Search

Long noncoding RNAs (lncRNAs) involved in the regulation of antiviral innate immune responses need to be further identified. By functionally screening the lncRNAs in macrophages, here we identified lncRNA Malat1, abundant in the nucleus but significantly down-regulated after viral infection, as a negative regulator of antiviral type I IFN (IFN-I) production. Malat1 directly bound to the transactive response DNA-binding protein (TDP43) in the nucleus and prevented activation of TDP43 by blocking the activated caspase-3-mediated TDP43 cleavage to TDP35. The cleaved TDP35 increased the nuclear IRF3 protein level by binding and degrading Rbck1 pre-mRNA to prevent IRF3 proteasomal degradation upon viral infection, thus selectively promoting antiviral IFN-I production. Deficiency of Malat1 enhanced antiviral innate responses in vivo, accompanying the increased IFN-I production and reduced viral burden. Importantly, the reduced MALAT1, augmented IRF3, and increased IFNA mRNA were found in peripheral blood mononuclear cells (PBMCs) from systemic lupus erythematosus (SLE) patients. Therefore, the down-regulation of MALAT1 in virus-infected cells or in human cells from autoimmune diseases will increase host resistance against viral infection or lead to autoinflammatory interferonopathies via the increased type I IFN production. Our results demonstrate that the nuclear Malat1 suppresses antiviral innate responses by targeting TDP43 activation via RNA-RBP interactive network, adding insight to the molecular regulation of innate responses and autoimmune pathogenesis.

Query optimizers rely on accurate cardinality estimates to produce good execution plans. Despite decades of research, existing cardinality estimators are inaccurate for complex queries, due to making lossy modeling assumptions and not capturing inter-table correlations. In this work, we show that it is possible to learn the correlations across all tables in a database without any independence assumptions. We present NeuroCard, a join cardinality estimator that builds a single neural density estimator over an entire database. Leveraging join sampling and modern deep autoregressive models, NeuroCard makes no inter-table or inter-column independence assumptions in its probabilistic modeling. NeuroCard achieves orders of magnitude higher accuracy than the best prior methods (a new state-of-the-art result of 8.5$\times$ maximum error on JOB-light), scales to dozens of tables, while being compact in space (several MBs) and efficient to construct or update (seconds to minutes)., Comment: VLDB 2021

While SARS-CoV-2 pathogenesis has been intensively investigated, the host mechanisms of viral clearance and inflammation resolution are still elusive because of the ethical limitation of human studies based on COVID-19 convalescents. Here we infected Syrian hamsters by authentic SARS-CoV-2 and built an ideal model to simulate the natural recovery process of SARS-CoV-2 infection from severe pneumonia1,2. We developed and applied a spatial transcriptomic sequencing technique with subcellular resolution and tissue-scale extensibility, i.e., Stereo-seq3, together with single-cell RNA sequencing (scRNA-seq), to the entire lung lobes of 45 hamsters and obtained an elaborate map of the pulmonary spatiotemporal changes from acute infection, severe pneumonia to the late viral clearance and inflammation resolution. While SARS-CoV-2 infection caused massive damages to the hamster lungs, including naïve T cell infection and deaths related to lymphopenia, we identified a group of monocyte-derived proliferating Slamf9+Spp1+ macrophages, which were SARS-CoV-2 infection-inducible and cell death-resistant, recruiting neutrophils to clear viruses together. After viral clearance, the Slamf9+Spp1+ macrophages differentiated into Trem2+ and Fbp1+ macrophages, both responsible for inflammation resolution and replenishment of alveolar macrophages. The existence of this specific macrophage subpopulation and its descendants were validated by RNAscope in hamsters, immunofluorescence in hACE2 mice, and public human autopsy scRNA-seq data of COVID-19 patients. The spatiotemporal landscape of SARS-CoV-2 infection in hamster lungs and the identification of Slamf9+Spp1+ macrophages that is pivotal to viral clearance and inflammation resolution are important to better understand the critical molecular and cellular players of COVID-19 host defense and also develop potential interventions of COVID-19 immunopathology.

The helicase RIG-I initiates an antiviral immune response after recognition of pathogenic RNA. TRIM25, an E3 ubiquitin ligase, mediates K63-linked ubiquitination of RIG-I, which is crucial for RIG-I downstream signaling and the antiviral innate immune response. The components and mode of the RIG-I-initiated innate signaling remain to be fully understood. Here we identify a novel long noncoding RNA (Lnczc3h7a) that binds to TRIM25 and promotes RIG-I-mediated antiviral innate immune responses. Depletion of Lnczc3h7a impairs RIG-I signaling and the antiviral innate response to RNA viruses in vitro and in vivo. Mechanistically, Lnczc3h7a binds to both TRIM25 and activated RIG-I, serving as a molecular scaffold for stabilization of the RIG-I-TRIM25 complex at the early stage of viral infection. Lnczc3h7a facilitates TRIM25-mediated K63-linked ubiquitination of RIG-I and thus promotes downstream signaling transduction. Our findings reveal that host RNAs can enhance the response of innate immune sensors to foreign RNAs, ensuring effective antiviral defense.

Long noncoding RNAs (lncRNAs) involved in the regulation of antiviral innate immune responses need to be further identified. By functionally screening the lncRNAs in macrophages, here we identified lncRNA Malat1, abundant in the nucleus but significantly down-regulated after viral infection, as a negative regulator of antiviral type I IFN (IFN-I) production. Malat1 directly bound to the transactive response DNA-binding protein (TDP43) in the nucleus and prevented activation of TDP43 by blocking the activated caspase-3-mediated TDP43 cleavage to TDP35. The cleaved TDP35 increased the nuclear IRF3 protein level by binding and degrading Rbck1 pre-mRNA to prevent IRF3 proteasomal degradation upon viral infection, thus selectively promoting antiviral IFN-I production. Deficiency of Malat1 enhanced antiviral innate responses in vivo, accompanying the increased IFN-I production and reduced viral burden. Importantly, the reduced MALAT1, augmented IRF3, and increased IFNA mRNA were found in peripheral blood mononuclear cells (PBMCs) from systemic lupus erythematosus (SLE) patients. Therefore, the down-regulation of MALAT1 in virus-infected cells or in human cells from autoimmune diseases will increase host resistance against viral infection or lead to autoinflammatory interferonopathies via the increased type I IFN production. Our results demonstrate that the nuclear Malat1 suppresses antiviral innate responses by targeting TDP43 activation via RNA-RBP interactive network, adding insight to the molecular regulation of innate responses and autoimmune pathogenesis.

Corporations today collect data at an unprecedented and accelerating scale, making the need to run queries on large datasets increasingly important. Technologies such as columnar block-based data organization and compression have become standard practice in most commercial database systems. However, the problem of best assigning records to data blocks on storage is still open. For example, today's systems usually partition data by arrival time into row groups, or range/hash partition the data based on selected fields. For a given workload, however, such techniques are unable to optimize for the important metric of the number of blocks accessed by a query. This metric directly relates to the I/O cost, and therefore performance, of most analytical queries. Further, they are unable to exploit additional available storage to drive this metric down further. In this paper, we propose a new framework called a query-data routing tree, or qd-tree, to address this problem, and propose two algorithms for their construction based on greedy and deep reinforcement learning techniques. Experiments over benchmark and real workloads show that a qd-tree can provide physical speedups of more than an order of magnitude compared to current blocking schemes, and can reach within 2X of the lower bound for data skipping based on selectivity, while providing complete semantic descriptions of created blocks., ACM SIGMOD 2020

Cardinality estimation has long been grounded in statistical tools for density estimation. To capture the rich multivariate distributions of relational tables, we propose the use of a new type of high-capacity statistical model: deep autoregressive models. However, direct application of these models leads to a limited estimator that is prohibitively expensive to evaluate for range or wildcard predicates. To produce a truly usable estimator, we develop a Monte Carlo integration scheme on top of autoregressive models that can efficiently handle range queries with dozens of dimensions or more. Like classical synopses, our estimator summarizes the data without supervision. Unlike previous solutions, we approximate the joint data distribution without any independence assumptions. Evaluated on real-world datasets and compared against real systems and dominant families of techniques, our estimator achieves single-digit multiplicative error at tail, an up to 90$\times$ accuracy improvement over the second best method, and is space- and runtime-efficient., Comment: VLDB 2020. Updates since version 1: new title and new/revised content

The helicase RIG-I initiates an antiviral immune response after recognition of pathogenic RNA. TRIM25, an E3 ubiquitin ligase, mediates K63-linked ubiquitination of RIG-I, which is crucial for RIG-I downstream signaling and the antiviral innate immune response. The components and mode of the RIG-I-initiated innate signaling remain to be fully understood. Here we identify a novel long noncoding RNA (Lnczc3h7a) that binds to TRIM25 and promotes RIG-I-mediated antiviral innate immune responses. Depletion of Lnczc3h7a impairs RIG-I signaling and the antiviral innate response to RNA viruses in vitro and in vivo. Mechanistically, Lnczc3h7a binds to both TRIM25 and activated RIG-I, serving as a molecular scaffold for stabilization of the RIG-I-TRIM25 complex at the early stage of viral infection. Lnczc3h7a facilitates TRIM25-mediated K63-linked ubiquitination of RIG-I and thus promotes downstream signaling transduction. Our findings reveal that host RNAs can enhance the response of innate immune sensors to foreign RNAs, ensuring effective antiviral defense.

A text-to-speech synthesis system typically consists of multiple stages, such as a text analysis frontend, an acoustic model and an audio synthesis module. Building these components often requires extensive domain expertise and may contain brittle design choices. In this paper, we present Tacotron, an end-to-end generative text-to-speech model that synthesizes speech directly from characters. Given pairs, the model can be trained completely from scratch with random initialization. We present several key techniques to make the sequence-to-sequence framework perform well for this challenging task. Tacotron achieves a 3.82 subjective 5-scale mean opinion score on US English, outperforming a production parametric system in terms of naturalness. In addition, since Tacotron generates speech at the frame level, it's substantially faster than sample-level autoregressive methods., Comment: Submitted to Interspeech 2017. v2 changed paper title to be consistent with our conference submission (no content change other than typo fixes)

We present ZipG, a distributed memory-efficient graph store for serving interactive graph queries. ZipG achieves memory efficiency by storing the input graph data using a compressed representation. What differentiates ZipG from other graph stores is its ability to execute a wide range of graph queries directly on this compressed representation. ZipG can thus execute a larger fraction of queries in main memory, achieving query interactivity. ZipG exposes a minimal API that is functionally rich enough to implement published functionalities from several industrial graph stores. We demonstrate this by implementing and evaluating graph queries from Facebook TAO, LinkBench, Graph Search and several other workloads on top of ZipG. On a single server with 244GB memory, ZipG executes tens of thousands of queries from these workloads for raw graph data over half a TB; this leads to an order of magnitude (sometimes as much as 23×) higher throughput than Neo4j and Titan. We get similar gains in distributed settings compared to Titan.

The innate RNA sensor RIG-I is critical in the initiation of antiviral type I interferons (IFNs) production upon recognition of "non-self" viral RNAs. Here, we identify a host-derived, IFN-inducible long noncoding RNA, lnc-Lsm3b, that can compete with viral RNAs in the binding of RIG-I monomers and feedback inactivate the RIG-I innate function at late stage of innate response. Mechanistically, binding of lnc-Lsm3b restricts RIG-I protein's conformational shift and prevents downstream signaling, thereby terminating type I IFNs production. Multivalent structural motifs and long-stem structure are critical features of lnc-Lsm3b for RIG-I binding and inhibition. These data reveal a non-canonical self-recognition mode in the regulation of immune response and demonstrate an important role of an inducible "self" lncRNA acting as a potent molecular decoy actively saturating RIG-I binding sites to restrict the duration of "non-self" RNA-induced innate immune response and maintaining immune homeostasis, with potential utility in inflammatory disease management.

R is a popular statistical programming language with a number of extensions that support data processing and machine learning tasks. However, interactive data analysis in R is usually limited as the R runtime is single threaded and can only process data sets that fit in a single machine's memory. We present SparkR, an R package that provides a frontend to Apache Spark and uses Spark's distributed computation engine to enable large scale data analysis from the R shell. We describe the main design goals of SparkR, discuss how the high-level DataFrame API enables scalable computation and present some of the key details of our implementation.