and its regulatory function. Unfortunately, sequencing costs and technical constraints severely restrict access to high-quality Hi-C data for many cell types. Existing frameworks rely on a sparse Hi-C dataset or cheaper-to-acquire ChIP-seq data to predict Hi-C contact maps with high read coverage. However, these methods fail to generalize to sparse or cross-cell-type inputs because they do not account for the contributions of epigenomic features or the impact of the structural neighborhood in predicting Hi-C reads. We propose GrapHiC, which combines Hi-C and ChIP-seq in a graph representation, allowing more accurate embedding of structural and epigenomic features. Each node represents a binned genomic region, and we assign edge weights using the observed Hi-C reads. Additionally, we embed ChIP-seq and relative positional information as node attributes, allowing our representation to capture structural neighborhoods and the contributions of proteins and their modifications for predicting Hi-C reads. Our evaluations show that GrapHiC generalizes better than the current state-of-the-art on cross-cell-type settings and sparse Hi-C inputs. Moreover, we can utilize our framework to impute Hi-C reads even when no Hi-C contact map is available, thus making high-quality Hi-C data more accessible for many cell types. Availabilityhttps://github.com/rsinghlab/GrapHiC ACM Reference FormatGhulam Murtaza, Justin Wagner, Justin M. Zook, and Ritambhara Singh. 2018. GrapHiC: An integrative graph based approach for imputing missing Hi-C reads. In Proceedings of 22nd International Workshop on Data Mining in Bioinformatics (BioKDD 23). ACM, New York, NY, USA, 16 pages. https://doi.org/XXXXXXX.XXXXXXX

genomics↗

bioRxiv2022

Investigating the performance of deep learning methods for Hi-C resolution improvement

MotivationHi-C is a widely used technique to study the 3D organization of the genome. Due to its high sequencing cost, most of the generated datasets are of coarse resolution, which makes it impractical to study finer chromatin features such as Topologically Associating Domains (TADs) and chromatin loops. Multiple deep-learning-based methods have recently been proposed to increase the resolution of these data sets by imputing Hi-C reads (typically called upscaling). However, the existing works evaluate these methods on either synthetically downsampled or a small subset of experimentally generated sparse Hi-C datasets, making it hard to establish their generalizability in the real-world use case. We present our framework - Hi-CY - that compares existing Hi-C resolution upscaling methods on seven experimentally generated low-resolution Hi-C datasets belonging to various levels of read sparsities originating from three cell lines on a comprehensive set of evaluation metrics. Hi-CY also includes four downstream analysis tasks, such as TAD and chromatin loops recall, to provide a thorough report on the generalizability of these methods. ResultsWe observe that existing deep-learning methods fail to generalize to experimentally generated sparse Hi-C datasets showing a performance reduction of up to 57 %. As a potential solution, we find that retraining deep-learning based methods with experimentally generated Hi-C datasets improves performance by up to 31%. More importantly, Hi-CY shows that even with retraining, the existing deep-learning based methods struggle to recover biological features such as chromatin loops and TADs when provided with sparse Hi-C datasets. Our study, through Hi-CY framework, highlights the need for rigorous evaluation in future. We identify specific avenues for improvements in the current deep learning-based Hi-C upscaling methods, including but not limited to using experimentally generated datasets for training. Availabilityhttps://github.com/rsinghlab/Hi-CY Author SummaryWe evaluate deep learning-based Hi-C upscaling methods with our framework Hi-CY using seven datasets originating from three cell lines evaluated using three correlation metrics, four Hi-C similarity metrics, and four downstream analysis tasks, including TAD and chromatin loop recovery. We identify a distributional shift between Hi-C contact matrices generated from downsampled and experimentally generated sparse Hi-C datasets. We use Hi-CY to establish that the existing methods trained with downsampled Hi-C datasets tend to perform significantly worse on experimentally generated Hi-C datasets. We explore potential strategies to alleviate the drop in performance such as retraining models with experimentally generated datasets. Our results suggest that retraining improves performance up to 31 % on five sparse GM12878 datsets but provides marginal improvement in cross cell-type setting. Moreover, we observe that regardless of the training scheme, all deep-learning based methods struggle to recover biological features such as TADs and chromatin loops when provided with very sparse experimentally generated datasets as inputs.

bioinformatics↗

bioRxiv2021

Population structure and association mapping for hundred seed weight in mungbean minicore

Mungbean is an important legume rich in protein and carbohydrates. It is cultivated mostly in south to south east Asia and east Africa. One of the important yield contributing traits in mungbean is 100 seed weight. Variation for this trait is available in the mungbean germplasm, and to facilitate improvement of this trait through breeding, genome-wide association mapping for 100 seed weight was carried out in the World Vegetable Center mungbean minicore collection. A total of 24,870 single nucleotide polymorphic markers were tested for association with 100 see weight in plants grown in Pakistan, Bangladesh, Myanmar and Taiwan and candidate loci associated with seed weight on chromosomes 1, 4, 6, 8 and 9 were identified. None of the QTLs was stable across all environments, but loci on chromosomes 4 and 8 were significant in Pakistan and Bangladesh and other loci on chromosome 8 were significantly associated in plants grown in Myanmar and Taiwan.

bioinformatics↗