DeepKINET: a deep generative model for estimating single-cell RNA splicing and degradation rates.

Messenger RNA splicing and degradation are critical for gene expression regulation, the abnormality of which leads to diseases. Previous methods for estimating kinetic rates have limitations, assuming uniform rates across cells. DeepKINET is a deep generative model that estimates splicing and degradation rates at single-cell resolution from scRNA-seq data. DeepKINET outperforms existing methods on simulated and metabolic labeling datasets. Applied to forebrain and breast cancer data, it identifies RNA-binding proteins responsible for kinetic rate diversity. DeepKINET also analyzes the effects of splicing factor mutations on target genes in erythroid lineage cells. DeepKINET effectively reveals cellular heterogeneity in post-transcriptional regulation.

LineageVAE: Reconstructing Historical Cell States and Transcriptomes toward Unobserved Progenitors.

MOTIVATION: Single-cell RNA sequencing (scRNA-seq) enables comprehensive characterization of the cell state. However, its destructive nature prohibits measuring gene expression changes during dynamic processes such as embryogenesis. Although recent studies integrating scRNA-seq with lineage tracing have provided clonal insights between progenitor and mature cells, challenges remain. Because of their experimental nature, observations are sparse, and cells observed in the early state are not the exact progenitors of cells observed at later time points. To overcome these limitations, we developed LineageVAE, a novel computational methodology that utilizes deep learning based on the property that cells sharing barcodes have identical progenitors. RESULTS: LineageVAE is a deep generative model that transforms scRNA-seq observations with identical lineage barcodes into sequential trajectories toward a common progenitor in a latent cell state space. This method enables the reconstruction of unobservable cell state transitions, historical transcriptomes, and regulatory dynamics at a single-cell resolution. Applied to hematopoiesis and reprogrammed fibroblast datasets, LineageVAE demonstrated its ability to restore backward cell state transitions and infer progenitor heterogeneity and transcription factor activity along differentiation trajectories. AVAILABILITY AND IMPLEMENTATION: The LineageVAE model was implemented in Python using the PyTorch deep learning library. The code is available on GitHub at https://github.com/LzrRacer/LineageVAE/. SUPPLEMENTARY INFORMATION: Available at Bioinformatics online.

Spatial and single-cell colocalisation analysis reveals MDK-mediated immunosuppressive environment with regulatory T cells in colorectal carcinogenesis.

BACKGROUND: Cell-cell interaction factors that facilitate the progression of adenoma to sporadic colorectal cancer (CRC) remain unclear, thereby hindering patient survival. METHODS: We performed spatial transcriptomics on five early CRC cases, which included adenoma and carcinoma, and one advanced CRC. To elucidate cell-cell interactions within the tumour microenvironment (TME), we investigated the colocalisation network at single-cell resolution using a deep generative model for colocalisation analysis, combined with a single-cell transcriptome, and assessed the clinical significance in CRC patients. FINDINGS: CRC cells colocalised with regulatory T cells (Tregs) at the adenoma-carcinoma interface. At early-stage carcinogenesis, cell-cell interaction inference between colocalised adenoma and cancer epithelial cells and Tregs based on the spatial distribution of single cells highlighted midkine (MDK) as a prominent signalling molecule sent from tumour epithelial cells to Tregs. Interaction between MDK-high CRC cells and SPP1+ macrophages and stromal cells proved to be the mechanism underlying immunosuppression in the TME. Additionally, we identified syndecan4 (SDC4) as a receptor for MDK associated with Treg colocalisation. Finally, clinical analysis using CRC datasets indicated that increased MDK/SDC4 levels correlated with poor overall survival in CRC patients. INTERPRETATION: MDK is involved in the immune tolerance shown by Tregs to tumour growth. MDK-mediated formation of the TME could be a potential target for early diagnosis and treatment of CRC. FUNDING: Japan Society for the Promotion of Science (JSPS) Grant-in-Aid for Science Research; OITA Cancer Research Foundation; AMED under Grant Number; Japan Science and Technology Agency (JST); Takeda Science Foundation; The Princess Takamatsu Cancer Research Fund.

Single-cell colocalization analysis using a deep generative model.

Analyzing colocalization of single cells with heterogeneous molecular phenotypes is essential for understanding cell-cell interactions, and cellular responses to external stimuli and their biological functions in diseases and tissues. However, existing computational methodologies identified the colocalization patterns between predefined cell populations, which can obscure the molecular signatures arising from intercellular communication. Here, we introduce DeepCOLOR, a computational framework based on a deep generative model that recovers intercellular colocalization networks with single-cell resolution by the integration of single-cell and spatial transcriptomes. Along with colocalized population detection accuracy that is superior to existing methods in simulated dataset, DeepCOLOR identified plausible cell-cell interaction candidates between colocalized single cells and segregated cell populations defined by the colocalization relationships in mouse brain tissues, human squamous cell carcinoma samples, and human lung tissues infected with SARS-CoV-2. DeepCOLOR is applicable to studying cell-cell interactions behind various spatial niches. A record of this paper's transparent peer review process is included in the supplemental information.

Metastases and treatment-resistant lineages in patient-derived cancer cells of colorectal cancer.

Circulating tumor cells (CTCs) play an important role in metastasis and recurrence. However, which cells comprise the complex tumor lineages in recurrence and are key in metastasis are unknown in colorectal cancer (CRC). CRC with high expression of POU5F1 has a poor prognosis with a high incidence of liver metastatic recurrence. We aim to reveal the key cells promoting metastasis and identify treatment-resistant lineages with established EGFP-expressing organoids in two-dimensional culture (2DOs) under the POU5F1 promotor. POU5F1-expressing cells are highly present in relapsed clinical patients' blood as CTCs. Sorted POU5F1-expressing cells from 2DOs have cancer stem cell abilities and abundantly form liver metastases in vivo. Single-cell RNA sequencing of 2DOs identifies heterogeneous populations derived from POU5F1-expressing cells and the Wnt signaling pathway is enriched in POU5F1-expressing cells. Characteristic high expression of CTLA4 is observed in POU5F1-expressing cells and immunocytochemistry confirms the co-expression of POU5F1 and CTLA4. Demethylation in some CpG islands at the transcriptional start sites of POU5F1 and CTLA4 is observed. The Wnt/ß-catenin pathway inhibitor, XAV939, prevents the adhesion and survival of POU5F1-expressing cells in vitro. Early administration of XAV939 also completely inhibits liver metastasis induced by POU5F1-positive cells.

FXYD3 functionally demarcates an ancestral breast cancer stem cell subpopulation with features of drug-tolerant persisters.

The heterogeneity of cancer stem cells (CSCs) within tumors presents a challenge in therapeutic targeting. To decipher the cellular plasticity that fuels phenotypic heterogeneity, we undertook single-cell transcriptomics analysis in triple-negative breast cancer (TNBC) to identify subpopulations in CSCs. We found a subpopulation of CSCs with ancestral features that is marked by FXYD domain-containing ion transport regulator 3 (FXYD3), a component of the Na+/K+ pump. Accordingly, FXYD3+ CSCs evolve and proliferate, while displaying traits of alveolar progenitors that are normally induced during pregnancy. Clinically, FXYD3+ CSCs were persistent during neoadjuvant chemotherapy, hence linking them to drug-tolerant persisters (DTPs) and identifying them as crucial therapeutic targets. Importantly, FXYD3+ CSCs were sensitive to senolytic Na+/K+ pump inhibitors, such as cardiac glycosides. Together, our data indicate that FXYD3+ CSCs with ancestral features are drivers of plasticity and chemoresistance in TNBC. Targeting the Na+/K+ pump could be an effective strategy to eliminate CSCs with ancestral and DTP features that could improve TNBC prognosis.

Hypoxia activates SREBP2 through Golgi disassembly in bone marrow-derived monocytes for enhanced tumor growth.

Bone marrow-derived cells (BMDCs) infiltrate hypoxic tumors at a pre-angiogenic state and differentiate into mature macrophages, thereby inducing pro-tumorigenic immunity. A critical factor regulating this differentiation is activation of SREBP2-a well-known transcription factor participating in tumorigenesis progression-through unknown cellular mechanisms. Here, we show that hypoxia-induced Golgi disassembly and Golgi-ER fusion in monocytic myeloid cells result in nuclear translocation and activation of SREBP2 in a SCAP-independent manner. Notably, hypoxia-induced SREBP2 activation was only observed in an immature lineage of bone marrow-derived cells. Single-cell RNA-seq analysis revealed that SREBP2-mediated cholesterol biosynthesis was upregulated in HSCs and monocytes but not in macrophages in the hypoxic bone marrow niche. Moreover, inhibition of cholesterol biosynthesis impaired tumor growth through suppression of pro-tumorigenic immunity and angiogenesis. Thus, our findings indicate that Golgi-ER fusion regulates SREBP2-mediated metabolic alteration in lineage-specific BMDCs under hypoxia for tumor progression.

Acidic extracellular pH drives accumulation of N1-acetylspermidine and recruitment of protumor neutrophils.

An acidic tumor microenvironment plays a critical role in tumor progression. However, understanding of metabolic reprogramming of tumors in response to acidic extracellular pH has remained elusive. Using comprehensive metabolomic analyses, we demonstrated that acidic extracellular pH (pH 6.8) leads to the accumulation of N1-acetylspermidine, a protumor metabolite, through up-regulation of the expression of spermidine/spermine acetyltransferase 1 (SAT1). Inhibition of SAT1 expression suppressed the accumulation of intra- and extracellular N1-acetylspermidine at acidic pH. Conversely, overexpression of SAT1 increased intra- and extracellular N1-acetylspermidine levels, supporting the proposal that SAT1 is responsible for accumulation of N1-acetylspermidine. While inhibition of SAT1 expression only had a minor effect on cancer cell growth in vitro, SAT1 knockdown significantly decreased tumor growth in vivo, supporting a contribution of the SAT1-N1-acetylspermidine axis to protumor immunity. Immune cell profiling revealed that inhibition of SAT1 expression decreased neutrophil recruitment to the tumor, resulting in impaired angiogenesis and tumor growth. We showed that antineutrophil-neutralizing antibodies suppressed growth in control tumors to a similar extent to that seen in SAT1 knockdown tumors in vivo. Further, a SAT1 signature was found to be correlated with poor patient prognosis. Our findings demonstrate that extracellular acidity stimulates recruitment of protumor neutrophils via the SAT1-N1-acetylspermidine axis, which may represent a metabolic target for antitumor immune therapy.

UNMF: a unified nonnegative matrix factorization for multi-dimensional omics data.

Factor analysis, ranging from principal component analysis to nonnegative matrix factorization, represents a foremost approach in analyzing multi-dimensional data to extract valuable patterns, and is increasingly being applied in the context of multi-dimensional omics datasets represented in tensor form. However, traditional analytical methods are heavily dependent on the format and structure of the data itself, and if these change even slightly, the analyst must change their data analysis strategy and techniques and spend a considerable amount of time on data preprocessing. Additionally, many traditional methods cannot be applied as-is in the presence of missing values in the data. We present a new statistical framework, unified nonnegative matrix factorization (UNMF), for finding informative patterns in messy biological data sets. UNMF is designed for tidy data format and structure, making data analysis easier and simplifying the development of data analysis tools. UNMF can handle a wide range of data structures and formats, and works seamlessly with tensor data including missing observations and repeated measurements. The usefulness of UNMF is demonstrated through its application to several multi-dimensional omics data, offering user-friendly and unified features for analysis and integration. Its application holds great potential for the life science community. UNMF is implemented with R and is available from GitHub (https://github.com/abikoushi/moltenNMF).

Topological data analysis of protein structure and inter/intra-molecular interaction changes attributable to amino acid mutations.

The presence of some amino acid mutations in the amino acid sequence that determines a protein's structure can significantly affect that 3D structure and its biological function. However, the effects upon structural and functional changes differ for each displaced amino acid, and it is very difficult to predict these changes in advance. Although computer simulations are very effective at predicting conformational changes, they struggle to determine whether the amino acid mutation of interest induces sufficient conformational changes, unless the researcher is a specialist in molecular structure calculations. Therefore, we created a framework that efficiently utilizes molecular dynamics and persistent homology methods to identify amino acid mutations that induce structural changes. We show that this framework can be used not only to predict conformational changes produced by amino acid mutations but also to extract groups of mutations that significantly alter similar molecular interactions, by capturing the resultant protein-protein interaction changes.

Convergent genomic diversity and novel BCAA metabolism in intrahepatic cholangiocarcinoma.

BACKGROUND: Driver alterations may represent novel candidates for driver gene-guided therapy; however, intrahepatic cholangiocarcinoma (ICC) with multiple genomic aberrations makes them intractable. Therefore, the pathogenesis and metabolic changes of ICC need to be understood to develop new treatment strategies. We aimed to unravel the evolution of ICC and identify ICC-specific metabolic characteristics to investigate the metabolic pathway associated with ICC development using multiregional sampling to encompass the intra- and inter-tumoral heterogeneity. METHODS: We performed the genomic, transcriptomic, proteomic and metabolomic analysis of 39-77 ICC tumour samples and eleven normal samples. Further, we analysed their cell proliferation and viability. RESULTS: We demonstrated that intra-tumoral heterogeneity of ICCs with distinct driver genes per case exhibited neutral evolution, regardless of their tumour stage. Upregulation of BCAT1 and BCAT2 indicated the involvement of 'Val Leu Ile degradation pathway'. ICCs exhibit the accumulation of ubiquitous metabolites, such as branched-chain amino acids including valine, leucine, and isoleucine, to negatively affect cancer prognosis. We revealed that this metabolic pathway was almost ubiquitously altered in all cases with genomic diversity and might play important roles in tumour progression and overall survival. CONCLUSIONS: We propose a novel ICC onco-metabolic pathway that could enable the development of new therapeutic interventions.

Increased flexibility of the SARS-CoV-2 RNA-binding site causes resistance to remdesivir.

Mutations continue to accumulate within the SARS-CoV-2 genome, and the ongoing epidemic has shown no signs of ending. It is critical to predict problematic mutations that may arise in clinical environments and assess their properties in advance to quickly implement countermeasures against future variant infections. In this study, we identified mutations resistant to remdesivir, which is widely administered to SARS-CoV-2-infected patients, and discuss the cause of resistance. First, we simultaneously constructed eight recombinant viruses carrying the mutations detected in in vitro serial passages of SARS-CoV-2 in the presence of remdesivir. We confirmed that all the mutant viruses didn't gain the virus production efficiency without remdesivir treatment. Time course analyses of cellular virus infections showed significantly higher infectious titers and infection rates in mutant viruses than wild type virus under treatment with remdesivir. Next, we developed a mathematical model in consideration of the changing dynamic of cells infected with mutant viruses with distinct propagation properties and defined that mutations detected in in vitro passages canceled the antiviral activities of remdesivir without raising virus production capacity. Finally, molecular dynamics simulations of the NSP12 protein of SARS-CoV-2 revealed that the molecular vibration around the RNA-binding site was increased by the introduction of mutations on NSP12. Taken together, we identified multiple mutations that affected the flexibility of the RNA binding site and decreased the antiviral activity of remdesivir. Our new insights will contribute to developing further antiviral measures against SARS-CoV-2 infection.

Spatial and single-cell transcriptomics decipher the cellular environment containing HLA-G+ cancer cells and SPP1+ macrophages in colorectal cancer.

The cellular interactions in the tumor microenvironment of colorectal cancer (CRC) are poorly understood, hindering patient treatment. In the current study, we investigate whether events occurring at the invasion front are of particular importance for CRC treatment strategies. To this end, we analyze CRC tissues by combining spatial transcriptomics from patients with a public single-cell transcriptomic atlas to determine cell-cell interactions at the invasion front. We show that CRC cells are localized specifically at the invasion front. These cells induce human leukocyte antigen G (HLA-G) to produce secreted phosphoprotein 1 (SPP1)+ macrophages while conferring CRC cells with anti-tumor immunity, as well as proliferative and invasive properties. Taken together, these findings highlight the signaling between CRC cell populations and stromal cell populations at the cellular level.

Smoking enhances the expression of angiotensin-converting enzyme 2 involved in the efficiency of severe acute respiratory syndrome coronavirus 2 infection.

Smoking is one of the risk factors most closely related to the severity of coronavirus disease 2019 (COVID-19). However, the relationship between smoking history and severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) infectivity is unknown. In this study, we evaluated the ACE2 expression level in the lungs of current smokers, ex-smokers, and nonsmokers. The ACE2 expression level of ex-smokers who smoked cigarettes until recently (cessation period shorter than 6 months) was higher than that of nonsmokers and ex-smokers with a long history of nonsmoking (cessation period longer than 6 months). We also showed that the efficiency of SARS-CoV-2 infection was enhanced in a manner dependent on the angiotensin-converting enzyme 2 (ACE2) expression level. Using RNA-seq analysis on the lungs of smokers, we identified that the expression of inflammatory signaling genes was correlated with ACE2 expression. Notably, with increasing duration of smoking cessation among ex-smokers, not only ACE2 expression level but also the expression levels of inflammatory signaling genes decreased. These results indicated that smoking enhances the expression levels of ACE2 and inflammatory signaling genes. Our data suggest that the efficiency of SARS-CoV-2 infection is enhanced by smoking-mediated upregulation of ACE2 expression level.

Bayesian statistical method for detecting structural and topological diversity in polymorphic proteins.

Polymorphisms in immune-related proteins and viral spike proteins are high and complicate host-virus interactions. Therefore, diversity analysis of such protein structures is essential to understand the mechanism of the immune system. However, experimental methods, including X-ray crystallography, nuclear magnetic resonance, and cryo-electron microscopy, have several problems: (i) they are conducted under different conditions from the actual cellular environment, (ii) they are laborious, time-consuming, and expensive, and (iii) they do not provide information on the thermodynamic behaviors. In this paper, we propose a computational method to solve these problems by using MD simulations, persistent homology, and a Bayesian statistical model. We apply our method to eight types of HLA-DR complexes to evaluate the structural diversity. The results show that our method can correctly discriminate the intrinsic structural variations caused by amino acid mutations from the random fluctuations caused by thermal vibrations. In the end, we discuss the applicability of our method in combination with existing deep learning-based methods for protein structure analysis.

Detecting time-evolving phenotypic components of adverse reactions against BNT162b2 SARS-CoV-2 vaccine via non-negative tensor factorization.

Symptoms of adverse reactions to vaccines evolve over time, but traditional studies have focused only on the frequency and intensity of symptoms. Here, we attempt to extract the dynamic changes in vaccine adverse reaction symptoms as a small number of interpretable components by using non-negative tensor factorization. We recruited healthcare workers who received two doses of the BNT162b2 mRNA COVID-19 vaccine at Chiba University Hospital and collected information on adverse reactions using a smartphone/web-based platform. We analyzed the adverse-reaction data after each dose obtained for 1,516 participants who received two doses of vaccine. The non-negative tensor factorization revealed four time-evolving components that represent typical temporal patterns of adverse reactions for both doses. These components were differently associated with background factors and post-vaccine antibody titers. These results demonstrate that complex adverse reactions against vaccines can be explained by a limited number of time-evolving components identified by tensor factorization.

An analysis modality for vascular structures combining tissue-clearing technology and topological data analysis.

The blood and lymphatic vasculature networks are not yet fully understood even in mouse because of the inherent limitations of imaging systems and quantification methods. This study aims to evaluate the usefulness of the tissue-clearing technology for visualizing blood and lymphatic vessels in adult mouse. Clear, unobstructed brain/body imaging cocktails and computational analysis (CUBIC) enables us to capture the high-resolution 3D images of organ- or area-specific vascular structures. To evaluate these 3D structural images, signals are first classified from the original captured images by machine learning at pixel base. Then, these classified target signals are subjected to topological data analysis and non-homogeneous Poisson process model to extract geometric features. Consequently, the structural difference of vasculatures is successfully evaluated in mouse disease models. In conclusion, this study demonstrates the utility of CUBIC for analysis of vascular structures and presents its feasibility as an analysis modality in combination with 3D images and mathematical frameworks.

Metabolic remodeling of pyrimidine synthesis pathway and serine synthesis pathway in human glioblastoma.

Glioblastoma is the most common brain tumor with dismal outcomes in adults. Metabolic remodeling is now widely acknowledged as a hallmark of cancer cells, but glioblastoma-specific metabolic pathways remain unclear. Here we show, using a large-scale targeted proteomics platform and integrated molecular pathway-level analysis tool, that the de novo pyrimidine synthesis pathway and serine synthesis pathway (SSP) are the major enriched pathways in vivo for patients with glioblastoma. Among the enzymes associated with nucleotide synthesis, RRM1 and NME1 are significantly upregulated in glioblastoma. In the SSP, SHMT2 and PSPH are upregulated but the upstream enzyme PSAT1 is downregulated in glioblastoma. Kaplan-Meier curves of overall survival for the GSE16011 and The Cancer Genome Atlas datasets revealed that high SSP activity correlated with poor outcome. Enzymes relating to the pyrimidine synthesis pathway and SSP might offer therapeutic targets for new glioblastoma treatments.

Metabolome and transcriptome analysis on muscle of sporadic inclusion body myositis.

OBJECTIVE: Sporadic inclusion body myositis (sIBM) is the most common acquired myopathy in patients older than 50 years of age. sIBM is hardly responds to any immunosuppressing theraphies, and its pathophysiology remains elusive. This study aims to explore pathogenic pathways underlying sIBM and identify novel therapeutic targets using metabolomic and transcriptomic analyses. METHODS: In this retrospective observational study, we analyzed biopsied muscle samples from 14 sIBM patients and six non-diseased subjects to identify metabolic profiles. Frozen muscle samples were used to measure metabolites with cation and anion modes of capillary electrophoresis time of flight mass spectrometry. We validated the metabolic pathway altered in muscles of sIBM patients through RNA sequencing and histopathological studies. RESULTS: A total of 198 metabolites were identified. Metabolomic and transcriptomic analyses identified specific metabolite changes in sIBM muscle samples. The pathways of histamine biosynthesis and certain glycosaminoglycan biosynthesis were upregulated in sIBM patients, whereas those of carnitine metabolism and creatine metabolism were downregulated. Histopathological examination showed infiltration of mast cells and deposition of chondroitin sulfate in skeletal muscle samples, supporting the results of metabolomic and transcriptomic analyses. INTERPRETATION: We identified alterations of several metabolic pathways in muscle samples of sIBM patients. These results suggest that mast cells, chondroitin sulfate biosynthesis, carnitine, and creatine play roles in sIBM pathophysiology.