Biology subjects

Zafar, H.

Publications and source records attributed to Zafar, H..

At least 19 recordsLinked to original sources

bioRxiv2026

Ethanol-induced activation of BMP signaling and reprogramming of cardiomyocytes' transcriptome

Congenital heart disease (CHD) comprises a diverse group of structural heart defects present at birth due to complex interactions between genetic and environmental factors. Prenatal alcohol exposure (PAE) is a known environmental factor that disrupts fetal cardiogenesis and increases the risk of CHD. However, the molecular mechanisms behind ethanol (EtOH)-induced CHD remain obscure. This study investigated the effects of EtOH on bone morphogenetic protein (BMP) signaling and transcriptomic reprograming in HL-1 cardiomyocytes. HL-1 cells were treated with varying concentrations of EtOH (25, 50, and 100 mM) for 24 h. 100 mM of EtOH exposure significantly enhanced SMAD1/5 phosphorylation and upregulated BMP-responsive genes, namely Id1, Gata4, Mef2c, and Nkx2.5. Increased histone acetyltransferase activity further validated activation of BMP signaling through histone hyperacetylation. These effects were reversed by the BMP pathway inhibitor LDN-193189, confirming pathway-specific activation. Further, transcriptome analysis following 100 mM EtOH treatment identified 3,876 differentially expressed genes. KEGG enrichment analysis revealed significant dysregulation of cardiogenic pathways, including TGF-{beta}, Hedgehog, PI3K-Akt, Notch, FoxO, and calcium signaling pathways, along with extracellular matrix-receptor interaction and focal adhesion pathways. Gene Ontology analysis highlighted disturbances in heart development, cellular differentiation, apoptosis, extracellular matrix (ECM) organization, and chromatin regulation. Network analysis identified key hub genes, viz. Kras, Fn1, Col1a1, Prkaca, Fbn1, Col6a1, Col6a2, Ccnd1, Col1a2 and Myc which are upregulated and Hsp90aa1, Mdm2, Jun, Hras, Il6, Hsp90ab1, Pdgfra, Cdkn1a, Pparg, Fos and Hspa8 are downregulated which were subsequently validated by qRT-PCR. Collectively, these findings provide novel insights into the molecular basis of EtOH-induced CHD and identify potential biomolecule candidates for future therapeutic investigation.

cell biology↗

bioRxiv2026

Molecular mechanism of IF1- and IF2-driven translation initiation in bacteria

Bacterial translation initiation is a highly regulated process essential for accurate start codon selection and the assembly of an elongation-competent ribosome. Two initiation factors, 1 (IF1) and 3 (IF3), contribute to quality control for formation of 30S preinitiation complex (30S PIC), while the GTPase IF2 facilitates stable initiator tRNA binding and promotes subunit association. However, the molecular mechanism of IF1 action and the regulatory role of IF2-mediated GTP hydrolysis and inorganic phosphate (Pi) release remain poorly understood. Using ensemble cryo-EM integrated with fast-kinetics, we delineate the translation initiation pathway involving IF1 and IF2. We show that IF1 transiently associates with the 30S subunit and interferes with the formation of multiple inter-subunit bridges. IF2 promotes subunit association by stabilizing the 30S PIC through interactions mediated by its N-terminal domains. IF1 departure happens after or concomitant with GTP hydrolysis, following which the inter-subunit bridges establish. Then Pi release triggers remodeling of IF2 followed by its departure from the 70S initiation complex. These findings reveal how the coordinated interplay of IF1 and IF2 with the ribosome ensures translational fidelity and plays crucial role for formation of elongation-competent 70S.

biochemistry↗

bioRxiv2026

HDAC Inhibition Sensitizes Pancreatic Tumors to DNA Damage by Global Redistribution of the Transcriptional Machinery

The DNA damage response (DDR) is critical for pancreatic ductal adenocarcinoma (PDAC) development and therapeutic responses, including to genotoxic agents. While epigenetic modulators have been shown to contribute to the DDR, how chromatin regulation dictates responses to DNA damage in PDAC remains incompletely understood. Here, we identify Class I histone deacetylases (HDACs) as critical regulators of the DDR. HDAC1/2 directs the genomic distribution of H3K27ac, ensuring sufficient BRD4 and RNA polymerase II (Pol II) occupancy at DDR gene promoters. HDAC inhibition by entinostat shifts the balance of H3K27 acetylation preferentially towards intergenic regions, diverting BRD4 and Pol II from promoters, thereby suppressing DDR gene expression. In line with this, HDAC inhibition heightens DNA damage and sensitizes PDAC to diverse DNA-damaging and DDR-targeting agents. Since the clinical development of HDAC inhibitors has been limited by systemic toxicity, we developed bottlebrush prodrug (BPD) nanoparticles for tumor-selective entinostat delivery. Entinostat-BPD achieved tumor-specific HDAC inhibition while displaying potent efficacy and reduced systemic toxicity. These findings reveal an HDAC-dependent DDR vulnerability and offer combinational and precision targeting strategies to facilitate clinical translation and improve PDAC patient outcomes. SIGNIFICANCE STATEMENTThe ability of tumor cells to tolerate DNA damage limits the efficacy of many anticancer therapies. Our study reveals that pancreatic cancer cells enforce this resistance by sustaining expression of DNA damage response (DDR) genes through Class I histone deacetylases (HDACs). HDACs maintain genome-wide acetylation patterns required for efficient recruitment of the transcriptional machinery to DDR genes. Pharmacological HDAC inhibition disrupts this process and sensitizes pancreatic cancer cells to diverse DNA-damaging agents. To overcome systemic toxicity that limits translational potential, we further establish a bottlebrush prodrug nanoparticle platform that enables tumor-selective HDAC inhibition. Given the central role of the DDR in cancer, targeting HDAC-mediated DDR regulation through drug combinations and precision delivery may have broad therapeutic relevance across cancer types.

cancer biology↗

bioRxiv2026

A Comprehensive Benchmarking of Spatial Deconvolution andDomain Detection Methods across Diverse Tissues and SpatialTranscriptomic Technologies

Spatial transcriptomic technologies enable high-resolution characterization of gene expression patterns and reconstruction of cellular architecture within tissue contexts. Two key computational problems have emerged for analyzing these datasets: spatial deconvolution, for disentangling cell-type compositions at spatial locations, and spatial domain detection, for identifying spatially coherent regions within a tissue section. Although numerous methods have been developed for each task, a comprehensive and unified benchmarking study spanning diverse tissue types, spatial resolutions, and technological platforms remains lacking, hindering informed method selection by end users and impeding future methodological advancements. Here, we present spDDB (https://github.com/Zafar-Lab/spDDB), a comprehensive benchmarking framework for spatial deconvolution and domain detection methods across a large and diverse collection of datasets spanning multiple tissues, technologies, and biological conditions. We evaluated 21 deconvolution methods, including seven recently-developed approaches, across 37 datasets curated from brain, cancer, and organ tissues encompassing four distinct technologies. To enable rigorous evaluation, we introduced SynthST, a deep graph attention autoencoder-based simulator that generates realistic spatial cell-type distributions from spatial transcriptomic data, and employed a suite of spatial bivariate metrics including a novel bivariate Gearys C metric, alongside rare cell-type, and cell-shape characterization metrics, for multidimensional performance assessment. While Cell2location, RCTD and SONAR emerged as top-performing methods for spatial deconvolution across tissue types, deconvolution performance varied substantially based on tissue architecture, spatial technology, dataset scale, and cell type diversity. For domain detection, we benchmarked 18 methods across 36 datasets spanning six spatial technologies, identifying PROST, BASS, and SpaceFlow as the leading approaches, while revealing notable limitations of existing methods in handling large-scale datasets. Finally, we provide practical guidelines to assist end users in selecting optimal methods for both tasks across diverse experimental settings.

bioinformatics↗

bioRxiv2026

Myeloid Gi signaling acts as a weight-independent immunometabolic switch controlling systemic insulin sensitivity

Metabolic dysfunction does not necessarily correlate with adiposity. Metabolically healthy obese individuals and insulin-resistant lean individuals represent a fundamental paradox that implicates immune cell intrinsic mechanisms in the pathogenesis of type 2 diabetes. Here, we identify myeloid Gi signaling as a previously unrecognized determinant of whole-body glucose homeostasis. Single-cell transcriptomic analysis of adipose tissue macrophages from obese mice and humans reveals marked alteration in Gnai isoform, suggesting that myeloid Gi signaling is functionally engaged during metabolic disease. Using complementary myeloid-specific rodent models of Gi inhibition (pertussis toxin) and chemogenetic Gi activation (DREADD), we demonstrate that inhibition of Gi signaling improves glucose tolerance and enhances insulin sensitivity under both regular chow and high-fat diet conditions, independent of body weight and energy expenditure. Whereas acute Gi activation in lean mice modestly enhances glucose disposal, the same intervention during diet-induced obesity markedly impairs systemic glucose homeostasis, revealing context-dependent pathway function. Mechanistically, Gi inhibition amplifies macrophage cAMP-CREB signaling to drive IL-6 production, engaging STAT3- and AMPK-dependent pathways in adipose tissue and skeletal muscle to support insulin action. Conversely, Gi activation engages a previously uncharacterized G{beta}{gamma}-mTOR/AKT-JNK cascade, driving IL-1{beta} secretion that directly impairs insulin signaling in adipocytes and myotubes. Pharmacological IL-6 receptor blockade abolishes the metabolic benefits of Gi inhibition, whereas IL-1 receptor antagonism fully rescues Gi activation-induced metabolic dysfunction, establishing these cytokines as obligate downstream effectors. This signaling architecture is conserved in human macrophages, and ATAC-seq profiling reveals chromatin remodeling at cAMP-CREB and IL-6 regulatory pathway loci, consistent with the observed transcriptional reprogramming. Together, these findings establish myeloid Gi signaling as a weight-independent immunometabolic switch that couples opposing cytokine programs to systemic insulin sensitivity and identify this pathway as a therapeutic target in obesity-associated metabolic disease. O_FIG O_LINKSMALLFIG WIDTH=200 HEIGHT=141 SRC="FIGDIR/small/713834v1_ufig1.gif" ALT="Figure 1"> View larger version (56K): org.highwire.dtl.DTLVardef@15e4576org.highwire.dtl.DTLVardef@224cd5org.highwire.dtl.DTLVardef@1e24888org.highwire.dtl.DTLVardef@1080bf3_HPS_FORMAT_FIGEXP M_FIG Graphical Abstract C_FIG

physiology↗

bioRxiv2025

Acute activation of Gq-signaling in islet macrophages inhibits β-cell insulin secretion through AMPK-sphingolipid axis

Obesity-associated inflammation disrupts pancreatic {beta}-cell function, but the immune-derived signals that directly regulate insulin secretion remain incompletely defined. Here, we identify myeloid Gq signaling as a critical immunometabolic node that links macrophage activation to {beta}-cell dysfunction. For the first time, we employed a chemogenetic approach (DREADDs) to selectively and temporally activate Gq-coupled GPCR signaling in myeloid cells to examine its effect on islet function. Our findings reveal that acute Gq activation in islet-resident macrophages impaired glucose-stimulated insulin secretion, uncovering a previously unrecognized immune-endocrine axis. Conversely, myeloid-specific Gq deletion improves systemic glucose homeostasis, underscoring the physiological relevance of this pathway. Mechanistic analysis revealed that Gq activation in macrophages stimulates AMPK signaling and drives the secretion of sphingolipids. These lipids suppress insulin secretion and introduce a new mechanism for immune-islet communication, extending beyond traditional cytokine-based models. We further identify the lipid-sensing receptor GPR18 as an upstream activator of the Gq-AMPK pathway in macrophages. GPR18 stimulation recapitulated the Gq-dependent sphingolipid secretion and {beta}-cell inhibitory phenotype, which was abolished in myeloid Gq-deficient mice. Collectively, these findings establish a mechanistic framework whereby macrophage Gq signaling integrates lipid sensing and metabolic stress to modulate {beta}-cell function. This work reveals a previously unrecognized macrophage-{beta}-cell communication axis with therapeutic potential for restoring insulin secretion in metabolic diseases such as obesity and type 2 diabetes. O_FIG O_LINKSMALLFIG WIDTH=200 HEIGHT=109 SRC="FIGDIR/small/680858v1_ufig1.gif" ALT="Figure 1"> View larger version (44K): org.highwire.dtl.DTLVardef@790c4dorg.highwire.dtl.DTLVardef@116cceborg.highwire.dtl.DTLVardef@1e76631org.highwire.dtl.DTLVardef@e65770_HPS_FORMAT_FIGEXP M_FIG C_FIG

physiology↗

bioRxiv2025

Breaking Barriers: Transitioning from X-ray Crystallography to Cryo-EM for Structural Studies of ATAD2B

Cryo-electron microscopy (cryo-EM) has transformed structural biology by enabling near-atomic resolution of large macromolecular complexes without the need for crystallization. Here, we describe our laboratorys transition from X-ray crystallography to single-particle cryo-EM to investigate the ATPase family AAA+ domain-containing protein 2B (ATAD2B), a chromatin regulator implicated in epigenetic signaling. We outline the challenges encountered during protein expression, purification, and sample preparation, including co-purification of the chaperonin GroEL, and strategies employed to overcome these obstacles. Our workflow highlights critical steps in sample optimization, grid vitrification, and data processing using CryoSPARC, cisTEM, and Topaz, as well as computational requirements for high-resolution reconstructions. We also discuss model building, refinement, and validation approaches, emphasizing best practices for new cryo-EM users. This work provides practical insights for structural biologists adopting cryo-EM, particularly for large, flexible protein complexes, and underscores the importance of integrated approaches combining biochemical, computational, and imaging strategies.

biochemistry↗

bioRxiv2025

A Temporal Single-Cell Multi-Omics Atlas of Murine Pancreatic Islet Remodeling During Hyperglycaemia Progression

Pancreatic islets undergo coordinated cellular remodeling during obesity-induced insulin resistance (IR). However, the associated molecular changes across endocrine and non-endocrine compartments remain largely unexplored. Here, using longitudinal single-cell RNA sequencing (scRNA-seq) and single-cell ATAC sequencing (scATAC-seq) on islets from C57BL/6 mice subjected to high-fat diet (HFD) feeding for 8, 16, and 24 weeks, along with age-matched controls on regular chow, we mapped dynamic changes in islet cell composition and transcriptional states. Beta cells demonstrated pronounced stress-induced reprogramming, with the emergence of proliferative and dysfunctional subsets. Alpha and delta cell fractions declined under HFD, despite increased polyhormonal biosynthesis, suggesting functional rather than numerical adaptation. Immune profiling showed robust expansion of proinflammatory M1 macrophages and upregulation of NF-{kappa}B and chemotaxis pathways, particularly at 16 weeks. Notably, cell-cell communication analyses revealed diet-specific disruption in signaling networks. Under HFD conditions, intercellular communication among beta cells, macrophages, and delta cells was markedly altered, leading to the disruption of key signaling pathways such as the gastric inhibitory polypeptide receptor (GIPR) and major histocompatibility complex-I (MHC-I). Notably, C-C motif chemokine ligand 27A (Ccl27a) expression and chromatin accessibility were significantly altered in a distinct subpopulation of beta cells under HFD condition, indicative of a niche-specific regulatory mechanism. Integration with human islet datasets from obese and type 2 diabetes (T2D) donors confirmed conserved shifts in beta cell identity and immune activation. This study presents a comprehensive high-resolution atlas of islet remodeling under metabolic stress, identifying key communication nodes and transcriptional programs pertinent to T2D pathogenesis. Graphical Abstract O_FIG O_LINKSMALLFIG WIDTH=200 HEIGHT=141 SRC="FIGDIR/small/656754v1_ufig1.gif" ALT="Figure 1"> View larger version (52K): org.highwire.dtl.DTLVardef@64aee0org.highwire.dtl.DTLVardef@150beborg.highwire.dtl.DTLVardef@1c19c6dorg.highwire.dtl.DTLVardef@4d1a0_HPS_FORMAT_FIGEXP M_FIG C_FIG

genomics↗

bioRxiv2025

Intratumor Heterogeneity Through the Lens of Gene Regulatory Networks

Intratumor heterogeneity (ITH), the presence of heterogeneous genomic and transcriptomic populations in tumor and its implications for the cancer therapies has been one of the focal points of computational cancer biology in the past years. While ITH is well characterized at the genomic level, the regulatory wiring connecting clonal genotypes to transcriptional states remains largely unmapped. To understand whether the evidence for ITH at the gene regulatory level can be inferred from scRNA-seq data, we analyzed scRNA-seq datasets spanning triple-negative breast cancer, colorectal cancer, glioblastoma, and a longitudinal stage IV breast cancer cohort, inferring clone-specific gene regulatory networks (GRNs) and developing quantitative tools to compare them: a Wasserstein distance-based measure of regulatory divergence, a kth order neighborhood similarity framework distinguishing structural from functional rewiring, bootstrap-based edge confidence, and an integrated regulon classification pipeline. Our analyses demonstrate that tumor clones exhibit reproducibly distinct network architectures, and dominant clones show significantly greater regulatory divergence from normal cells than smaller clones, linking regulatory rewiring to clonal fitness. Treatment-induced remodeling exceeds clonal variation, with rewiring exhibiting structural divergence but functional convergence. Regulon analysis further resolved clone-variable, sample-specific, and broadly conserved programs, some of which containing transcription factors whose target genes independently were associated with patient survival.

bioinformatics↗

bioRxiv2025

multiHIVE: Hierarchical Multimodal Deep Generative Model for Single-cell Multiomics Integration

MotivationRecently developed single-cell multiomics technologies are enhancing our understanding of cellular heterogeneity by providing multiple views of a biological system. CITE-seq (cellular indexing of transcriptomes and epitopes by sequencing) is one such multiomics assay, with the ability to connect cell states to functions by simultaneously profiling RNA and surface proteins from the same cell. However, the distinct technical characteristics of these data modalities pose significant challenges to their integration into a cohesive representation of cellular identity. ResultsHere we present multiHIVE, a hierarchical multimodal deep generative model for inferring cellular embeddings by integrating CITE-seq data modalities. multiHIVE employs hierarchically stacked latent variables as well as modality-specific latent variables to capture shared and private information from the modalities respectively, facilitating integration, denoising and imputation tasks. Extensive benchmarking using gold-standard real and simulated datasets demonstrates multiHIVEs superiority in integrating CITE-seq datasets. Moreover, multiHIVE outperformed the state-of-the-art methods in imputing missing protein measurements and integration of CITE-seq dataset with unimodal dataset. Using a thymocyte development dataset, we showed that multiHIVEs cellular embeddings can lead to improved trajectory inference and gene trend identification. Finally, using datasets across development and disease, we demonstrated that factorization of multiHIVE-inferred denoised expression into gene expression programs aids in identifying biological processes at multiple levels of cellular hierarchy. Availability and ImplementationmultiHIVE is implemented in Python and is publicly available at https://github.com/Zafar-Lab/multiHIVE.

bioinformatics↗

bioRxiv2024

PHALCON: phylogeny-aware variant calling from large-scale single-cell panel sequencing datasets

Single-cell sequencing (SCS) enables variant detection and tumor phylogeny reconstruction for resolving intra-tumor heterogeneity (ITH), which causes drug resistance and cancer relapse. Recently emerged panel sequencing methods sequence disease-specific genes across thousands of cells, but existing variant callers and SCS-specific phylogenetic methods struggle with large-scale datasets and amplification biases in panel-based sequencing protocols. We present a statistical variant caller, PHALCON for scalable mutation detection from large-scale single-cell panel sequencing data by modeling tumor evolution under a finite-sites model along a clonal phylogeny. Across a wide variety of simulation settings, PHALCON outperformed state-of-the-art methods in variant calling, tumor phylogeny inference, and runtime. From triple negative breast cancer (TNBC) and acute myeloid leukemia (AML) datasets, PHALCON detected novel somatic mutations with high functional impact, resolved clonal substructure and rare clones. In AML, PHALCON also uncovered poor-survival subgroups harboring DNA methylation and chromatin/cohesin mutations and revealed novel cellular co-occurrence and exclusivity patterns of driver mutations.

bioinformatics↗

bioRxiv2024

Faster amylin aggregation on fibrillar collagen hastens diabetic progression through β cell death and loss of function

Amyloid deposition of the neuroendocrine peptide amylin in islet tissues is a hallmark of type 2 diabetes (T2DM), leading to {beta}-cell toxicity through nutrient deprivation, membrane rupture and apoptosis. Though accumulation of toxic amylin aggregates in islet matrices is well documented, the role of the islet extracellular matrix in mediating amylin aggregation and its pathological consequences remains elusive. Here, we address this question by probing amylin interaction with collagen I (Col I)--whose expression in the islet tissue increases during diabetes progression. By combining multiple biophysical techniques, we show that hydrophobic, hydrophilic & cation-{pi} interactions regulate amylin binding to Col I, with fibrillar collagen driving faster amylin aggregation. Amylin-entangled Col I matrices containing high amounts of amylin induce death and loss of function of INS1E {beta}-cells. Together, our results illustrate how amylin incorporation in islet matrices through amylin-Col interactions drives T2DM progression by impacting {beta}-cell viability and insulin secretion.

biophysics↗

bioRxiv2023

Temporal recording of mammalian development and precancer

Key to understanding many biological phenomena is knowing the temporal ordering of cellular events, which often require continuous direct observations [1, 2]. An alternative solution involves the utilization of irreversible genetic changes, such as naturally occurring mutations, to create indelible markers that enables retrospective temporal ordering [3-8]. Using NSC-seq, a newly designed and validated multi-purpose single-cell CRISPR platform, we developed a molecular clock approach to record the timing of cellular events and clonality in vivo, while incorporating assigned cell state and lineage information. Using this approach, we uncovered precise timing of tissue-specific cell expansion during murine embryonic development and identified new intestinal epithelial progenitor states by their unique genetic histories. NSC-seq analysis of murine adenomas and single-cell multi-omic profiling of human precancers as part of the Human Tumor Atlas Network (HTAN), including 116 scRNA-seq datasets and clonal analysis of 418 human polyps, demonstrated the occurrence of polyancestral initiation in 15-30% of colonic precancers, revealing their origins from multiple normal founders. Thus, our multimodal framework augments existing single-cell analyses and lays the foundation for in vivo multimodal recording, enabling the tracking of lineage and temporal events during development and tumorigenesis.

developmental biology↗

bioRxiv2023

Thrombospondin 2 is a key determinant of fibrogenesis in NAFLD

Hepatic overexpression of the thrombospondin 2 gene (THBS2) and elevated levels of circulating thrombospondin 2 (TSP2) have been observed in patients with chronic liver disease. The current study aimed to identify the specific cells expressing THBS2/TSP2 in non-alcoholic fatty liver disease (NAFLD) and investigate the underlying mechanism behind THBS2/TSP2 up-regulation. Comprehensive NAFLD liver gene datasets, including single-cell RNA sequencing (scRNA-seq), in-house NAFLD liver tissue, and LX-2 cells derived from human hepatic stellate cells (HSCs), were analyzed using a combination of computational biology, genetic, immunological, and pharmacological approaches. Analysis of the genetic dataset revealed the presence of 1433 variable genes in patients with advanced fibrosis NAFLD, with THBS2 ranked among the top 2 genes. Quantitative polymerase chain reaction examination of NAFLD livers showed a significant correlation between THBS2 expression and fibrosis stage (r=0.349, p<0.001). In support of this, scRNA-seq data and in situ hybridization demonstrated that the THBS2 gene was highly expressed in HSCs of NAFLD patients with advanced fibrosis. Pathway analysis of the gene dataset revealed THBS2 expression to be associated with the transforming growth factor beta (TGF{beta}) pathway and collagen gene activation. Moreover, the activation of LX-2 cells with TGF{beta} increased THBS2/TSP2 and collagen expression independently of the TGF{beta}-SMAD2/3 pathway. THBS2 gene knockdown significantly decreased collagen expression in LX-2 cells. In conclusion, THBS2/TSP2 is highly expressed in HSCs and plays a role in regulating fibrogenesis in NAFLD patients. THBS2/TSP2 may therefore represent a potential target for anti-fibrotic therapy in NAFLD. (241 words) One-sentence summariesThrombospondin 2 represent a potential target for anti-fibrotic therapy in NAFLD.

molecular biology↗

bioRxiv2023

Charting spatial ligand-target activity using Renoir

The advancement of single-cell RNA sequencing (scRNA-seq) and spatial transcriptomics has enabled the inference of cellular interactions in a tissue microenvironment. Despite the development of cell-cell interaction inference methods, there is a lack of methods capable of mapping the influence of ligands on downstream target genes across a spatial topology with specific cell type composition, with the potential to shed light on niche-specific relationship between ligands and their downstream targets. Here we present Renoir for charting the ligand-target activities across a spatial topology and delineating spatial communication niches harboring specific ligand-target activities. Renoir also spatially maps pathway-level activity of ligand-target genesets and identifies domain-specific ligand-target activities. Across spatial datasets with varying resolution (spot to single-cell) ranging from development to disease, Renoir inferred cellular niches with distinct ligand-target interactions, spatially mapped pathway activities, and identified context-specific novel cell-cell interactions. Renoir uncovers biological insights and therapeutically-relevant cellular crosstalk from spatial transcriptomics data.

bioinformatics↗

bioRxiv2022

scDREAMER: atlas-level integration of single-cell datasets using deep generative model paired with adversarial classifier

Integration of heterogeneous single-cell sequencing datasets generated across multiple tissue locations, time and conditions is essential for a comprehensive understanding of the cellular states and expression programs underlying complex biological systems. Here, we present scDREAMER (https://github.com/Zafar-Lab/scDREAMER), a novel data integration framework that employs a novel adversarial variational autoencoder for learning lower-dimensional cellular embeddings and a batch classifier neural network for the removal of batch effects. Using five real benchmarking datasets, we demonstrated that scDREAMER can overcome critical challenges including the presence of skewed cell types among batches, nested batch effects, large number of batches and conservation of development trajectory across different batches. Moreover, our benchmarking demonstrated that scDREAMER outperformed state-of-the-art methods in batch-correction and conservation of biological variation. Using a 1 million cells dataset, we further showed that scDREAMER is scalable and can perform atlas-level integration across species (e.g., human and mouse) while being faster than other deep-learning-based methods.

bioinformatics↗

bioRxiv2022

FiMO: Inferring the Temporal Order of Mutations on Clonal Phylogeny under Finite-sites Models

Determining the temporal order of somatic mutations that drives cancer progression is essential for understanding the intra-tumor heterogeneity (ITH) and designing personalized therapy. Recently emerged single-cell DNA sequencing (SCS) technologies provide high-resolution datasets suitable for elucidating the temporal order of mutation. However, this task is challenged by technical artifacts associated with single-cell sequencing. While computational methods have been developed to account for these errors, these methods rely on using infinite sites assumption which gets violated in human cancers due to deletion, loss of heterozygosity and parallel mutations. Here, we propose a novel method FiMO, which employs a Bayesian frameowrk for inferring the temporal order of somatic mutations from noisy SCS mutational profiles under tumor evolutionary models that account for mutation recurrence and losses. Using synthetic datasets generated under a wide variety of settings, we show that FiMO outperforms the state-of-the-art methods in inferring the temporal order of mutations. We also applied FiMO on two experimental colon cancer datasets for inferring the temporal order of somatic mutations and quantifying their posterior probability. FiMO is publicly available at https://github.com/aveshag/FiMO.

bioinformatics↗

bioRxiv2022

Phylovar: Towards scalable phylogeny-aware inference of single-nucleotide variations from single-cell DNA sequencing data

Single-nucleotide variants (SNVs) are the most common variations in the human genome. Recently developed methods for SNV detection from single-cell DNA sequencing (scDNAseq) data, such as SCI{Phi} and scVILP, leverage the evolutionary history of the cells to overcome the technical errors associated with single-cell sequencing protocols. Despite being accurate, these methods are not scalable to the extensive genomic breadth of single-cell whole-genome (scWGS) and whole-exome sequencing (scWES) data. Here we report on a new scalable method, Phylovar, which extends the phylogeny-guided variant calling approach to sequencing datasets containing millions of loci. Through benchmarking on simulated datasets under different settings, we show that, Phylovar outperforms SCI{Phi} in terms of running time while being more accurate than Monovar (which is not phylogeny-aware) in terms of SNV detection. Furthermore, we applied Phylovar to two real biological datasets: an scWES triple-negative breast cancer data consisting of 32 cells and 3375 loci as well as an scWGS data of neuron cells from a normal human brain containing 16 cells and approximately 2.5 million loci. For the cancer data, Phylovar detected somatic SNVs with high or moderate functional impact that were also supported by bulk sequencing dataset and for the neuron dataset, Phylovar identified 5745 SNVs with non-synonymous effects some of which were associated with neurodegenerative diseases. We implemented Phylovar and made it publicly available at https://github.com/mae6/Phylovar.git.

cancer biology↗