Efficacy of JAK1/2 inhibition in murine immune bone marrow failure.

Immune aplastic anemia (AA) is a severe blood disease characterized by T-lymphocyte- mediated stem cell destruction. Hematopoietic stem cell transplantation and immunosuppression are effective, but they entail costs and risks, and are not always successful. The Janus kinase (JAK) 1/2 inhibitor ruxolitinib (RUX) suppresses cytotoxic T-cell activation and inhibits cytokine production in models of graft-versus-host disease. We tested RUX in murine immune AA for potential therapeutic benefit. After infusion of lymph node (LN) cells mismatched at the major histocompatibility complex [C67BL/6 (B6)⇒CByB6F1], RUX, administered as a food additive (Rux-chow), attenuated bone marrow hypoplasia, ameliorated peripheral blood pancytopenia, preserved hematopoietic progenitors, and prevented mortality, when used either prophylactically or therapeutically. RUX suppressed the infiltration, proliferation, and activation of effector T cells in the bone marrow and mitigated Fas-mediated apoptotic destruction of target hematopoietic cells. Similar effects were obtained when Rux-chow was fed to C.B10 mice in a minor histocompatibility antigen mismatched (B6⇒C.B10) AA model. RUX only modestly suppressed lymphoid and erythroid hematopoiesis in normal and irradiated CByB6F1 mice. Our data support clinical trials of JAK/STAT inhibitors in human AA and other immune bone marrow failure syndromes.

CAISC: A software to integrate copy number variations and single nucleotide mutations for genetic heterogeneity profiling and subclone detection by single-cell RNA sequencing.

BACKGROUND: Although both copy number variations (CNVs) and single nucleotide variations (SNVs) detected by single-cell RNA sequencing (scRNA-seq) are used to study intratumor heterogeneity and detect clonal groups, a software that integrates these two types of data in the same cells is unavailable. RESULTS: We developed Clonal Architecture with Integration of SNV and CNV (CAISC), an R package for scRNA-seq data analysis that clusters single cells into distinct subclones by integrating CNV and SNV genotype matrices using an entropy weighted approach. The performance of CAISC was tested on simulation data and four real datasets, which confirmed its high accuracy in sub-clonal identification and assignment, including subclones which cannot be identified using one type of data alone. Furthermore, integration of SNV and CNV allowed for accurate examination of expression changes between subclones, as demonstrated by the results from trisomy 8 clones of the myelodysplastic syndromes (MDS) dataset. CONCLUSIONS: CAISC is a powerful tool for integration of CNV and SNV data from scRNA-seq to identify clonal clusters with better accuracy than obtained from a single type of data. CAISC allows users to interactively examine clonal assignments.

Comprehensive network modeling from single cell RNA sequencing of human and mouse reveals well conserved transcription regulation of hematopoiesis.

BACKGROUND: Presently, there is no comprehensive analysis of the transcription regulation network in hematopoiesis. Comparison of networks arising from gene co-expression across species can facilitate an understanding of the conservation of functional gene modules in hematopoiesis. RESULTS: We used single-cell RNA sequencing to profile bone marrow from human and mouse, and inferred transcription regulatory networks in each species in order to characterize transcriptional programs governing hematopoietic stem cell differentiation. We designed an algorithm for network reconstruction to conduct comparative transcriptomic analysis of hematopoietic gene co-expression and transcription regulation in human and mouse bone marrow cells. Co-expression network connectivity of hematopoiesis-related genes was found to be well conserved between mouse and human. The co-expression network showed "small-world" and "scale-free" architecture. The gene regulatory network formed a hierarchical structure, and hematopoiesis transcription factors localized to the hierarchy's middle level. CONCLUSIONS: Transcriptional regulatory networks are well conserved between human and mouse. The hierarchical organization of transcription factors may provide insights into hematopoietic cell lineage commitment, and to signal processing, cell survival and disease initiation.

Conventional Co-Housing Modulates Murine Gut Microbiota and Hematopoietic Gene Expression.

Specific-pathogen-free (SPF) mice have improved hematopoietic characteristics relative to germ-free mice, however, it is not clear whether improvements in hematopoietic traits will continue when the level of microorganism exposure is further increased. We co-housed SPF C57BL/6 mice in a conventional facility (CVT) and found a significant increase in gut microbiota diversity along with increased levels of myeloid cells and T cells, especially effector memory T cells. Through single cell RNA sequencing of sorted KL (c-Kit+Lin-) cells, we imputed a decline in long-term hematopoietic stem cells and an increase in granulocyte-monocyte progenitors in CVT mice with up-regulation of genes associated with cell survival. Bone marrow transplantation through competitive repopulation revealed a significant increase in KSL (c-Kit+Sca-1+Lin-) cell reconstitution in recipients of CVT donor cells which occurred when donors were co-housed for both one and twelve months. However, there was minimal to no gain in mature blood cell engraftment in recipients of CVT donor cells relative to those receiving SPF donor cells. We conclude that co-housing SPF mice with mice born in a conventional facility increased gut microbiota diversity, augmented myeloid cell production and T cell activation, stimulated KSL cell reconstitution, and altered hematopoietic gene expression.

Single-cell RNA-seq reveals a distinct transcriptome signature of aneuploid hematopoietic cells.

Cancer cells frequently exhibit chromosomal abnormalities. Specific cytogenetic aberrations often are predictors of outcome, especially in hematologic neoplasms, such as monosomy 7 in myeloid malignancies. The functional consequences of aneuploidy at the cellular level are difficult to assess because of a lack of convenient markers to distinguish abnormal from diploid cells. We performed single-cell RNA sequencing (scRNA-seq) to study hematopoietic stem and progenitor cells from the bone marrow of 4 healthy donors and 5 patients with bone marrow failure and chromosome gain or loss. In total, transcriptome sequences were obtained from 391 control cells and 588 cells from patients. We characterized normal hematopoiesis as binary differentiation from stem cells to erythroid and myeloid-lymphoid pathways. Aneuploid cells were distinguished from diploid cells in patient samples by computational analyses of read fractions and gene expression of individual chromosomes. We confirmed assignment of aneuploidy to individual cells quantitatively, by copy-number variation, and qualitatively, by loss of heterozygosity. When we projected patients' single cells onto the map of normal hematopoiesis, diverse patterns were observed, broadly reflecting clinical phenotypes. Patients' monosomy 7 cells showed downregulation of genes involved in immune response and DNA damage checkpoint and apoptosis pathways, which may contribute to the clonal expansion of monosomy 7 cells with accumulated gene mutations. scRNA-seq is a powerful technique through which to infer the functional consequences of chromosome gain and loss and explore gene targets for directed therapy.

Long noncoding RNAs of single hematopoietic stem and progenitor cells in healthy and dysplastic human bone marrow.

Long noncoding RNAs (lncRNAs) are regulators of cell differentiation and development. The lncRNA transcriptome in human hematopoietic stem and progenitor cells is not comprehensively defined. We investigated lncRNAs in 979 human bone marrow-derived CD34+ cells by single cell RNA sequencing followed by de novo transcriptome reconstruction. We identified 3,173 lncRNAs in total, among which 2,365 were previously unknown, and we characterized lncRNA stem, differentiation, and maturation signatures. lncRNA expression exhibited high cell-to-cell variation, which was only apparent in single cell analysis. lncRNA expression followed a lineage-specific and highly dynamic pattern during early hematopoiesis. lncRNAs in hematopoietic cells closely correlated with protein-coding genes of known functions in the regulation of hematopoiesis and cell fate decisions, and the potential regulatory roles of lncRNAs in hematopoiesis were imputed by projection from protein-coding genes with a "guilt-by-association" approach. We characterized lncRNAs preferentially expressed in hematopoietic stem cells and in various downstream differentiated lineage progenitors. We also profiled lncRNA expression in single cells from patients with myelodysplastic syndromes and in aneuploid cells in particular. Our study provides a global view of lncRNAs in human hematopoietic stem and progenitor cells. We observed a highly ordered pattern of lncRNA expression and participation in regulation of early hematopoiesis, and coordinate aberrant messenger RNA and lncRNA transcriptomes in dysplastic hematopoiesis. (Registered at clinicaltrials.gov with identifiers: 00001620, 00001397).

Abnormal RNA splicing and genomic instability after induction of DNMT3A mutations by CRISPR/Cas9 gene editing.

DNA methyltransferase 3A (DNMT3A) mediates de novo DNA methylation. Mutations in DNMT3A are associated with hematological malignancies, most frequently acute myeloid leukemia. DNMT3A mutations are hypothesized to establish a pre-leukemic state, rendering cells vulnerable to secondary oncogenic mutations and malignant transformation. However, the mechanisms by which DNMT3A mutations contribute to leukemogenesis are not well-defined. Here, we successfully created four DNMT3A-mutated K562 cell lines with frameshift mutations resulting in truncated DNMT3A proteins. DNMT3A-mutated cell lines exhibited significantly impaired growth and increased apoptotic activity compared to wild-type (WT) cells. Consistent with previous studies, DNMT3A-mutated cells displayed impaired differentiation capacity. RNA-seq was used to compare transcriptomes of DNMT3A-mutated and WT cells; DNMT3A ablation resulted in downregulation of genes involved in spliceosome function, causing dysfunction of RNA splicing. Unexpectedly, we observed DNMT3A-mutated cells to exhibit marked genomic instability and an impaired DNA damage response compared to WT. CRISPR/Cas9-mediated DNMT3A-mutated K562 cells may be used to model effects of DNMT3A mutations in human cells. Our findings implicate aberrant splicing and induction of genomic instability as potential mechanisms by which DNMT3A mutations might predispose to malignancy.

Cross-Sectional Positive Association of Serum Lipids and Blood Pressure With Serum Sodium Within the Normal Reference Range of 135-145 mmol/L.

OBJECTIVE: Serum sodium concentration is maintained by osmoregulation within normal range of 135 to 145 mmol/L. Previous analysis of data from the ARIC study (Atherosclerosis Risk in Communities) showed association of serum sodium with the 10-year risk scores of coronary heart disease and stroke. Current study evaluated the association of within-normal-range serum sodium with cardiovascular risk factors. APPROACH AND RESULTS: Only participants who did not take cholesterol or blood pressure medications and had sodium within normal 135 to 145 mmol/L range were included (n=8615), and the cohort was stratified based on race, sex, and smoking status. Multiple linear regression analysis of data from ARIC study was performed, with adjustment for age, blood glucose, insulin, glomerular filtration rate, body mass index, waist to hip ratio, and calorie intake. The analysis showed positive associations with sodium of total cholesterol, low-density lipoprotein cholesterol, and total cholesterol to high-density lipoprotein cholesterol ratio; apolipoprotein B; and systolic and diastolic blood pressure. Increases in lipids and blood pressure associated with 10 mmol/L increase in sodium are similar to the increases associated with 7 to 10 years of aging. Analysis of sodium measurements made 3 years apart demonstrated that it is stable within 2 to 3 mmol/L, explaining its association with long-term health outcomes. Furthermore, elevated sodium promoted lipid accumulation in cultured adipocytes, suggesting direct causative effects on lipid metabolism. CONCLUSIONS: Serum sodium concentration is a cardiovascular risk factor even within the normal reference range. Thus, decreasing sodium to the lower end of the normal range by modification of water and salt intake is a personalizable strategy for decreasing cardiovascular risks.

[Differential expression and bioinformatics analysis of microRNA in peripheral blood mononuclear cells from patients with aplastic anemia].

OBJECTIVE: To study differential expression of microRNA (miRNA) in peripheral blood mononuclear cells (PBMNC) from patients with different types of aplastic anemia (AA) and explore the role of miRNA in the pathogenesis of AA. METHODS: miRNA microarray were used to determine the differential expression profile of miRNA in PBMNC from patients with AA. Real-time quantitative polymerase china reaction (RQ-PCR) was used to verify the differential expression of miRNA. Candidate miRNA were analyzed with bioinformatics tools. RESULTS: Compared with the normal controls, 6 miRNAs were up-regulated and 10 were down-regulated in patients with severe aplastic anemia (SAA), while 24 miRNAs were up-regulated and 12 were down-regulated in patients with chronic non-severe aplastic anemia (CAA). Compared with CAA patients, 4 miRNAs were up-regulated and 11 were down-regulated in SAA patients. Compared with normal controls, 3 miRNAs were up-regulated and 4 were down-regulated in both SAA and CAA patients. As verified by RQ-PCR, expression of miR-155-5p and miR-1260b were increased in both CAA and SAA patients compared with the normal controls (P<0.01). The expression of miR-155-5p and miR-1260b of CAA patients were higher than that of SAA patients (P<0.01). Bioinformatics analysis showed that target genes of miR-155-5p and miR-1260b may be involved in regulation of cell metabolism, gene expression and transcription, TNF signaling pathway, B cell receptor signaling pathway, Fc gamma R-mediated phagocytosis and other signaling process. CONCLUSION: There are characteristic differential expression profiles of miRNA in PBMNC from CAA and SAA patients, in which miRNA-155-5p and miRNA-1260b are both up-regulated. The common target gene predicted for miRNA-155-5p and miRNA-1260b is ETS1. miRNA-155-5p and miRNA-1260b may act synergistically to inhibit the expression of ETS1 and promote differentiation of Th17, therefore play an important role in the pathogenesis of AA.

Peptide affinity analysis of proteins that bind to an unstructured NH2-terminal region of the osmoprotective transcription factor NFAT5.

NFAT5 is an osmoregulated transcription factor that particularly increases expression of genes involved in protection against hypertonicity. Transcription factors often contain unstructured regions that bind co-regulatory proteins that are crucial for their function. The NH2-terminal region of NFAT5 contains regions predicted to be intrinsically disordered. We used peptide aptamer-based affinity chromatography coupled with mass spectrometry to identify protein preys pulled down by one or more overlapping 20 amino acid peptide baits within a predicted NH2-terminal unstructured region of NFAT5. We identify a total of 467 unique protein preys that associate with at least one NH2-terminal peptide bait from NFAT5 in either cytoplasmic or nuclear extracts from HEK293 cells treated with elevated, normal, or reduced NaCl concentrations. Different sets of proteins are pulled down from nuclear vs. cytoplasmic extracts. We used GeneCards to ascertain known functions of the protein preys. The protein preys include many that were previously known, but also many novel ones. Consideration of the novel ones suggests many aspects of NFAT5 regulation, interaction and function that were not previously appreciated, for example, hypertonicity inhibits NFAT5 by sumoylating it and the NFAT5 protein preys include components of the CHTOP complex that desumoylate proteins, an action that should contribute to activation of NFAT5.

Identification of highly synchronized subnetworks from gene expression data.

BACKGROUND: There has been a growing interest in identifying context-specific active protein-protein interaction (PPI) subnetworks through integration of PPI and time course gene expression data. However the interaction dynamics during the biological process under study has not been sufficiently considered previously. METHODS: Here we propose a topology-phase locking (TopoPL) based scoring metric for identifying active PPI subnetworks from time series expression data. First the temporal coordination in gene expression changes is evaluated through phase locking analysis; The results are subsequently integrated with PPI to define an activity score for each PPI subnetwork, based on individual member expression, as well topological characteristics of the PPI network and of the expression temporal coordination network; Lastly, the subnetworks with the top scores in the whole PPI network are identified through simulated annealing search. RESULTS: Application of TopoPL to simulated data and to the yeast cell cycle data showed that it can more sensitively identify biologically meaningful subnetworks than the method that only utilizes the static PPI topology, or the additive scoring method. Using TopoPL we identified a core subnetwork with 49 genes important to yeast cell cycle. Interestingly, this core contains a protein complex known to be related to arrangement of ribosome subunits that exhibit extremely high gene expression synchronization. CONCLUSIONS: Inclusion of interaction dynamics is important to the identification of relevant gene networks.

Granulocytic myeloid-derived suppressor cells to prevent and treat murine immune-mediated bone marrow failure.

Myeloid-derived suppressor cells (MDSCs) are immature myeloid cells that originate in the bone marrow (BM) and have immunoregulatory functions. MDSCs have been implicated in the pathogenesis of several autoimmune diseases but have not been investigated in immune aplastic anemia (AA). We examined the roles of granulocytic-MDSCs (G-MDSCs) in murine models of human AA and BM failure (BMF). As both prophylaxis and therapy, BM-derived G-MDSCs improved pancytopenia and BM cellularity and suppressed BM T-cell infiltration in major histocompatibility complex (MHC)-matched C.B10 BMF mice. These effects were not obtained in the MHC-mismatched CByB6F1 AA model, likely because of MHC disparity between G-MDSCs and donor T cells. Single-cell RNA sequencing demonstrated that G-MDSCs downregulated cell cycle-related genes in BM-infiltrated T cells, consistent with suppression of T-cell proliferation by G-MDSCs through reactive oxygen species pathways. Clearance of G-MDSCs in the MHC-mismatched CByB6F1 model using anti-Ly6G antibody facilitated T cell-mediated BM destruction, suggesting an intrinsic immunosuppressive property of G-MDSCs. However, the same anti-Ly6G antibody in the MHC-matched C.B10 AA model mildly mitigated BMF, associated with expansion of an intermediate Ly6G population. Our results demonstrate that G-MDSC eradication and therapeutic efficacy are immune context-dependent.

Early activation of inflammatory pathways in UBA1-mutated hematopoietic stem and progenitor cells in VEXAS.

VEXAS (vacuoles, E1 enzyme, X-linked, autoinflammatory, somatic) syndrome is a pleiotropic, severe autoinflammatory disease caused by somatic mutations in the ubiquitin-like modifier activating enzyme 1 (UBA1) gene. To elucidate VEXAS pathophysiology, we performed transcriptome sequencing of single bone marrow mononuclear cells and hematopoietic stem and progenitor cells (HSPCs) from VEXAS patients. HSPCs are biased toward myeloid (granulocytic) differentiation, and against lymphoid differentiation in VEXAS. Activation of multiple inflammatory pathways (interferons and tumor necrosis factor alpha) occurs ontogenically early in primitive hematopoietic cells and particularly in the myeloid lineage in VEXAS, and inflammation is prominent in UBA1-mutated cells. Dysregulation in protein degradation likely leads to higher stress response in VEXAS HSPCs, which positively correlates with inflammation. TCR usage is restricted and there are increased cytotoxicity and IFN-γ signaling in T cells. In VEXAS syndrome, both aberrant inflammation and myeloid predominance appear intrinsic to hematopoietic stem cells mutated in UBA1.

CellCallEXT: Analysis of Ligand-Receptor and Transcription Factor Activities in Cell-Cell Communication of Tumor Immune Microenvironment.

(1) Background: Single-cell RNA sequencing (scRNA-seq) data are useful for decoding cell-cell communication. CellCall is a tool that is used to infer inter- and intracellular communication pathways by integrating paired ligand-receptor (L-R) and transcription factor (TF) activities from steady-state data and thus cannot directly handle two-condition comparisons. For tumor and healthy status, it can only individually analyze cells from tumor or healthy tissue and examine L-R pairs only identified in either tumor or healthy controls, but not both together. Furthermore, CellCall is highly affected by gene expression specificity in tissues. (2) Methods: CellCallEXT is an extension of CellCall that deconvolutes intercellular communication and related internal regulatory signals based on scRNA-seq. Information on Reactome was retrieved and integrated with prior knowledge of L-R-TF signaling and gene regulation datasets of CellCall. (3) Results: CellCallEXT was successfully applied to examine tumors and immune cell microenvironments and to identify the altered L-R pairs and downstream gene regulatory networks among immune cells. Application of CellCallEXT to scRNA-seq data from patients with deficiency of adenosine deaminase 2 demonstrated its ability to impute dysfunctional intercellular communication and related transcriptional factor activities. (4) Conclusions: CellCallEXT provides a practical tool to examine intercellular communication in disease based on scRNA-seq data.

Time-Varying Gene Expression Network Analysis Reveals Conserved Transition States in Hematopoietic Differentiation between Human and Mouse.

(1) Background: analyses of gene networks can elucidate hematopoietic differentiation from single-cell gene expression data, but most algorithms generate only a single, static network. Because gene interactions change over time, it is biologically meaningful to examine time-varying structures and to capture dynamic, even transient states, and cell-cell relationships. (2) Methods: a transcriptomic atlas of hematopoietic stem and progenitor cells was used for network analysis. After pseudo-time ordering with Monocle 2, LOGGLE was used to infer time-varying networks and to explore changes of differentiation gene networks over time. A range of network analysis tools were used to examine properties and genes in the inferred networks. (3) Results: shared characteristics of attributes during the evolution of differentiation gene networks showed a "U" shape of network density over time for all three branches for human and mouse. Differentiation appeared as a continuous process, originating from stem cells, through a brief transition state marked by fewer gene interactions, before stabilizing in a progenitor state. Human and mouse shared hub genes in evolutionary networks. (4) Conclusions: the conservation of network dynamics in the hematopoietic systems of mouse and human was reflected by shared hub genes and network topological changes during differentiation.

Single-cell RNA sequencing coupled to TCR profiling of large granular lymphocyte leukemia T cells.

T-cell large granular lymphocyte leukemia (T-LGLL) is a lymphoproliferative disease and bone marrow failure syndrome which responds to immunosuppressive therapies. We show single-cell TCR coupled with RNA sequencing of CD3+ T cells from 13 patients, sampled before and after alemtuzumab treatments. Effector memory T cells and loss of T cell receptor (TCR) repertoire diversity are prevalent in T-LGLL. Shared TCRA and TCRB clonotypes are absent. Deregulation of cell survival and apoptosis gene programs, and marked downregulation of apoptosis genes in CD8+ clones, are prominent features of T-LGLL cells. Apoptosis genes are upregulated after alemtuzumab treatment, especially in responders than non-responders; baseline expression levels of apoptosis genes are predictive of hematologic response. Alemtuzumab does not attenuate TCR clonality, and TCR diversity is further skewed after treatment. Inferences made from analysis of single cell data inform understanding of the pathophysiologic mechanisms of clonal expansion and persistence in T-LGLL.

Single-cell profiling of T lymphocytes in deficiency of adenosine deaminase 2.

Deficiency of adenosine deaminase 2 (DADA2) is a monogenic vasculitis syndrome caused by autosomal-recessive loss-of-function mutations in the ADA2 gene (previously known as CECR1). Vasculitis, vasculopathy, and inflammation are dominant clinical features of this disease; the spectrum of manifestations includes immunodeficiency and lymphoproliferation as well as hematologic manifestations. ADA2 is primarily secreted by stimulated monocytes and macrophages. Aberrant monocyte differentiation to macrophages and neutrophils are important in the pathogenesis of DADA2, but little is known about T lymphocytes in this disease. We performed combined single-cell RNA sequencing and single-cell TCR sequencing in order to profile T cell repertoires in 10 patients with DADA2. Although there were no significant alterations of T cell subsets, we observed activation of both CD8+ and CD4+ T cells. There was no clonal expansion of T cells: most TCRs were expressed at basal levels in patients and healthy donors. TCR usage was private to individual patients and not disease specific, indicating as unlikely a common pathogenic background or predisposition to a common pathogen. We recognized activation of IFN pathways as a signature of T cells and STAT1 as a hub gene in the gene network of T cell activation and cytotoxicity. Overall, T cells in DADA2 patients showed distinct cell-cell interactions with monocytes, as compared with healthy donors, and many of these ligand-receptor interactions likely drove up-regulation of STAT1 in both T cells and other immune cells in patients. Our analysis reveals previously undercharacterized cell characteristics in DADA2.

Quantitative utilization of prior biological knowledge in the Bayesian network modeling of gene expression data.

BACKGROUND: Bayesian Network (BN) is a powerful approach to reconstructing genetic regulatory networks from gene expression data. However, expression data by itself suffers from high noise and lack of power. Incorporating prior biological knowledge can improve the performance. As each type of prior knowledge on its own may be incomplete or limited by quality issues, integrating multiple sources of prior knowledge to utilize their consensus is desirable. RESULTS: We introduce a new method to incorporate the quantitative information from multiple sources of prior knowledge. It first uses the Naïve Bayesian classifier to assess the likelihood of functional linkage between gene pairs based on prior knowledge. In this study we included cocitation in PubMed and schematic similarity in Gene Ontology annotation. A candidate network edge reservoir is then created in which the copy number of each edge is proportional to the estimated likelihood of linkage between the two corresponding genes. In network simulation the Markov Chain Monte Carlo sampling algorithm is adopted, and samples from this reservoir at each iteration to generate new candidate networks. We evaluated the new algorithm using both simulated and real gene expression data including that from a yeast cell cycle and a mouse pancreas development/growth study. Incorporating prior knowledge led to a ~2 fold increase in the number of known transcription regulations recovered, without significant change in false positive rate. In contrast, without the prior knowledge BN modeling is not always better than a random selection, demonstrating the necessity in network modeling to supplement the gene expression data with additional information. CONCLUSION: our new development provides a statistical means to utilize the quantitative information in prior biological knowledge in the BN modeling of gene expression data, which significantly improves the performance.