nipalsMCIA: Flexible Multi-Block Dimensionality Reduction in R via Non-linear Iterative Partial Least Squares.

Motivation: With the increased reliance on multi-omics data for bulk and single cell analyses, the availability of robust approaches to perform unsupervised analysis for clustering, visualization, and feature selection is imperative. Joint dimensionality reduction methods can be applied to multi-omics datasets to derive a global sample embedding analogous to single-omic techniques such as Principal Components Analysis (PCA). Multiple co-inertia analysis (MCIA) is a method for joint dimensionality reduction that maximizes the covariance between block- and global-level embeddings. Current implementations for MCIA are not optimized for large datasets such such as those arising from single cell studies, and lack capabilities with respect to embedding new data. Results: We introduce nipalsMCIA, an MCIA implementation that solves the objective function using an extension to Non-linear Iterative Partial Least Squares (NIPALS), and shows significant speed-up over earlier implementations that rely on eigendecompositions for single cell multi-omics data. It also removes the dependence on an eigendecomposition for calculating the variance explained, and allows users to perform out-of-sample embedding for new data. nipalsMCIA provides users with a variety of pre-processing and parameter options, as well as ease of functionality for down-stream analysis of single-omic and global-embedding factors. Availability: nipalsMCIA is available as a BioConductor package at https://bioconductor.org/packages/release/bioc/html/nipalsMCIA.html, and includes detailed documentation and application vignettes. Supplementary Materials are available online.

Predicting gene expression state and prioritizing putative enhancers using 5hmC signal.

BACKGROUND: Like its parent base 5-methylcytosine (5mC), 5-hydroxymethylcytosine (5hmC) is a direct epigenetic modification of cytosines in the context of CpG dinucleotides. 5hmC is the most abundant oxidized form of 5mC, generated through the action of TET dioxygenases at gene bodies of actively-transcribed genes and at active or lineage-specific enhancers. Although such enrichments are reported for 5hmC, to date, predictive models of gene expression state or putative regulatory regions for genes using 5hmC have not been developed. RESULTS: Here, by using only 5hmC enrichment in genic regions and their vicinity, we develop neural network models that predict gene expression state across 49 cell types. We show that our deep neural network models distinguish high vs low expression state utilizing only 5hmC levels and these predictive models generalize to unseen cell types. Further, in order to leverage 5hmC signal in distal enhancers for expression prediction, we employ an Activity-by-Contact model and also develop a graph convolutional neural network model with both utilizing Hi-C data and 5hmC enrichment to prioritize enhancer-promoter links. These approaches identify known and novel putative enhancers for key genes in multiple immune cell subsets. CONCLUSIONS: Our work highlights the importance of 5hmC in gene regulation through proximal and distal mechanisms and provides a framework to link it to genome function. With the recent advances in 6-letter DNA sequencing by short and long-read techniques, profiling of 5mC and 5hmC may be done routinely in the near future, hence, providing a broad range of applications for the methods developed here.

Loop Catalog: a comprehensive HiChIP database of human and mouse samples.

HiChIP enables cost-effective and high-resolution profiling of regulatory and structural loops. To leverage the increasing number of publicly available HiChIP datasets from diverse cell lines and primary cells, we developed the Loop Catalog (https://loopcatalog.lji.org), a web-based database featuring HiChIP loop calls for 1319 samples across 133 studies and 44 high-resolution Hi-C loop calls. We demonstrate its utility in interpreting fine-mapped GWAS variants (SNP-to-gene linking), in identifying enriched sequence motifs and motif pairs at loop anchors, and in network-level analysis of loops connecting regulatory elements (community detection). Our comprehensive catalog, spanning over 4M unique 5kb loops, along with the accompanying analysis modalities constitutes an important resource for studies in gene regulation and genome organization.

A multi-omics systems vaccinology resource to develop and test computational models of immunity.

Systems vaccinology studies have identified factors affecting individual vaccine responses, but comparing these findings is challenging due to varying study designs. To address this lack of reproducibility, we established a community resource for comparing Bordetella pertussis booster responses and to host annual contests for predicting patients' vaccination outcomes. We report here on our experiences with the "dry-run" prediction contest. We found that, among 20+ models adopted from the literature, the most successful model predicting vaccination outcome was based on age alone. This confirms our concerns about the reproducibility of conclusions between different vaccinology studies. Further, we found that, for newly trained models, handling of baseline information on the target variables was crucial. Overall, multiple co-inertia analysis gave the best results of the tested modeling approaches. Our goal is to engage community in these prediction challenges by making data and models available and opening a public contest in August 2024.

Chromatin structure and var2csa - a tango in regulation of var gene expression in the human malaria parasite, Plasmodium falciparum?

Over the last few decades, novel methods have been developed to study how chromosome positioning within the nucleus may play a role in gene regulation. Adaptation of these methods in the human malaria parasite, Plasmodium falciparum, has recently led to the discovery that the three-dimensional structure of chromatin within the nucleus may be critical in controlling expression of virulence genes (var genes). Recent work has implicated an unusual, highly conserved var gene called var2csa in contributing to coordinated transcriptional switching, however how this gene functions in this capacity is unknown. To further understand how var2csa influences var gene switching, targeted DNA double-strand breaks (DSBs) within the sub-telomeric region of chromosome 12 were used to delete the gene and the surrounding chromosomal region. To characterize the changes in chromatin architecture stemming from this deletion and how these changes could affect var gene expression, we used a combination of RNA-seq, Chip-seq and Hi-C to pinpoint epigenetic and chromatin structural modifications in regions of differential gene expression. We observed a net gain of interactions in sub-telomeric regions and internal var gene regions following var2csa knockout, indicating an increase of tightly controlled heterochromatin structures. Our results suggest that disruption of var2csa results not only in changes in var gene transcriptional regulation but also a significant tightening of heterochromatin clusters thereby disrupting coordinated activation of var genes throughout the genome. Altogether our result confirms a strong link between the var2csa locus, chromatin structure and var gene expression.

LIGHT signaling through LTßR and HVEM in keratinocytes promotes psoriasis and atopic dermatitis-like skin inflammation.

Psoriasis (PS) and atopic dermatitis (AD) are common skin inflammatory diseases characterized by hyper-responsive keratinocytes. Although, some cytokines have been suggested to be specific for each disease, other cytokines might be central to both diseases. Here, we show that Tumor necrosis factor superfamily member 14 (TNFSF14), known as LIGHT, is required for experimental PS, similar to its requirement in experimental AD. Mice devoid of LIGHT, or deletion of either of its receptors, lymphotoxin ß receptor (LTßR) and herpesvirus entry mediator (HVEM), in keratinocytes, were protected from developing imiquimod-induced psoriatic features, including epidermal thickening and hyperplasia, and expression of PS-related genes. Correspondingly, in single cell RNA-seq analysis of PS patient biopsies, LTßR transcripts were found strongly expressed with HVEM in keratinocytes, and LIGHT was upregulated in T cells. Similar transcript expression profiles were also seen in AD biopsies, and LTßR deletion in keratinocytes also protected mice from allergen-induced AD features. Moreover, in vitro, LIGHT upregulated a broad spectrum of genes in human keratinocytes that are clinical features of both PS and AD skin lesions. Our data suggest that agents blocking LIGHT activity might be useful for therapeutic intervention in PS as well as in AD.

Atypical B cells consist of subsets with distinct functional profiles.

Atypical B cells are a population of activated B cells that are commonly enriched in individuals with chronic immune activation but are also part of a normal immune response to infection or vaccination. To better define the role of atypical B cells in the human adaptive immune response, we performed single-cell sequencing of transcriptomes, cell surface markers, and B cell receptors in individuals with chronic exposure to the malaria parasite Plasmodium falciparum, a condition known to lead to accumulation of circulating atypical B cells. We identified three previously uncharacterized populations of atypical B cells with distinct transcriptional and functional profiles and observed marked differences among these three subsets in their ability to produce immunoglobulin G upon T-cell-dependent activation. Our findings help explain the conflicting observations in prior studies regarding the function of atypical B cells and highlight their different roles in the adaptive immune response in chronic inflammatory conditions.

Lineage-specific 3D genome organization is assembled at multiple scales by IKAROS.

A generic level of chromatin organization generated by the interplay between cohesin and CTCF suffices to limit promiscuous interactions between regulatory elements, but a lineage-specific chromatin assembly that supersedes these constraints is required to configure the genome to guide gene expression changes that drive faithful lineage progression. Loss-of-function approaches in B cell precursors show that IKAROS assembles interactions across megabase distances in preparation for lymphoid development. Interactions emanating from IKAROS-bound enhancers override CTCF-imposed boundaries to assemble lineage-specific regulatory units built on a backbone of smaller invariant topological domains. Gain of function in epithelial cells confirms IKAROS' ability to reconfigure chromatin architecture at multiple scales. Although the compaction of the Igκ locus required for genome editing represents a function of IKAROS unique to lymphocytes, the more general function to preconfigure the genome to support lineage-specific gene expression and suppress activation of extra-lineage genes provides a paradigm for lineage restriction.

MuDCoD: multi-subject community detection in personalized dynamic gene networks from single-cell RNA sequencing.

MOTIVATION: With the wide availability of single-cell RNA-seq (scRNA-seq) technology, population-scale scRNA-seq datasets across multiple individuals and time points are emerging. While the initial investigations of these datasets tend to focus on standard analysis of clustering and differential expression, leveraging the power of scRNA-seq data at the personalized dynamic gene co-expression network level has the potential to unlock subject and/or time-specific network-level variation, which is critical for understanding phenotypic differences. Community detection from co-expression networks of multiple time points or conditions has been well-studied; however, none of the existing settings included networks from multiple subjects and multiple time points simultaneously. To address this, we develop Multi-subject Dynamic Community Detection (MuDCoD) for multi-subject community detection in personalized dynamic gene networks from scRNA-seq. MuDCoD builds on the spectral clustering framework and promotes information sharing among the networks of the subjects as well as networks at different time points. It clusters genes in the personalized dynamic gene networks and reveals gene communities that are variable or shared not only across time but also among subjects. RESULTS: Evaluation and benchmarking of MuDCoD against existing approaches reveal that MuDCoD effectively leverages apparent shared signals among networks of the subjects at individual time points, and performs robustly when there is no or little information sharing among the networks. Applications to population-scale scRNA-seq datasets of human-induced pluripotent stem cells during dopaminergic neuron differentiation and CD4+ T cell activation indicate that MuDCoD enables robust inference for identifying time-varying personalized gene modules. Our results illustrate how personalized dynamic community detection can aid in the exploration of subject-specific biological processes that vary across time. AVAILABILITY AND IMPLEMENTATION: MuDCoD is publicly available at https://github.com/bo1929/MuDCoD as a Python package. Implementation includes simulation and real-data experiments together with extensive documentation.

A systems vaccinology resource to develop and test computational models of immunity.

Computational models that predict an individual's response to a vaccine offer the potential for mechanistic insights and personalized vaccination strategies. These models are increasingly derived from systems vaccinology studies that generate immune profiles from human cohorts pre- and post-vaccination. Most of these studies involve relatively small cohorts and profile the response to a single vaccine. The ability to assess the performance of the resulting models would be improved by comparing their performance on independent datasets, as has been done with great success in other areas of biology such as protein structure predictions. To transfer this approach to system vaccinology studies, we established a prototype platform that focuses on the evaluation of Computational Models of Immunity to Pertussis Booster vaccinations (CMI-PB). A community resource, CMI-PB generates experimental data for the explicit purpose of model evaluation, which is performed through a series of annual data releases and associated contests. We here report on our experience with the first such 'dry run' for a contest where the goal was to predict individual immune responses based on pre-vaccination multi-omic profiles. Over 30 models adopted from the literature were tested, but only one was predictive, and was based on age alone. The performance of new models built using CMI-PB training data was much better, but varied significantly based on the choice of pre-vaccination features used and the model building strategy. This suggests that previously published models developed for other vaccines do not generalize well to Pertussis Booster vaccination. Overall, these results reinforced the need for comparative analysis across models and datasets that CMI-PB aims to achieve. We are seeking wider community engagement for our first public prediction contest, which will open in early 2024.

Rewiring of the 3D genome during acquisition of carboplatin resistance in a triple-negative breast cancer patient-derived xenograft.

Changes in the three-dimensional (3D) structure of the genome are an emerging hallmark of cancer. Cancer-associated copy number variants and single nucleotide polymorphisms promote rewiring of chromatin loops, disruption of topologically associating domains (TADs), active/inactive chromatin state switching, leading to oncogene expression and silencing of tumor suppressors. However, little is known about 3D changes during cancer progression to a chemotherapy-resistant state. We integrated chromatin conformation capture (Hi-C), RNA-seq, and whole-genome sequencing obtained from triple-negative breast cancer patient-derived xenograft primary tumors (UCD52) and carboplatin-resistant samples and found increased short-range (< 2 Mb) interactions, chromatin looping, formation of TAD, chromatin state switching into a more active state, and amplification of ATP-binding cassette transporters. Transcriptome changes suggested the role of long-noncoding RNAs in carboplatin resistance. Rewiring of the 3D genome was associated with TP53, TP63, BATF, FOS-JUN family of transcription factors and led to activation of aggressiveness-, metastasis- and other cancer-related pathways. Integrative analysis highlighted increased ribosome biogenesis and oxidative phosphorylation, suggesting the role of mitochondrial energy metabolism. Our results suggest that 3D genome remodeling may be a key mechanism underlying carboplatin resistance.

3D genome mapping identifies subgroup-specific chromosome conformations and tumor-dependency genes in ependymoma.

Ependymoma is a tumor of the brain or spinal cord. The two most common and aggressive molecular groups of ependymoma are the supratentorial ZFTA-fusion associated and the posterior fossa ependymoma group A. In both groups, tumors occur mainly in young children and frequently recur after treatment. Although molecular mechanisms underlying these diseases have recently been uncovered, they remain difficult to target and innovative therapeutic approaches are urgently needed. Here, we use genome-wide chromosome conformation capture (Hi-C), complemented with CTCF and H3K27ac ChIP-seq, as well as gene expression and DNA methylation analysis in primary and relapsed ependymoma tumors, to identify chromosomal conformations and regulatory mechanisms associated with aberrant gene expression. In particular, we observe the formation of new topologically associating domains ('neo-TADs') caused by structural variants, group-specific 3D chromatin loops, and the replacement of CTCF insulators by DNA hyper-methylation. Through inhibition experiments, we validate that genes implicated by these 3D genome conformations are essential for the survival of patient-derived ependymoma models in a group-specific manner. Thus, this study extends our ability to reveal tumor-dependency genes by 3D genome conformations even in tumors that lack targetable genetic alterations.

Lymphotoxin beta receptor signaling directly controls airway smooth muscle deregulation and asthmatic lung dysfunction.

BACKGROUND: Dysregulation of airway smooth muscle cells (ASM) is central to the severity of asthma. Which molecules dominantly control ASM in asthma is unclear. High levels of the cytokine LIGHT (aka TNFSF14) have been linked to asthma severity and lower baseline predicted FEV1 percentage, implying that signals through its receptors might directly control ASM dysfunction. OBJECTIVE: Our study sought to determine whether signaling via lymphotoxin beta receptor (LTßR) or herpesvirus entry mediator from LIGHT dominantly drives ASM hyperreactivity induced by allergen. METHODS: Conditional knockout mice deficient for LTßR or herpesvirus entry mediator in smooth muscle cells were used to determine their role in ASM deregulation and airway hyperresponsiveness (AHR) in vivo. Human ASM were used to study signals induced by LTßR. RESULTS: LTßR was strongly expressed in ASM from normal and asthmatic subjects compared to several other receptors implicated in smooth muscle deregulation. Correspondingly, conditional deletion of LTßR only in smooth muscle cells in smMHCCreLTßRfl/fl mice minimized changes in their numbers and mass as well as AHR induced by house dust mite allergen in a model of severe asthma. Intratracheal LIGHT administration independently induced ASM hypertrophy and AHR in vivo dependent on direct LTßR signals to ASM. LIGHT promoted contractility, hypertrophy, and hyperplasia of human ASM in vitro. Distinguishing LTßR from the receptors for IL-13, TNF, and IL-17, which have also been implicated in smooth muscle dysregulation, LIGHT promoted NF-κB-inducing kinase-dependent noncanonical nuclear factor kappa-light-chain enhancer of activated B cells in ASM in vitro, leading to sustained accumulation of F-actin, phosphorylation of myosin light chain kinase, and contractile activity. CONCLUSIONS: LTßR signals directly and dominantly drive airway smooth muscle hyperresponsiveness relevant for pathogenesis of airway remodeling in severe asthma.

Brd2 is dispensable for genome compartmentalization and replication timing.

Replication Timing (RT) refers to the temporal order in which the genome is replicated during S phase. Early replicating regions correlate with the transcriptionally active, accessible euchromatin (A) compartment, while late replicating regions correlate with the heterochromatin (B) compartment and repressive histone marks. Previously, widespread A/B genome compartmentalization changes were reported following Brd2 depletion. Since RT and A/B compartmentalization are two of the most highly correlated chromosome properties, we evaluated the effects of Brd2 depletion on RT. We performed E/L Repli-Seq following Brd2 depletion in the previously described Brd2 conditional degron cell line and found no significant alterations in RT after Brd2 KD. This finding prompted us to re-analyze the Micro-C data from the previous publication. We report that we were unable to detect any compartmentalization changes in Brd2 depleted cells compared to DMSO control using the same data. Taken together, our findings demonstrate that Brd2 depletion alone does not affect A/B compartmentalization or RT in mouse embryonic stem cells.

dcHiC detects differential compartments across multiple Hi-C datasets.

The compartmental organization of mammalian genomes and its changes play important roles in distinct biological processes. Here, we introduce dcHiC, which utilizes a multivariate distance measure to identify significant changes in compartmentalization among multiple contact maps. Evaluating dcHiC on four collections of bulk and single-cell contact maps from in vitro mouse neural differentiation (n = 3), mouse hematopoiesis (n = 10), human LCLs (n = 20) and post-natal mouse brain development (n = 3 stages), we show its effectiveness and sensitivity in detecting biologically relevant changes, including those orthogonally validated. dcHiC reported regions with dynamically regulated genes associated with cell identity, along with correlated changes in chromatin states, subcompartments, replication timing and lamin association. With its efficient implementation, dcHiC enables high-resolution compartment analysis as well as standalone browser visualization, differential interaction identification and time-series clustering. dcHiC is an essential addition to the Hi-C analysis toolbox for the ever-growing number of bulk and single-cell contact maps. Available at: https://github.com/ay-lab/dcHiC .

A double-negative thymocyte-specific enhancer augments Notch1 signaling to direct early T cell progenitor expansion, lineage restriction and ß-selection.

T cell differentiation requires Notch1 signaling. In the present study, we show that an enhancer upstream of Notch1 active in double-negative (DN) mouse thymocytes is responsible for raising Notch1 signaling intrathymically. This enhancer is required to expand multipotent progenitors intrathymically while delaying early differentiation until lineage restrictions have been established. Early thymic progenitors lacking the enhancer show accelerated differentiation through the DN stages and increased frequency of B, innate lymphoid (IL) and natural killer (NK) cell differentiation. Transcription regulators for T cell lineage restriction and commitment are expressed normally, but IL and NK cell gene expression persists after T cell lineage commitment and T cell receptor ß VDJ recombination, Cd3 expression and ß-selection have been impaired. This Notch1 enhancer is inactive in double-positive (DP) thymocytes. Its aberrant reactivation at this stage in Ikaros mutants is required for leukemogenesis. Thus, the DN-specific Notch1 enhancer harnesses the regulatory architecture of DN and DP thymocytes to achieve carefully orchestrated changes in Notch1 signaling required for early lineage restrictions and normal T cell differentiation.

Intermittent PI3Kδ inhibition sustains anti-tumour immunity and curbs irAEs.

Phosphoinositide 3-kinase δ (PI3Kδ) has a key role in lymphocytes, and inhibitors that target this PI3K have been approved for treatment of B cell malignancies1-3. Although studies in mouse models of solid tumours have demonstrated that PI3Kδ inhibitors (PI3Kδi) can induce anti-tumour immunity4,5, its effect on solid tumours in humans remains unclear. Here we assessed the effects of the PI3Kδi AMG319 in human patients with head and neck cancer in a neoadjuvant, double-blind, placebo-controlled randomized phase II trial (EudraCT no. 2014-004388-20). PI3Kδ inhibition decreased the number of tumour-infiltrating regulatory T (Treg) cells and enhanced the cytotoxic potential of tumour-infiltrating T cells. At the tested doses of AMG319, immune-related adverse events (irAEs) required treatment to be discontinued in 12 out of 21 of patients treated with AMG319, suggestive of systemic effects on Treg cells. Accordingly, in mouse models, PI3Kδi decreased the number of Treg cells systemically and caused colitis. Single-cell RNA-sequencing analysis revealed a PI3Kδi-driven loss of tissue-resident colonic ST2 Treg cells, accompanied by expansion of pathogenic T helper 17 (TH17) and type 17 CD8+ T (TC17) cells, which probably contributed to toxicity; this points towards a specific mode of action for the emergence of irAEs. A modified treatment regimen with intermittent dosing of PI3Kδi in mouse models led to a significant decrease in tumour growth without inducing pathogenic T cells in colonic tissue, indicating that alternative dosing regimens might limit toxicity.

Single-cell eQTL analysis of activated T cell subsets reveals activation and cell type-dependent effects of disease-risk variants.

The impact of genetic variants on cells challenged in biologically relevant contexts has not been fully explored. Here, we activated CD4+ T cells from 89 healthy donors and performed a single-cell RNA sequencing assay with >1 million cells to examine cell type-specific and activation-dependent effects of genetic variants. Single-cell expression quantitative trait loci (sc-eQTL) analysis of 19 distinct CD4+ T cell subsets showed that the expression of over 4000 genes is significantly associated with common genetic polymorphisms and that most of these genes show their most prominent effects in specific cell types. These genes included many that encode for molecules important for activation, differentiation, and effector functions of T cells. We also found new gene associations for disease-risk variants identified from genome-wide association studies and highlighted the cell types in which their effects are most prominent. We found that biological sex has a major influence on activation-dependent gene expression in CD4+ T cell subsets. Sex-biased transcripts were significantly enriched in several pathways that are essential for the initiation and execution of effector functions by CD4+ T cells like TCR signaling, cytokines, cytokine receptors, costimulatory, apoptosis, and cell-cell adhesion pathways. Overall, this DICE (Database of Immune Cell Expression, eQTLs, and Epigenomics) subproject highlights the power of sc-eQTL studies for simultaneously exploring the activation and cell type-dependent effects of common genetic variants on gene expression (https://dice-database.org).