Biology subjects

Karimzadeh, M.

Publications and source records attributed to Karimzadeh, M..

At least 19 recordsLinked to original sources

bioRxiv2026

A high-resolution protein-RNA binding atlas for structure-aware interaction prediction and post-transcriptional regulatory modeling

RNA-binding proteins (RBPs) control post-transcriptional gene expression by regulating RNA processing, localization, stability, and translation. However, our understanding of RNA regulation remains limited by the resolution, scale, and uniformity of available RBP binding maps. Here we introduce eclair CLIP (ecCLIP), a high-throughput CLIP-seq platform that combines endogenous protein tagging for uniform affinity capture with ligation-free library preparation to generate high-resolution RNA-protein binding profiles. We applied ecCLIP to 67 endogenously tagged RBPs in HEK293T cells, generating a high-quality transcriptome-scale atlas that recovered known binding preferences and captured post-transcriptional regulons linked to RNA half-life. To test the utility of the atlas, we incorporated ecCLIP into Borzuki, a Borzoi-derived sequence-to-function model trained on complementary RNA degradation, expression, CLIP, and Perturb-seq data. Inclusion of ecCLIP data improved RNA half-life prediction and assessment of 3' UTR variant effects on RNA stability, demonstrating that our atlas helps capture RNA dynamics. To directly model the cis-regulatory grammar underlying these RBP-RNA interactions, we developed Pythia, a structure-aware neural network architecture that incorporates context-free grammar-inspired fixed convolutions to explicitly encode RNA base-pairing features. Pythia predicted RBP binding from sequence and structural context and decomposed predictions into sequence- and structure-dependent components. We tested this decomposition for five RBPs using massively parallel reporter CLIP (MPRC) binding assays in cells, using synthetic RNA libraries containing mutations designed to preserve or disrupt predicted RNA structures across nominated binding sites. We used the same MPRC strategy to confirm allele-specific binding effects for variants prioritized by Pythia. Together, ecCLIP and Pythia provide a resource for structure-aware decoding of RNA-protein interaction grammar and interpretation of post-transcriptional regulatory variation.

genomics↗

bioRxiv2026

X-Cell: Scaling Causal Perturbation Prediction Across Diverse Cellular Contexts via Diffusion Language Models

Causal models of cellular systems hold the promise to empower broad biological discovery, including the systematic identification of novel targets for drug discovery. Predicting how genetic and pathway perturbations reshape gene expression across diverse cellular contexts is a prerequisite for building generalizable cellular foundation models. However, current methods typically fail to extrapolate beyond their training distributions because they rely predominantly on observational expression atlases rather than interventional perturbation data. We present X-Atlas/Pisces, the largest genome-wide CRISPRi Perturb-seq compendium to date, comprising 25.6 million perturbed single-cell transcriptomes across 16 biologically diverse contexts, including widely used cell lines, induced pluripotent stem cells (iPSCs), resting and CD3/CD28 activated Jurkat T lymphoma cells, and multi-lineage differentiating iPSCs. Leveraging this resource, we develop X-Cell, a diffusion language model that predicts perturbation responses by iteratively refining control-to-perturbed state transitions through cross-attention to multi-modal biological priors derived from natural language, protein language models, interaction networks, genetic dependency maps, and morphological profiles. X-Cell outperforms existing state-of-the-art models by up to five-fold on key metrics such as Pearson{Delta} (correlation between predicted and observed perturbation-induced log-fold changes), and demonstrates zero-shot prediction of T cell inactivating perturbations in stimulated Jurkat cells. We scale X-Cell to 4.9 billion parameters (X-Cell-Ultra), the largest causal perturbation model to date. We demonstrate for the first time that perturbation prediction follows power-law scaling with an exponent matching large language models. X-Cell-Ultra demonstrates zero-shot generalization to novel biological contexts, including unseen iPSC-derived melanocyte progenitors and primary human CD4+ T cells from multiple donors, and outperforms all baselines after self-supervised test-time adaptation. These results demonstrate that coordinated scaling of causal perturbation data and model capacity yields foundation models capable of generalizable perturbation prediction across cellular contexts, with potential applications for improving computational target identification, validation, and context-specific therapeutic prioritization.

systems biology↗

bioRxiv2025

Orphan Non-Coding RNAs Drive Tumorigenesis and Enable Pre-Diagnostic Detection in HPV-Negative Head and Neck Cancer

HPV-negative head and neck squamous cell carcinoma (HNSCC) lacks reliable biomarkers for early detection and molecular stratification. We recently described orphan non-coding RNAs (oncRNAs) as a class of cancer-emergent small RNAs that arise from cryptic promoters largely silent in normal tissues. Here, we define the comprehensive landscape of oncRNAs in HNSCC. Using TCGA and CPTAC-3 data, we demonstrate that these transcripts are not only tumor-specific but provide digital fingerprints that accurately stratify tumors by anatomical site. Moving from association to function, we investigated whether these emergent transcripts actively drive tumorigenesis. We focused on oncRNA64585, a transcript highly enriched in oral cavity tumors and associated with tumor progression. Its depletion significantly impaired proliferation in vitro and suppressed tumor growth in syngeneic murine models. Mechanistically, transcriptomic and chromatin accessibility profiling revealed that oncRNA64585 regulates transcriptional programs governing cell cycle progression and apoptosis through interaction with the Polycomb Repressive Complex 2 (PRC2). Finally, we demonstrate the clinical utility of these signatures: oncRNA profiles in plasma predicted oral cavity cancer with >90% accuracy up to four years prior to clinical diagnosis. These findings establish oncRNAs as active drivers of HNSCC biology and potent tools for early detection and therapeutic targeting.

cancer biology↗

bioRxiv2025

Discovery of a tRNA-regulatory transcription factor that suppresses breast cancer metastasis

Transfer RNAs (tRNAs), once viewed as static adaptors in translation, are now recognized as dynamic regulators of gene expression. While recent studies have illuminated roles for tRNA stability, the upstream mechanisms governing tRNA transcription remain poorly understood. To address this gap, we generated the EXpression atlas of tRNA (EXTRNA), a high-resolution tRNA expression dataset spanning 24 cell lines across 9 human tissues. EXTRNA revealed both tissue-type-specific expression programs ("tRNAomes") and unexpected intra-tissue heterogeneity across breast cancer samples. Integrating EXTRNA with computational network analysis and data from other publicly available datasets, we identified Zinc Finger ZZ-Type And EF-Hand Domain Containing 1 (ZZEF1) as the first sequence-specific transcription factor of a particular tRNA. ZZEF1 promoted tRNA-LysUUU transcription by partnering with the ATP-dependent chromatin remodeler Chromodomain Helicase DNA Binding Protein 6 (CHD6), enhancing chromatin accessibility at tRNA-Lys-TTT-3 loci. ZZEF1 deficiency reduced tRNA-LysUUU abundance, decreased the translational efficiency of AAR codon-enriched mRNAs--including the tumor suppressor Serine/Threonine Kinase 3 (STK3)--and promoted metastatic progression in breast cancer in vivo. Together, our findings establish a previously unrecognized mechanism for RNA polymerase III-mediated tRNA transcription and define a regulatory circuit linking chromatin remodeling, codon-specific translation, and tumor suppression. More broadly, this work introduces a framework for dissecting the regulatory logic of the tRNAome and highlights tRNA expression control as a promising avenue for therapeutic intervention.

genomics↗

bioRxiv2025

A multi-modal cell-free RNA language model for liquid biopsy applications

Cell-free RNA (cfRNA) profiling has emerged as a powerful tool for non-invasive disease detection, but its application is limited by data sparsity and complexity, especially in settings with constrained sample availability. We introduce Exai-1, a multi-modal, transformer-based generative foundation model that integrates RNA sequence embeddings with cfRNA abundance data to capture biologically meaningful representations of circulating RNAs. By leveraging both sequence and expression modalities, Exai-1 captures a biologically meaningful latent structure of cfRNA profiles. Pre-trained on over 306 billion tokens from 8,339 samples, Exai-1 enhances signal fidelity, reduces technical noise, and improves disease detection by generating synthetic cfRNA profiles. We show that self-attention and variational inference are particularly important for preservation of biological signals and contextual relationships. Additionally, Exai-1 facilitates cross-biofluid translation and assay compatibility through disentangling biological signals from confounders. By uniting sequence-informed embeddings with cfRNA expression patterns, Exai-1 establishes a transfer learning foundation for liquid biopsy, offering a scalable and adaptable framework for next-generation cfRNA-based diagnostics.

bioinformatics↗

bioRxiv2025

Targeting Osteosarcoma heterogeneity to improve therapeutic response

Intra tumor heterogeneity complicates cancer therapy by providing tumors with the ability to alter their phenotypes and become more therapy resistant. Here, we tested the hypothesis that identifying and modulating expression of key state-specific transcription factors could be used as a strategy for driving cells to a more therapy-sensitive state. Recent single-cell studies have explored the inter and intra tumoral heterogeneity of osteosarcoma and identified gene pathways enriched in specific cell states. For example, metastatic tumors are characterized by an expression of genes in the TNF-, PI3K, TGF{beta} and mTOR pathways. We identified similar profiles in osteosarcoma patient-derived xenograft-derived cell lines and potential transcription factor drivers of these states. We then used perturb-seq to downregulate expression of key transcription factors and evaluated the effect of these modulations on single cell RNA profiles and drug responses. Knockdown of NFE2L3 or NR0B1 increased the proportion of cells sensitive to targeted therapy. This approach, which could potentially be applied to other cancers, could be used as a strategy to increase the response to targeted therapies by increasing the proportion of cells in a drug-sensitive state. HighlightsO_LIDistinct transcriptomic states were identified in osteosarcoma cell lines using single-cell RNA sequencing. C_LIO_LILineage tracing identified states with differential sensitivity to therapy. C_LIO_LIUsing perturb-seq, we identified transcription factors that drive cells towards a more sensitive state. C_LIO_LIThe transcription factor NFE2L3 was identified as targets capable of reprogramming cells to a sensitive state. C_LI

genomics↗

bioRxiv2024

An intrinsically disordered RNA-binding region provides local target selectivity and is essential for LARP6 function.

Intrinsically disordered regions (IDRs) are prevalent in RNA-binding proteins (RBPs), yet their roles in RNA interactions remain poorly defined. Here, we examined the structured and disordered RNA-binding activities of LARP6, an RBP with a diverse RNA-binding repertoire. Using mass spectrometry-based RNA interaction mapping in living cells, we identified direct LARP6-RNA contacts within both the structured La-module and its flanking IDRs. Mutagenesis combined with individual-nucleotide resolution UV crosslinking and immunoprecipitation (iCLIP) revealed the La-module, but not the IDRs, as essential for LARP6 binding to RNA. Deletion of the N-terminal IDR broadened the footprints of LARP6 on RNA, uncovering a role in RNA-binding selectivity. Mechanistically, this is achieved through restricting the conformational flexibility of the adjacent La-module. The IDR-mediated RNA-binding selectivity is critical for LARP6-driven cancer cell viability and invasion. Our findings uncover a previously unrecognised critical function for IDRs in promoting selective RBP-RNA interactions, which operates through conformational restriction.

biochemistry↗

bioRxiv2023

Integrative identification of non-coding regulatory regions driving metastatic prostate cancer

Large-scale sequencing efforts of thousands of tumor samples have been undertaken to understand the mutational landscape of the coding genome. However, the vast majority of germline and somatic variants occur within non-coding portions of the genome. These genomic regions do not directly encode for specific proteins, but can play key roles in cancer progression, for example by driving aberrant gene expression control. Here, we designed an integrative computational and experimental framework to identify recurrently mutated non-coding regulatory regions that drive tumor progression. Application of this approach to whole-genome sequencing (WGS) data from a large cohort of metastatic castration-resistant prostate cancer (mCRPC) revealed a large set of recurrently mutated regions. We used (i) in silico prioritization of functional non-coding mutations, (ii) massively parallel reporter assays, and (iii) in vivo CRISPR-interference (CRISPRi) screens in xenografted mice to systematically identify and validate driver regulatory regions that drive mCRPC. We discovered that one of these enhancer regions, GH22I030351, acts on a bidirectional promoter to simultaneously modulate expression of U2-associated splicing factor SF3A1 and chromosomal protein CCDC157. We found that both SF3A1 and CCDC157 are promoters of tumor growth in xenograft models of prostate cancer. We nominated a number of transcription factors, including SOX6, to be responsible for higher expression of SF3A1 and CCDC157. Collectively, we have established and confirmed an integrative computational and experimental approach that enables the systematic detection of non-coding regulatory regions that drive the progression of human cancers.

cancer biology↗

bioRxiv2023

Disease Modifying Osteoarthritis Drug Discovery Using A Temporal Phenotypic Reporter In Primary Human Chondrocytes

Osteoarthritis is a significant and growing problem with no disease modifying drugs in the clinic. Current screening platforms typically use 2D culture, immortalized or non-human cells in a hyperoxic environment. To challenge this paradigm and identify new drugs, we engineered primary human chondrocytes with a secreted luciferase reporter under the control of the articular cartilage marker, type II collagen. We then successfully screened a natural product library using a high throughput model with COL2A1-Gaussia luciferase primary human chondrocyte reporter cells in 3D aggregates under physioxia. We identified several candidate compounds that increased type II collagen over controls, with aromoline being the best candidate. Aromoline is a bisbenzylisoquinoline alkaloid that has been studied for its anti-proliferative, anti-inflammatory, and anti-microbial properties, and we are the first to explore its effects on chondrocytes and chondrogenesis. In silico analysis of predicted targets narrowed by RNA-Seq data on expression provided an unexpected initial candidate target protein: the dopamine receptor D4 (DRD4). The researchers confirmed upregulation in the expression of DRD4 and type II collagen after treatment with aromoline. This novel approach combining in silico and in vitro methods provides a platform for drug discovery in a challenging and under-researched area. In conclusion, a novel drug (aromoline) and target receptor (dopamine receptor D4) were identified as stimulating type II collagen, with the goal to treat or prevent osteoarthritis.

pharmacology and toxicology↗

bioRxiv2022

Micronutrient Optimization Using Design of Experiments Approach in Tissue Engineered Articular Cartilage for Production of Type II Collagen

Tissue Engineering of cartilage has been hampered by the inability of engineered tissue to express native levels of type II collagen in vitro. Inadequate levels of type II collagen are, in part, due to a failure to recapitulate the physiological environment in culture. In this study, we engineered primary rabbit chondrocytes to express a secreted reporter, Gaussia Luciferase, driven by the type II collagen promoter, and applied a Design of Experiments approach to assess chondrogenic differentiation in micronutrient-supplemented medium. Using a Response Surface Model, 240 combinations of micronutrients absent in standard chondrogenic differentiation medium, were screened and assessed for type II collagen expression. Five conditions predicted to produce the greatest Luciferase expression were selected for further study. Validation of these conditions in 3D aggregates identified an optimal condition for type II collagen expression. Engineered cartilage grown in this condition, showed a 170% increase in type II collagen expression (Day 22 Luminescence) and in Youngs tensile modulus compared to engineered cartilage in basal media alone. Collagen cross-linking analysis confirmed formation of type II-type : II collagen and type II-type : IX collagen cross-linked heteropolymeric fibrils, characteristic of mature native cartilage. Combining a Design of Experiments approach and secreted reporter cells in 3D aggregate culture enabled a high-throughput platform that can be used to identify more optimal physiological culture parameters for chondrogenesis.

cell biology↗

bioRxiv2022

Optimizing bioink composition for human chondrocyte expression of lubricin

The surface zone of articular cartilage is the first area impacted by cartilage defects, commonly resulting in osteoarthritis. Chondrocytes in the surface zone of articular cartilage synthesize and secrete lubricin, a proteoglycan that functions as a lubricant protecting the deeper layers from shear stress. 3D bioprinting is a tissue engineering technique that uses cells encapsulated in biomaterials to fabricate 3D constructs. Gelatin methacrylate (GelMA) is a frequently used biomaterial for 3D bioprinting cartilage. Oxidized methacrylated alginate (OMA) is a chemically modified alginate designed for its tunable degradation rate and mechanical properties. To determine an optimal combination of GelMA and OMA for lubricin expression, we used our novel high-throughput human articular chondrocyte reporter system. Primary human chondrocytes were transduced with PRG4 (lubricin) promoter-driven Gaussia luciferase, allowing for temporal assessment of lubricin expression. A lubricin expression driven Design of Experiment screen and subsequent validation identified 14% GelMA/2% OMA for further study. Therefore, 14% GelMA/2% OMA, 14% GelMA and 16% GelMA were 3D bioprinted. The combination of lubricin protein expression and shape retention over the 22 days in culture, determined the 14% GelMA/2%OMA to be the optimal formulation for lubricin secretion.

bioengineering↗

bioRxiv2022

The molecular consequences of androgen activity in the human breast

The mammary gland has been extensively studied for estrogen and progesterone reactivity, but the molecular effects of androgen in the breast remain largely unexplored. Transgender men are recorded female at birth but identify as male and may undergo gender-affirming androgen therapy to align their physical characteristics and gender identity. Here we perform single cell resolution transcriptome, chromatin, and spatial profiling of androgen treated breasts from transgender men. We find male-biased androgen receptor gene targets are upregulated in cells expressing androgen receptor, and that paracrine signaling drives sex-relevant changes in other cell types. We observe an altered epithelium, shifts in immune populations, and a reduction of capillary vasculature. Finally, we find evidence of the metabolic impact of androgen and identify a gene regulatory network driving androgen-directed fat loss. This work elucidates the molecular consequences of androgen in the human breast at single cell resolution.

genomics↗

bioRxiv2021

Single-cell profiling of healthy human kidney reveals features of sex-based transcriptional programs and tissue-specific immunity

Maintaining organ homeostasis requires complex functional synergy between distinct cell types, a snapshot of which is glimpsed through the simultaneously broad and granular analysis provided by single-cell atlases. Knowledge of the transcriptional programs underpinning the complex and specialized functions of human kidney cell populations at homeostasis is limited by difficulty accessing healthy, fresh tissue. Here, we present a single-cell perspective of healthy human kidney from 19 living donors, with equal contribution from males and females, profiling the transcriptome of 27677 high-quality cells to map healthy kidney at high resolution. Our sex-balanced dataset revealed sex-based differences in gene expression within proximal tubular cells, specifically, increased anti-oxidant metallothionein genes in females and the predominance of aerobic metabolism-related genes in males. Functional differences in metabolism were confirmed between male and female proximal tubular cells, with male cells exhibiting higher oxidative phosphorylation and higher levels of energy precursor metabolites. Within the immune niche, we identified kidney-specific lymphocyte populations with unique transcriptional profiles indicative of kidney-adapted functions and validated findings by flow cytometry. We observed significant heterogeneity in resident myeloid populations and identified an MRC1+ LYVE1+ FOLR2+ C1QC+ population as the predominant myeloid population in healthy kidney. This study provides a detailed cellular map of healthy human kidney, revealing novel insights into the complexity of renal parenchymal cells and kidney-resident immune populations.

immunology↗

bioRxiv2021

Multiscale interactome analysis coupled with off-target drug predictions reveals drug repurposing candidates for human coronavirus disease

The COVID-19 pandemic has led to an urgent need for the identification of new antiviral drug therapies that can be rapidly deployed to treat patients with this disease. COVID-19 is caused by infection with the human coronavirus SARS-CoV-2. We developed a computational approach to identify new antiviral drug targets and repurpose clinically-relevant drug compounds for the treatment of COVID-19. Our approach is based on graph convolutional networks (GCN) and involves multiscale host-virus interactome analysis coupled to off-target drug predictions. Cellbased experimental assessment reveals several clinically-relevant repurposing drug candidates predicted by the in silico analyses to have antiviral activity against human coronavirus infection. In particular, we identify the MET inhibitor capmatinib as having potent and broad antiviral activity against several coronaviruses in a MET-independent manner, as well as novel roles for host cell proteins such as IRAK1/4 in supporting human coronavirus infection, which can inform further drug discovery studies.

cell biology↗

bioRxiv2020

Early temporal dynamics of cellular responses to SARS-CoV-2

Type I interferons (IFNs) are our first line of defence against a virus. Protein over-expression studies have suggested the ability of SARS-CoV-2 proteins to block IFN responses. Emerging data also suggest that timing and extent of IFN production is associated with manifestation of COVID-19 severity. In spite of progress in understanding how SARS-CoV-2 activates antiviral responses, mechanistic studies into wildtype SARS-CoV-2-mediated induction and inhibition of human type I IFN responses are lacking. Here we demonstrate that SARS-CoV-2 infection induces a mild type I IFN response in vitro and in moderate cases of COVID-19. In vitro stimulation of type I IFN expression and signaling in human airway epithelial cells is associated with activation of canonical transcriptions factors, and SARS-CoV-2 is unable to inhibit exogenous induction of these responses. Our data demonstrate that SARS-CoV-2 is not adept in blocking type I IFN responses and provide support for ongoing IFN clinical trials. O_FIG O_LINKSMALLFIG WIDTH=200 HEIGHT=200 SRC="FIGDIR/small/158154v2_ufig1.gif" ALT="Figure 1"> View larger version (35K): org.highwire.dtl.DTLVardef@193c540org.highwire.dtl.DTLVardef@7b106forg.highwire.dtl.DTLVardef@1741cfforg.highwire.dtl.DTLVardef@1bde68_HPS_FORMAT_FIGEXP M_FIG GRAPHICAL SUMMARY C_FIG

immunology↗

bioRxiv2020

Massively multiplex single-molecule oligonucleosome footprinting

Our understanding of the beads-on-a-string arrangement of nucleosomes has been built largely on high-resolution sequence-agnostic imaging methods and sequence-resolved bulk biochemical techniques. To bridge the divide between these approaches, we present the single-molecule adenine methylated oligonucleosome sequencing assay (SAMOSA). SAMOSA is a high-throughput single-molecule sequencing method that combines adenine methyltransferase footprinting and single-molecule real-time DNA sequencing to natively and nondestructively measure nucleosome positions on individual chromatin fibres. SAMOSA data allows unbiased classification of single-molecular states of nucleosome occupancy on individual chromatin fibres. We leverage this to estimate nucleosome regularity and spacing on single chromatin fibres genome-wide, at predicted transcription factor binding motifs, and across both active and silent human epigenomic domains. Our analyses suggest that chromatin is comprised of a diverse array of both regular and irregular single-molecular oligonucleosome patterns that differ subtly in their relative abundance across epigenomic domains. This irregularity is particularly striking in constitutive heterochromatin, which has typically been viewed as a conformationally static entity. Our proof-of-concept study provides a powerful new methodology for studying nucleosome organization at a previously intractable resolution, and offers up new avenues for modeling and visualizing higher-order chromatin structure. 1-sentence summaryHigh-throughput single-molecule real-time footprinting of chromatin arrays reveals heterogeneous patterns of oligonucleosome occupancy.

genomics↗

bioRxiv2020

Viral integration transforms chromatin to drive oncogenesis

Human papillomavirus (HPV) drives almost all cervical cancers and up to [~]70% of head and neck cancers. Frequent integration into the host genome occurs only for tumourigenic strains of HPV. We hypothesized that changes in the epigenome and transcriptome contribute to the tumourigenicity of HPV. We found that viral integration events often occurred along with changes in chromatin state and expression of genes near the integration site. We investigated whether introduction of new transcription factor binding sites due to HPV integration could invoke these changes. Some regions within the HPV genome, particularly the position of a conserved CTCF binding site, showed enriched chromatin accessibility signal. ChIP-seq revealed that the conserved CTCF binding site within the HPV genome bound CTCF in 4 HPV+ cancer cell lines. Significant changes in CTCF binding pattern and increases in chromatin accessibility occurred exclusively within 100 kbp of HPV integration sites. The chromatin changes co-occurred with out-sized changes in transcription and alternative splicing of local genes. We analyzed the essentiality of genes upregulated around HPV integration sites of The Cancer Genome Atlas (TCGA) HPV+ tumours. HPV integration upregulated genes which had significantly higher essentiality scores compared to randomly selected upregulated genes from the same tumours. Our results suggest that introduction of a new CTCF binding site due to HPV integration reorganizes chromatin and upregulates genes essential for tumour viability in some HPV+ tumours. These findings emphasize a newly recognized role of HPV integration in oncogenesis.

cancer biology↗

bioRxiv2018

Virtual ChIP-seq: Predicting transcription factor binding by learning from the transcriptome

MotivationIdentifying transcription factor binding sites is the first step in pinpointing non-coding mutations that disrupt the regulatory function of transcription factors and promote disease. ChIP-seq is the most common method for identifying binding sites, but performing it on patient samples is hampered by the amount of available biological material and the cost of the experiment. Existing methods for computational prediction of regulatory elements primarily predict binding in genomic regions with sequence similarity to known transcription factor sequence preferences. This has limited efficacy since most binding sites do not resemble known transcription factor sequence motifs, and many transcription factors are not even sequence-specific. ResultsWe developed Virtual ChIP-seq, which predicts binding of individual transcription factors in new cell types using an artificial neural network that integrates ChIP-seq results from other cell types and chromatin accessibility data in the new cell type. Virtual ChIP-seq also uses learned associations between gene expression and transcription factor binding at specific genomic regions. This approach outperforms methods that predict TF binding solely based on sequence preference, pre-dicting binding for 36 transcription factors (Matthews correlation coefficient > 0.3). AvailabilityThe datasets we used for training and validation are available at https://virchip.hoffmanlab.org. We have deposited in Zenodo the current version of our software (http://doi.org/10.5281/zenodo.1066928), datasets (http://doi.org/10.5281/zenodo.823297), predictions for 36 transcription factors on Roadmap Epigenomics cell types (http://doi.org/10.5281/zenodo.1455759), and predictions in Cistrome as well as ENCODE-DREAM in vivo TF Binding Site Prediction Challenge (http://doi.org/10.5281/zenodo.1209308).

bioinformatics↗