In vivo CRISPR screens identify the E3 ligase Cop1 as a modulator of macrophage infiltration and cancer immunotherapy target.

Despite remarkable clinical efficacy of immune checkpoint blockade (ICB) in cancer treatment, ICB benefits for triple-negative breast cancer (TNBC) remain limited. Through pooled in vivo CRISPR knockout (KO) screens in syngeneic TNBC mouse models, we found that deletion of the E3 ubiquitin ligase Cop1 in cancer cells decreases secretion of macrophage-associated chemokines, reduces tumor macrophage infiltration, enhances anti-tumor immunity, and strengthens ICB response. Transcriptomics, epigenomics, and proteomics analyses revealed that Cop1 functions through proteasomal degradation of the C/ebpδ protein. The Cop1 substrate Trib2 functions as a scaffold linking Cop1 and C/ebpδ, which leads to polyubiquitination of C/ebpδ. In addition, deletion of the E3 ubiquitin ligase Cop1 in cancer cells stabilizes C/ebpδ to suppress expression of macrophage chemoattractant genes. Our integrated approach implicates Cop1 as a target for improving cancer immunotherapy efficacy in TNBC by regulating chemokine secretion and macrophage infiltration in the tumor microenvironment.

Transfer learning enables predictions in network biology.

Mapping gene networks requires large amounts of transcriptomic data to learn the connections between genes, which impedes discoveries in settings with limited data, including rare diseases and diseases affecting clinically inaccessible tissues. Recently, transfer learning has revolutionized fields such as natural language understanding1,2 and computer vision3 by leveraging deep learning models pretrained on large-scale general datasets that can then be fine-tuned towards a vast array of downstream tasks with limited task-specific data. Here, we developed a context-aware, attention-based deep learning model, Geneformer, pretrained on a large-scale corpus of about 30 million single-cell transcriptomes to enable context-specific predictions in settings with limited data in network biology. During pretraining, Geneformer gained a fundamental understanding of network dynamics, encoding network hierarchy in the attention weights of the model in a completely self-supervised manner. Fine-tuning towards a diverse panel of downstream tasks relevant to chromatin and network dynamics using limited task-specific data demonstrated that Geneformer consistently boosted predictive accuracy. Applied to disease modelling with limited patient data, Geneformer identified candidate therapeutic targets for cardiomyopathy. Overall, Geneformer represents a pretrained deep learning model from which fine-tuning towards a broad range of downstream applications can be pursued to accelerate discovery of key network regulators and candidate therapeutic targets.

Green means go: Green light promotes hypocotyl elongation via brassinosteroid signaling.

Although many studies have elucidated the mechanisms by which different wavelengths of light (blue, red, far-red, or ultraviolet-B [UV-B]) regulate plant development, whether and how green light regulates plant development remains largely unknown. Previous studies reported that green light participates in regulating growth and development in land plants, but these studies have reported conflicting results, likely due to technical problems. For example, commercial green light-emitting diode light sources emit a little blue or red light. Here, using a pure green light source, we determined that unlike blue, red, far-red, or UV-B light, which inhibits hypocotyl elongation, green light promotes hypocotyl elongation in Arabidopsis thaliana and several other plants during the first 2-3 d after planting. Phytochromes, cryptochromes, and other known photoreceptors do not mediate green-light-promoted hypocotyl elongation, but the brassinosteroid (BR) signaling pathway is involved in this process. Green light promotes the DNA binding activity of BRI1-EMS-SUPPRESSOR 1 (BES1), a master transcription factor of the BR pathway, thus regulating gene transcription to promote hypocotyl elongation. Our results indicate that pure green light promotes elongation via BR signaling and acts as a shade signal to enable plants to adapt their development to a green-light-dominant environment under a canopy.

ABHD6 suppresses colorectal cancer progression via AKT signaling pathway.

Colorectal cancer (CRC) continues to be a prevalent malignancy, posing a significant risk to human health. The involvement of alpha/beta hydrolase domain 6 (ABHD6), a serine hydrolase family member, in CRC development was suggested by our analysis of clinical data. However, the role of ABHD6 in CRC remains unclear. This study seeks to elucidate the clinical relevance, biological function, and potential molecular mechanisms of ABHD6 in CRC. We investigated the role of ABHD6 in clinical settings, conducting proliferation, migration, and cell cycle assays. To determine the influence of ABHD6 expression levels on Oxaliplatin sensitivity, we also performed apoptosis assays. RNA sequencing and KEGG analysis were utilized to uncover the potential molecular mechanisms of ABHD6. Furthermore, we validated its expression levels using Western blot and reactive oxygen species (ROS) detection assays. Our results demonstrated that ABHD6 expression in CRC tissues was notably lower compared to adjacent normal tissues. This low expression correlated with a poorer prognosis for CRC patients. Moreover, ABHD6 overexpression impeded CRC cell proliferation and migration while inducing G0/G1 cell cycle arrest. In vivo experiments revealed that downregulation of ABHD6 resulted in an increase in tumor weight and volume. Mechanistically, ABHD6 overexpression inhibited the activation of the AKT signaling pathway and decreased ROS levels in CRC cells, suggesting the role of ABHD6 in CRC progression via the AKT signaling pathway. Our findings demonstrate that ABHD6 functions as a tumor suppressor, primarily by inhibiting the AKT signaling pathway. This role establishes ABHD6 as a promising prognostic biomarker and a potential therapeutic target for CRC patients.

A community challenge to predict clinical outcomes after immune checkpoint blockade in non-small cell lung cancer.

BACKGROUND: Predictive biomarkers of immune checkpoint inhibitor (ICI) efficacy are currently lacking for non-small cell lung cancer (NSCLC). Here, we describe the results from the Anti-PD-1 Response Prediction DREAM Challenge, a crowdsourced initiative that enabled the assessment of predictive models by using data from two randomized controlled clinical trials (RCTs) of ICIs in first-line metastatic NSCLC. METHODS: Participants developed and trained models using public resources. These were evaluated with data from the CheckMate 026 trial (NCT02041533), according to the model-to-data paradigm to maintain patient confidentiality. The generalizability of the models with the best predictive performance was assessed using data from the CheckMate 227 trial (NCT02477826). Both trials were phase III RCTs with a chemotherapy control arm, which supported the differentiation between predictive and prognostic models. Isolated model containers were evaluated using a bespoke strategy that considered the challenges of handling transcriptome data from clinical trials. RESULTS: A total of 59 teams participated, with 417 models submitted. Multiple predictive models, as opposed to a prognostic model, were generated for predicting overall survival, progression-free survival, and progressive disease status with ICIs. Variables within the models submitted by participants included tumor mutational burden (TMB), programmed death ligand 1 (PD-L1) expression, and gene-expression-based signatures. The best-performing models showed improved predictive power over reference variables, including TMB or PD-L1. CONCLUSIONS: This DREAM Challenge is the first successful attempt to use protected phase III clinical data for a crowdsourced effort towards generating predictive models for ICI clinical outcomes and could serve as a blueprint for similar efforts in other tumor types and disease states, setting a benchmark for future studies aiming to identify biomarkers predictive of ICI efficacy. TRIAL REGISTRATION: CheckMate 026; NCT02041533, registered January 22, 2014. CheckMate 227; NCT02477826, registered June 23, 2015.

TISMO: syngeneic mouse tumor database to model tumor immunity and immunotherapy response.

Syngeneic mouse models are tumors derived from murine cancer cells engrafted on genetically identical mouse strains. They are widely used tools for studying tumor immunity and immunotherapy response in the context of a fully functional murine immune system. Large volumes of syngeneic mouse tumor expression profiles under different immunotherapy treatments have been generated, although a lack of systematic collection and analysis makes data reuse challenging. We present Tumor Immune Syngeneic MOuse (TISMO), a database with an extensive collection of syngeneic mouse model profiles with interactive visualization features. TISMO contains 605 in vitro RNA-seq samples from 49 syngeneic cancer cell lines across 23 cancer types, of which 195 underwent cytokine treatment. TISMO also includes 1518 in vivo RNA-seq samples from 68 syngeneic mouse tumor models across 19 cancer types, of which 832 were from immune checkpoint blockade (ICB) studies. We manually annotated the sample metadata, such as cell line, mouse strain, transplantation site, treatment, and response status, and uniformly processed and quality-controlled the RNA-seq data. Besides data download, TISMO provides interactive web interfaces to investigate whether specific gene expression, pathway enrichment, or immune infiltration level is associated with differential immunotherapy response. TISMO is available at http://tismo.cistrome.org.

Sialylated IgG in epithelial cancers inhibits antitumor function of T cells via Siglec-7.

Although effective, immune checkpoint blockade induces response in only a subset of cancer patients. There is an urgent need to discover new immune checkpoint targets. Recently, it was found that a class of sialic acid-binding immunoglobulin-like lectins (Siglecs) expressed on the surface of T cells in cancer patients inhibit T cell activation through their intracellular immunosuppressive motifs by recognizing sialic acid-carrying glycans, sialoglycans. However, ligands of Siglecs remain elusive. Here, we report sialylated IgG (SIA-IgG), a ligand to Siglec-7, that is highly expressed in epithelial cancer cells. SIA-IgG binds Siglec-7 directly and inhibits TCR signals. Blocking of either SIA-IgG or Siglec-7 elicited potent antitumor immunity in T cells. Our study suggests that blocking of Siglec-7/SIA-IgG offers an opportunity to enhance immune function while simultaneously sensitizing cancer cells to immune attack.

TIMER2.0 for analysis of tumor-infiltrating immune cells.

Tumor progression and the efficacy of immunotherapy are strongly influenced by the composition and abundance of immune cells in the tumor microenvironment. Due to the limitations of direct measurement methods, computational algorithms are often used to infer immune cell composition from bulk tumor transcriptome profiles. These estimated tumor immune infiltrate populations have been associated with genomic and transcriptomic changes in the tumors, providing insight into tumor-immune interactions. However, such investigations on large-scale public data remain challenging. To lower the barriers for the analysis of complex tumor-immune interactions, we significantly improved our previous web platform TIMER. Instead of just using one algorithm, TIMER2.0 (http://timer.cistrome.org/) provides more robust estimation of immune infiltration levels for The Cancer Genome Atlas (TCGA) or user-provided tumor profiles using six state-of-the-art algorithms. TIMER2.0 provides four modules for investigating the associations between immune infiltrates and genetic or clinical features, and four modules for exploring cancer-related associations in the TCGA cohorts. Each module can generate a functional heatmap table, enabling the user to easily identify significant associations in multiple cancer types simultaneously. Overall, the TIMER2.0 web server provides comprehensive analysis and visualization functions of tumor infiltrating immune cells.

Deep learning for cancer type classification and driver gene identification.

BACKGROUND: Genetic information is becoming more readily available and is increasingly being used to predict patient cancer types as well as their subtypes. Most classification methods thus far utilize somatic mutations as independent features for classification and are limited by study power. We aim to develop a novel method to effectively explore the landscape of genetic variants, including germline variants, and small insertions and deletions for cancer type prediction. RESULTS: We proposed DeepCues, a deep learning model that utilizes convolutional neural networks to unbiasedly derive features from raw cancer DNA sequencing data for disease classification and relevant gene discovery. Using raw whole-exome sequencing as features, germline variants and somatic mutations, including insertions and deletions, were interactively amalgamated for feature generation and cancer prediction. We applied DeepCues to a dataset from TCGA to classify seven different types of major cancers and obtained an overall accuracy of 77.6%. We compared DeepCues to conventional methods and demonstrated a significant overall improvement (p < 0.001). Strikingly, using DeepCues, the top 20 breast cancer relevant genes we have identified, had a 40% overlap with the top 20 known breast cancer driver genes. CONCLUSION: Our results support DeepCues as a novel method to improve the representational resolution of DNA sequencings and its power in deriving features from raw sequences for cancer type prediction, as well as discovering new cancer relevant genes.

Progesterone receptor antagonists reverse stem cell expansion and the paracrine effectors of progesterone action in the mouse mammary gland.

BACKGROUND: The ovarian hormones estrogen and progesterone (EP) are implicated in breast cancer causation. A specific consequence of progesterone exposure is the expansion of the mammary stem cell (MSC) and luminal progenitor (LP) compartments. We hypothesized that this effect, and its molecular facilitators, could be abrogated by progesterone receptor (PR) antagonists administered in a mouse model. METHODS: Ovariectomized FVB mice were randomized to 14 days of treatment: sham, EP, EP + telapristone (EP + TPA), EP + mifepristone (EP + MFP). Mice were then sacrificed, mammary glands harvested, and mammary epithelial cell lineages separated by flow cytometry using cell surface markers. RNA from each lineage was sequenced and differential gene expression was analyzed using DESeq. Quantitative PCR was performed to confirm the candidate genes discovered in RNA seq. ANOVA with Tukey post hoc analysis was performed to compare relative expression. Alternative splicing events were examined using the rMATs multivariate analysis tool. RESULTS: Significant increases in the MSC and luminal mature (LM) cell fractions were observed following EP treatment compared to control (p < 0.01 and p < 0.05, respectively), whereas the LP fraction was significantly reduced (p < 0.05). These hormone-induced effects were reversed upon exposure to TPA and MFP (p < 0.01 for both). Gene Ontology analysis of RNA-sequencing data showed EP-induced enrichment of several pathways, with the largest effect on Wnt signaling in MSC, significantly repressed by PR inhibitors. In LP cells, significant induction of Wnt4 and Rankl, and Wnt pathway intermediates Lrp2 and Axin2 (confirmed by qRTPCR) were reversed by TPA and MFP (p < 0.0001). Downstream signaling intermediates of these pathways (Lrp5, Mmp7) showed similar effects. Expression of markers of epithelial-mesenchymal transition (Cdh1, Cdh3) and the induction of EMT regulators (Zeb1, Zeb2, Gli3, Snai1, and Ptch2) were significantly responsive to progesterone. EP treatment was associated with large-scale alternative splicing events, with an enrichment of motifs associated with Srsf, Esrp, and Rbfox families. Exon skipping was observed in Cdh1, Enah, and Brd4. CONCLUSIONS: PR inhibition reverses known tumorigenic pathways in the mammary gland and suppresses a previously unknown effect of progesterone on RNA splicing events. In total, our results strengthen the case for reconsideration of PR inhibitors for breast cancer prevention.

Cancer classification and pathway discovery using non-negative matrix factorization.

OBJECTIVES: Extracting genetic information from a full range of sequencing data is important for understanding disease. We propose a novel method to effectively explore the landscape of genetic mutations and aggregate them to predict cancer type. DESIGN: We applied non-smooth non-negative matrix factorization (nsNMF) and support vector machine (SVM) to utilize the full range of sequencing data, aiming to better aggregate genetic mutations and improve their power to predict disease type. More specifically, we introduce a novel classifier to distinguish cancer types using somatic mutations obtained from whole-exome sequencing data. Mutations were identified from multiple cancers and scored using SIFT, PP2, and CADD, and collapsed at the individual gene level. nsNMF was then applied to reduce dimensionality and obtain coefficient and basis matrices. A feature matrix was derived from the obtained matrices to train a classifier for cancer type classification with the SVM model. RESULTS: We have demonstrated that the classifier was able to distinguish four cancer types with reasonable accuracy. In five-fold cross-validations using mutation counts as features, the average prediction accuracy was 80% (SEM = 0.1%), significantly outperforming baselines and outperforming models using mutation scores as features. CONCLUSION: Using the factor matrices derived from the nsNMF, we identified multiple genes and pathways that are significantly associated with each cancer type. This study presents a generic and complete pipeline to study the associations between somatic mutations and cancers. The proposed method can be adapted to other studies for disease status classification and pathway discovery.

Early prediction of acute kidney injury following ICU admission using a multivariate panel of physiological measurements.

BACKGROUND: The development of acute kidney injury (AKI) during an intensive care unit (ICU) admission is associated with increased morbidity and mortality. METHODS: Our objective was to develop and validate a data driven multivariable clinical predictive model for early detection of AKI among a large cohort of adult critical care patients. We utilized data form the Medical Information Mart for Intensive Care III (MIMIC-III) for all patients who had a creatinine measured for 3 days following ICU admission and excluded patients with pre-existing condition of Chronic Kidney Disease and Acute Kidney Injury on admission. Data extracted included patient age, gender, ethnicity, creatinine, other vital signs and lab values during the first day of ICU admission, whether the patient was mechanically ventilated during the first day of ICU admission, and the hourly rate of urine output during the first day of ICU admission. RESULTS: Utilizing the demographics, the clinical data and the laboratory test measurements from Day 1 of ICU admission, we accurately predicted max serum creatinine level during Day 2 and Day 3 with a root mean square error of 0.224 mg/dL. We demonstrated that using machine learning models (multivariate logistic regression, random forest and artificial neural networks) with demographics and physiologic features can predict AKI onset as defined by the current clinical guideline with a competitive AUC (mean AUC 0.783 by our all-feature, logistic-regression model), while previous models aimed at more specific patient cohorts. CONCLUSIONS: Experimental results suggest that our model has the potential to assist clinicians in identifying patients at greater risk of new onset of AKI in critical care setting. Prospective trials with independent model training and external validation cohorts are needed to further evaluate the clinical utility of this approach and potentially instituting interventions to decrease the likelihood of developing AKI.

Using natural language processing and machine learning to identify breast cancer local recurrence.

BACKGROUND: Identifying local recurrences in breast cancer from patient data sets is important for clinical research and practice. Developing a model using natural language processing and machine learning to identify local recurrences in breast cancer patients can reduce the time-consuming work of a manual chart review. METHODS: We design a novel concept-based filter and a prediction model to detect local recurrences using EHRs. In the training dataset, we manually review a development corpus of 50 progress notes and extract partial sentences that indicate breast cancer local recurrence. We process these partial sentences to obtain a set of Unified Medical Language System (UMLS) concepts using MetaMap, and we call it positive concept set. We apply MetaMap on patients' progress notes and retain only the concepts that fall within the positive concept set. These features combined with the number of pathology reports recorded for each patient are used to train a support vector machine to identify local recurrences. RESULTS: We compared our model with three baseline classifiers using either full MetaMap concepts, filtered MetaMap concepts, or bag of words. Our model achieved the best AUC (0.93 in cross-validation, 0.87 in held-out testing). CONCLUSIONS: Compared to a labor-intensive chart review, our model provides an automated way to identify breast cancer local recurrences. We expect that by minimally adapting the positive concept set, this study has the potential to be replicated at other institutions with a moderately sized training dataset.

Discovering causal interactions using Bayesian network scoring and information gain.

BACKGROUND: The problem of learning causal influences from data has recently attracted much attention. Standard statistical methods can have difficulty learning discrete causes, which interacting to affect a target, because the assumptions in these methods often do not model discrete causal relationships well. An important task then is to learn such interactions from data. Motivated by the problem of learning epistatic interactions from datasets developed in genome-wide association studies (GWAS), researchers conceived new methods for learning discrete interactions. However, many of these methods do not differentiate a model representing a true interaction from a model representing non-interacting causes with strong individual affects. The recent algorithm MBS-IGain addresses this difficulty by using Bayesian network learning and information gain to discover interactions from high-dimensional datasets. However, MBS-IGain requires marginal effects to detect interactions containing more than two causes. If the dataset is not high-dimensional, we can avoid this shortcoming by doing an exhaustive search. RESULTS: We develop Exhaustive-IGain, which is like MBS-IGain but does an exhaustive search. We compare the performance of Exhaustive-IGain to MBS-IGain using low-dimensional simulated datasets based on interactions with marginal effects and ones based on interactions without marginal effects. Their performance is similar on the datasets based on marginal effects. However, Exhaustive-IGain compellingly outperforms MBS-IGain on the datasets based on 3 and 4-cause interactions without marginal effects. We apply Exhaustive-IGain to investigate how clinical variables interact to affect breast cancer survival, and obtain results that agree with judgements of a breast cancer oncologist. CONCLUSIONS: We conclude that the combined use of information gain and Bayesian network scoring enables us to discover higher order interactions with no marginal effects if we perform an exhaustive search. We further conclude that Exhaustive-IGain can be effective when applied to real data.

Tumor aerobic glycolysis confers immune evasion through modulating sensitivity to T cell-mediated bystander killing via TNF-α.

Metabolic reprogramming toward glycolysis is a hallmark of cancer malignancy. The molecular mechanisms by which the tumor glycolysis pathway promotes immune evasion remain to be elucidated. Here, by performing genome-wide CRISPR screens in murine tumor cells co-cultured with cytotoxic T cells (CTLs), we identified that deficiency of two important glycolysis enzymes, Glut1 (glucose transporter 1) and Gpi1 (glucose-6-phosphate isomerase 1), resulted in enhanced killing of tumor cells by CTLs. Mechanistically, Glut1 inactivation causes metabolic rewiring toward oxidative phosphorylation, which generates an excessive amount of reactive oxygen species (ROS). Accumulated ROS potentiate tumor cell death mediated by tumor necrosis factor alpha (TNF-α) in a caspase-8- and Fadd-dependent manner. Genetic and pharmacological inactivation of Glut1 sensitizes tumors to anti-tumor immunity and synergizes with anti-PD-1 therapy through the TNF-α pathway. The mechanistic interplay between tumor-intrinsic glycolysis and TNF-α-induced killing provides new therapeutic strategies to enhance anti-tumor immunity.

MetaTiME integrates single-cell gene expression to characterize the meta-components of the tumor immune microenvironment.

Recent advances in single-cell RNA sequencing have shown heterogeneous cell types and gene expression states in the non-cancerous cells in tumors. The integration of multiple scRNA-seq datasets across tumors can indicate common cell types and states in the tumor microenvironment (TME). We develop a data driven framework, MetaTiME, to overcome the limitations in resolution and consistency that result from manual labelling using known gene markers. Using millions of TME single cells, MetaTiME learns meta-components that encode independent components of gene expression observed across cancer types. The meta-components are biologically interpretable as cell types, cell states, and signaling activities. By projecting onto the MetaTiME space, we provide a tool to annotate cell states and signature continuums for TME scRNA-seq data. Leveraging epigenetics data, MetaTiME reveals critical transcriptional regulators for the cell states. Overall, MetaTiME learns data-driven meta-components that depict cellular states and gene regulators for tumor immunity and cancer immunotherapy.

Tutorial: integrative computational analysis of bulk RNA-sequencing data to characterize tumor immunity using RIMA.

RNA-sequencing (RNA-seq) has become an increasingly cost-effective technique for molecular profiling and immune characterization of tumors. In the past decade, many computational tools have been developed to characterize tumor immunity from gene expression data. However, the analysis of large-scale RNA-seq data requires bioinformatics proficiency, large computational resources and cancer genomics and immunology knowledge. In this tutorial, we provide an overview of computational analysis of bulk RNA-seq data for immune characterization of tumors and introduce commonly used computational tools with relevance to cancer immunology and immunotherapy. These tools have diverse functions such as evaluation of expression signatures, estimation of immune infiltration, inference of the immune repertoire, prediction of immunotherapy response, neoantigen detection and microbiome quantification. We describe the RNA-seq IMmune Analysis (RIMA) pipeline integrating many of these tools to streamline RNA-seq analysis. We also developed a comprehensive and user-friendly guide in the form of a GitBook with text and video demos to assist users in analyzing bulk RNA-seq data for immune characterization at both individual sample and cohort levels by using RIMA.

Discovery of Targets for Immune-Metabolic Antitumor Drugs Identifies Estrogen-Related Receptor Alpha.

Drugs that kill tumors through multiple mechanisms have the potential for broad clinical benefits. Here, we first developed an in silico multiomics approach (BipotentR) to find cancer cell-specific regulators that simultaneously modulate tumor immunity and another oncogenic pathway and then used it to identify 38 candidate immune-metabolic regulators. We show the tumor activities of these regulators stratify patients with melanoma by their response to anti-PD-1 using machine learning and deep neural approaches, which improve the predictive power of current biomarkers. The topmost identified regulator, ESRRA, is activated in immunotherapy-resistant tumors. Its inhibition killed tumors by suppressing energy metabolism and activating two immune mechanisms: (i) cytokine induction, causing proinflammatory macrophage polarization, and (ii) antigen-presentation stimulation, recruiting CD8+ T cells into tumors. We also demonstrate a wide utility of BipotentR by applying it to angiogenesis and growth suppressor evasion pathways. BipotentR (http://bipotentr.dfci.harvard.edu/) provides a resource for evaluating patient response and discovering drug targets that act simultaneously through multiple mechanisms. SIGNIFICANCE: BipotentR presents resources for evaluating patient response and identifying targets for drugs that can kill tumors through multiple mechanisms concurrently. Inhibition of the topmost candidate target killed tumors by suppressing energy metabolism and effects on two immune mechanisms. This article is highlighted in the In This Issue feature, p. 517.

Using an Unsupervised Clustering Model to Detect the Early Spread of SARS-CoV-2 Worldwide.

Deciphering the population structure of SARS-CoV-2 is critical to inform public health management and reduce the risk of future dissemination. With the continuous accruing of SARS-CoV-2 genomes worldwide, discovering an effective way to group these genomes is critical for organizing the landscape of the population structure of the virus. Taking advantage of recently published state-of-the-art machine learning algorithms, we used an unsupervised deep learning clustering algorithm to group a total of 16,873 SARS-CoV-2 genomes. Using single nucleotide polymorphisms as input features, we identified six major subtypes of SARS-CoV-2. The proportions of the clusters across the continents revealed distinct geographical distributions. Comprehensive analysis indicated that both genetic factors and human migration factors shaped the specific geographical distribution of the population structure. This study provides a different approach using clustering methods to study the population structure of a never-seen-before and fast-growing species such as SARS-CoV-2. Moreover, clustering techniques can be used for further studies of local population structures of the proliferating virus.

Statistical and machine learning methods for spatially resolved transcriptomics data analysis.

The recent advancement in spatial transcriptomics technology has enabled multiplexed profiling of cellular transcriptomes and spatial locations. As the capacity and efficiency of the experimental technologies continue to improve, there is an emerging need for the development of analytical approaches. Furthermore, with the continuous evolution of sequencing protocols, the underlying assumptions of current analytical methods need to be re-evaluated and adjusted to harness the increasing data complexity. To motivate and aid future model development, we herein review the recent development of statistical and machine learning methods in spatial transcriptomics, summarize useful resources, and highlight the challenges and opportunities ahead.