A mixture-of-experts deep generative model for integrated analysis of single-cell multiomics data

Liam tackles complex multimodal single-cell data integration challenges

Multi-omics characterization of single cells holds outstanding potential for profiling the dynamics and relations of gene regulatory states of thousands of cells. How to integrate multimodal data is an open problem, especially when aiming to combine data from multiple sources or conditions containing both biological and technical variation. We introduce liam, a flexible model for the simultaneous horizontal and vertical integration of paired single-cell multimodal data and mosaic integration of paired with unimodal data. Liam learns a joint low-dimensional representation of the measured modalities, which proves beneficial when the information content or quality of the modalities differ. Its integration accounts for complex batch effects using a tunable combination of conditional and adversarial training, which can be optimized using replicate information while retaining selected biological variation. We demonstrate liam's superior performance on multiple paired multimodal data types, including Multiome and CITE-seq data, and in mosaic integration scenarios. Our detailed benchmarking experiments illustrate the complexities and challenges remaining for integration and the meaningful assessment of its success.

Multi-modal generative modeling for joint analysis of single-cell T cell receptor and gene expression data

Recent advances in single-cell immune profiling have enabled the simultaneous measurement of transcriptome and T cell receptor (TCR) sequences, offering great potential for studying immune responses at the cellular level. However, integrating these diverse modalities across datasets is challenging due to their unique data characteristics and technical variations. Here, to address this, we develop the multimodal generative model mvTCR to fuse modality-specific information across transcriptome and TCR into a shared representation. Our analysis demonstrates the added value of multimodal over unimodal approaches to capture antigen specificity. Notably, we use mvTCR to distinguish T cell subpopulations binding to SARS-CoV-2 antigens from bystander cells. Furthermore, when combined with reference mapping approaches, mvTCR can map newly generated datasets to extensive T cell references, facilitating knowledge transfer. In summary, we envision mvTCR to enable a scalable analysis of multimodal immune profiling data and advance our understanding of immune responses.

Deciphering spatial domains from spatial multi-omics with SpatialGlue

Advances in spatial omics technologies now allow multiple types of data to be acquired from the same tissue slice. To realize the full potential of such data, we need spatially informed methods for data integration. Here, we introduce SpatialGlue, a graph neural network model with a dual-attention mechanism that deciphers spatial domains by intra-omics integration of spatial location and omics measurement followed by cross-omics integration. We demonstrated SpatialGlue on data acquired from different tissue types using different technologies, including spatial epigenome-transcriptome and transcriptome-proteome modalities. Compared to other methods, SpatialGlue captured more anatomical details and more accurately resolved spatial domains such as the cortex layers of the brain. Our method also identified cell types like spleen macrophage subsets located at three different zones that were not available in the original data annotations. SpatialGlue scales well with data size and can be used to integrate three modalities. Our spatial multi-omics analysis tool combines the information from complementary omics modalities to obtain a holistic view of cellular and tissue properties.

Protocol to perform integrative analysis of high-dimensional single-cell multimodal data using an interpretable deep learning technique

The advent of single-cell multi-omics sequencing technology makes it possible for researchers to leverage multiple modalities for individual cells. Here, we present a protocol to perform integrative analysis of high-dimensional single-cell multimodal data using an interpretable deep learning technique called moETM. We describe steps for data preprocessing, multi-omics integration, inclusion of prior pathway knowledge, and cross-omics imputation. As a demonstration, we used the single-cell multi-omics data collected from bone marrow mononuclear cells (GSE194122) as in our original study. For complete details on the use and execution of this protocol, please refer to Zhou et al.1.

Interpretable deep learning in single-cell omics

MOTIVATION

Single-cell omics technologies have enabled the quantification of molecular profiles in individual cells at an unparalleled resolution. Deep learning, a rapidly evolving sub-field of machine learning, has instilled a significant interest in single-cell omics research due to its remarkable success in analysing heterogeneous high-dimensional single-cell omics data. Nevertheless, the inherent multi-layer nonlinear architecture of deep learning models often makes them 'black boxes' as the reasoning behind predictions is often unknown and not transparent to the user. This has stimulated an increasing body of research for addressing the lack of interpretability in deep learning models, especially in single-cell omics data analyses, where the identification and understanding of molecular regulators are crucial for interpreting model predictions and directing downstream experimental validations.

RESULTS

In this work, we introduce the basics of single-cell omics technologies and the concept of interpretable deep learning. This is followed by a review of the recent interpretable deep learning models applied to various single-cell omics research. Lastly, we highlight the current limitations and discuss potential future directions.

Scalable integration of multiomic single-cell data using generative adversarial networks

MOTIVATION

Single-cell profiling has become a common practice to investigate the complexity of tissues, organs, and organisms. Recent technological advances are expanding our capabilities to profile various molecular layers beyond the transcriptome such as, but not limited to, the genome, the epigenome, and the proteome. Depending on the experimental procedure, these data can be obtained from separate assays or the very same cells. Yet, integration of more than two assays is currently not supported by the majority of the computational frameworks avaiable.

RESULTS

We here propose a Multi-Omic data integration framework based on Wasserstein Generative Adversarial Networks suitable for the analysis of paired or unpaired data with a high number of modalities (>2). At the core of our strategy is a single network trained on all modalities together, limiting the computational burden when many molecular layers are evaluated.

AVAILABILITY AND IMPLEMENTATION

Source code of our framework is available at https://github.com/vgiansanti/MOWGAN.

scmFormer Integrates Large-Scale Single-Cell Proteomics and Transcriptomics Data by Multi-Task Transformer

Transformer-based models have revolutionized single cell RNA-seq (scRNA-seq) data analysis. However, their applicability is challenged by the complexity and scale of single-cell multi-omics data. Here a novel single-cell multi-modal/multi-task transformer (scmFormer) is proposed to fill up the existing blank of integrating single-cell proteomics with other omics data. Through systematic benchmarking, it is demonstrated that scmFormer excels in integrating large-scale single-cell multimodal data and heterogeneous multi-batch paired multi-omics data, while preserving shared information across batchs and distinct biological information. scmFormer achieves 54.5% higher average F1 score compared to the second method in transferring cell-type labels from single-cell transcriptomics to proteomics data. Using COVID-19 datasets, it is presented that scmFormer successfully integrates over 1.48 million cells on a personal computer. Moreover, it is also proved that scmFormer performs better than existing methods on generating the unmeasured modality and is well-suited for spatial multi-omic data. Thus, scmFormer is a powerful and comprehensive tool for analyzing single-cell multi-omics data.

DANCE: a deep learning library and benchmark platform for single-cell analysis

DANCE is the first standard, generic, and extensible benchmark platform for accessing and evaluating computational methods across the spectrum of benchmark datasets for numerous single-cell analysis tasks. Currently, DANCE supports 3 modules and 8 popular tasks with 32 state-of-art methods on 21 benchmark datasets. People can easily reproduce the results of supported algorithms across major benchmark datasets via minimal efforts, such as using only one command line. In addition, DANCE provides an ecosystem of deep learning architectures and tools for researchers to facilitate their own model development. DANCE is an open-source Python package that welcomes all kinds of contributions.

InClust+: the deep generative framework with mask modules for multimodal data integration, imputation, and cross-modal generation

BACKGROUND

With the development of single-cell technology, many cell traits can be measured. Furthermore, the multi-omics profiling technology could jointly measure two or more traits in a single cell simultaneously. In order to process the various data accumulated rapidly, computational methods for multimodal data integration are needed.

RESULTS

Here, we present inClust+, a deep generative framework for the multi-omics. It's built on previous inClust that is specific for transcriptome data, and augmented with two mask modules designed for multimodal data processing: an input-mask module in front of the encoder and an output-mask module behind the decoder. InClust+ was first used to integrate scRNA-seq and MERFISH data from similar cell populations, and to impute MERFISH data based on scRNA-seq data. Then, inClust+ was shown to have the capability to integrate the multimodal data (e.g. tri-modal data with gene expression, chromatin accessibility and protein abundance) with batch effect. Finally, inClust+ was used to integrate an unlabeled monomodal scRNA-seq dataset and two labeled multimodal CITE-seq datasets, transfer labels from CITE-seq datasets to scRNA-seq dataset, and generate the missing modality of protein abundance in monomodal scRNA-seq data. In the above examples, the performance of inClust+ is better than or comparable to the most recent tools in the corresponding task.

CONCLUSIONS

The inClust+ is a suitable framework for handling multimodal data. Meanwhile, the successful implementation of mask in inClust+ means that it can be applied to other deep learning methods with similar encoder-decoder architecture to broaden the application scope of these models.

Single-cell spatial metabolomics with cell-type specific protein profiling for tissue systems biology

Metabolic reprogramming in cancer and immune cells occurs to support their increasing energy needs in biological tissues. Here we propose Single Cell Spatially resolved Metabolic (scSpaMet) framework for joint protein-metabolite profiling of single immune and cancer cells in male human tissues by incorporating untargeted spatial metabolomics and targeted multiplexed protein imaging in a single pipeline. We utilized the scSpaMet to profile cell types and spatial metabolomic maps of 19507, 31156, and 8215 single cells in human lung cancer, tonsil, and endometrium tissues, respectively. The scSpaMet analysis revealed cell type-dependent metabolite profiles and local metabolite competition of neighboring single cells in human tissues. Deep learning-based joint embedding revealed unique metabolite states within cell types. Trajectory inference showed metabolic patterns along cell differentiation paths. Here we show scSpaMet's ability to quantify and visualize the cell-type specific and spatially resolved metabolic-protein mapping as an emerging tool for systems-level understanding of tissue biology.

Single-Cell Multimodal Prediction via Transformers

The recent development of multimodal single-cell technology has made the possibility of acquiring multiple omics data from individual cells, thereby enabling a deeper understanding of cellular states and dynamics. Nevertheless, the proliferation of multimodal single-cell data also introduces tremendous challenges in modeling the complex interactions among different modalities. The recently advanced methods focus on constructing static interaction graphs and applying graph neural networks (GNNs) to learn from multimodal data. However, such static graphs can be suboptimal as they do not take advantage of the downstream task information; meanwhile GNNs also have some inherent limitations when deeply stacking GNN layers. To tackle these issues, in this work, we investigate how to leverage transformers for multimodal single-cell data in an end-to-end manner while exploiting downstream task information. In particular, we propose a scMoFormer framework which can readily incorporate external domain knowledge and model the interactions within each modality and cross modalities. Extensive experiments demonstrate that scMoFormer achieves superior performance on various benchmark datasets. Remarkably, scMoFormer won a Kaggle silver medal with the rank of 24/1221 (Top 2%) without ensemble in a NeurIPS 2022 competition. Our implementation is publicly available at Github.

Deep Learning Methods for Omics Data Imputation

One common problem in omics data analysis is missing values, which can arise due to various reasons, such as poor tissue quality and insufficient sample volumes. Instead of discarding missing values and related data, imputation approaches offer an alternative means of handling missing data. However, the imputation of missing omics data is a non-trivial task. Difficulties mainly come from high dimensionality, non-linear or non-monotonic relationships within features, technical variations introduced by sampling methods, sample heterogeneity, and the non-random missingness mechanism. Several advanced imputation methods, including deep learning-based methods, have been proposed to address these challenges. Due to its capability of modeling complex patterns and relationships in large and high-dimensional datasets, many researchers have adopted deep learning models to impute missing omics data. This review provides a comprehensive overview of the currently available deep learning-based methods for omics imputation from the perspective of deep generative model architectures such as autoencoder, variational autoencoder, generative adversarial networks, and Transformer, with an emphasis on multi-omics data imputation. In addition, this review also discusses the opportunities that deep learning brings and the challenges that it might face in this field.

Contrastively generative self-expression model for single-cell and spatial multimodal data

Advances in single-cell multi-omics technology provide an unprecedented opportunity to fully understand cellular heterogeneity. However, integrating omics data from multiple modalities is challenging due to the individual characteristics of each measurement. Here, to solve such a problem, we propose a contrastive and generative deep self-expression model, called single-cell multimodal self-expressive integration (scMSI), which integrates the heterogeneous multimodal data into a unified manifold space. Specifically, scMSI first learns each omics-specific latent representation and self-expression relationship to consider the characteristics of different omics data by deep self-expressive generative model. Then, scMSI combines these omics-specific self-expression relations through contrastive learning. In such a way, scMSI provides a paradigm to integrate multiple omics data even with weak relation, which effectively achieves the representation learning and data integration into a unified framework. We demonstrate that scMSI provides a cohesive solution for a variety of analysis tasks, such as integration analysis, data denoising, batch correction and spatial domain detection. We have applied scMSI on various single-cell and spatial multimodal datasets to validate its high effectiveness and robustness in diverse data types and application scenarios.

Single-Cell Multiomics

Single-cell RNA sequencing methods have led to improved understanding of the heterogeneity and transcriptomic states present in complex biological systems. Recently, the development of novel single-cell technologies for assaying additional modalities, specifically genomic, epigenomic, proteomic, and spatial data, allows for unprecedented insight into cellular biology. While certain technologies collect multiple measurements from the same cells simultaneously, even when modalities are separately assayed in different cells, we can apply novel computational methods to integrate these data. The application of computational integration methods to multimodal paired and unpaired data results in rich information about the identities of the cells present and the interactions between different levels of biology, such as between genetic variation and transcription. In this review, we both discuss the single-cell technologies for measuring these modalities and describe and characterize a variety of computational integration methods for combining the resulting data to leverage multimodal information toward greater biological insight.

Omics data integration in computational biology viewed through the prism of machine learning paradigms

Important quantities of biological data can today be acquired to characterize cell types and states, from various sources and using a wide diversity of methods, providing scientists with more and more information to answer challenging biological questions. Unfortunately, working with this amount of data comes at the price of ever-increasing data complexity. This is caused by the multiplication of data types and batch effects, which hinders the joint usage of all available data within common analyses. Data integration describes a set of tasks geared towards embedding several datasets of different origins or modalities into a joint representation that can then be used to carry out downstream analyses. In the last decade, dozens of methods have been proposed to tackle the different facets of the data integration problem, relying on various paradigms. This review introduces the most common data types encountered in computational biology and provides systematic definitions of the data integration problems. We then present how machine learning innovations were leveraged to build effective data integration algorithms, that are widely used today by computational biologists. We discuss the current state of data integration and important pitfalls to consider when working with data integration tools. We eventually detail a set of challenges the field will have to overcome in the coming years.

MultiVI: deep generative model for the integration of multimodal data

Jointly profiling the transcriptome, chromatin accessibility and other molecular properties of single cells offers a powerful way to study cellular diversity. Here we present MultiVI, a probabilistic model to analyze such multiomic data and leverage it to enhance single-modality datasets. MultiVI creates a joint representation that allows an analysis of all modalities included in the multiomic input data, even for cells for which one or more modalities are missing. It is available at scvi-tools.org .

Statistical and machine learning methods for immunoprofiling based on single-cell data

Immunoprofiling has become a crucial tool for understanding the complex interactions between the immune system and diseases or interventions, such as therapies and vaccinations. Immune response biomarkers are critical for understanding those relationships and potentially developing personalized intervention strategies. Single-cell data have emerged as a promising source for identifying immune response biomarkers. In this review, we discuss the current state-of-the-art methods for immunoprofiling, including those for reducing the dimensionality of high-dimensional single-cell data and methods for clustering, classification, and prediction. We also draw attention to recent developments in data integration.

scJVAE: A novel method for integrative analysis of multimodal single-cell data

The study of cellular decision-making can be approached comprehensively using multimodal single-cell omics technology. Recent advances in multimodal single-cell technology have enabled simultaneous profiling of more than one modality from the same cell, providing more significant insights into cell characteristics. However, learning the joint representation of multimodal single-cell data is challenging due to batch effects. Here we present a novel method, scJVAE (single-cell Joint Variational AutoEncoder), for batch effect removal and joint representation of multimodal single-cell data. The scJVAE integrates and learns joint embedding of paired scRNA-seq and scATAC-seq data modalities. We evaluate and demonstrate the ability of scJVAE to remove batch effects using various datasets with paired gene expression and open chromatin. We also consider scJVAE for downstream analysis, such as lower dimensional representation, cell-type clustering, and time and memory requirement. We find scJVAE a robust and scalable method outperforming existing state-of-the-art batch effect removal and integration methods.

Efficient Generation of Paired Single-Cell Multiomics Profiles by Deep Learning

Recent advances in single-cell sequencing technology have made it possible to measure multiple paired omics simultaneously in a single cell such as cellular indexing of transcriptomes and epitopes by sequencing (CITE-seq) and single-nucleus chromatin accessibility and mRNA expression sequencing (SNARE-seq). However, the widespread application of these single-cell multiomics profiling technologies has been limited by their experimental complexity, noise in nature, and high cost. In addition, single-omics sequencing technologies have generated tremendous and high-quality single-cell datasets but have yet to be fully utilized. Here, single-cell multiomics generation (scMOG), a deep learning-based framework to generate single-cell assay for transposase-accessible chromatin (ATAC) data in silico is developed from experimentally available single-cell RNA-seq measurements and vice versa. The results demonstrate that scMOG can accurately perform cross-omics generation between RNA and ATAC, and generate paired multiomics data with biological meanings when one omics is experimentally unavailable and out of training datasets. The generated ATAC, either alone or in combination with measured RNA, exhibits equivalent or superior performance to that of the experimentally measured counterparts throughout multiple downstream analyses. scMOG is also applied to human lymphoma data, which proves to be more effective in identifying tumor samples than the experimentally measured ATAC data. Finally, the performance of scMOG is investigated in other omics such as proteomics and it still shows robust performance on surface protein generation.

Tumor cell plasticity in targeted therapy-induced resistance: mechanisms and new strategies

Despite the success of targeted therapies in cancer treatment, therapy-induced resistance remains a major obstacle to a complete cure. Tumor cells evade treatments and relapse via phenotypic switching driven by intrinsic or induced cell plasticity. Several reversible mechanisms have been proposed to circumvent tumor cell plasticity, including epigenetic modifications, regulation of transcription factors, activation or suppression of key signaling pathways, as well as modification of the tumor environment. Epithelial-to-mesenchymal transition, tumor cell and cancer stem cell formation also serve as roads towards tumor cell plasticity. Corresponding treatment strategies have recently been developed that either target plasticity-related mechanisms or employ combination treatments. In this review, we delineate the formation of tumor cell plasticity and its manipulation of tumor evasion from targeted therapy. We discuss the non-genetic mechanisms of targeted drug-induced tumor cell plasticity in various types of tumors and provide insights into the contribution of tumor cell plasticity to acquired drug resistance. New therapeutic strategies such as inhibition or reversal of tumor cell plasticity are also presented. We also discuss the multitude of clinical trials that are ongoing worldwide with the intention of improving clinical outcomes. These advances provide a direction for developing novel therapeutic strategies and combination therapy regimens that target tumor cell plasticity.

A data-fusion approach to identifying developmental dyslexia from multi-omics datasets

This exploratory study tested and validated the use of data fusion and machine learning techniques to probe high-throughput omics and clinical data with a goal of exploring the etiology of developmental dyslexia. Developmental dyslexia is the leading learning disability in school aged children affecting roughly 5-10% of the US population. The complex biological and neurological phenotype of this life altering disability complicates its diagnosis. Phenome, exome, and metabolome data was collected allowing us to fully explore this system from a behavioral, cellular, and molecular point of view. This study provides a proof of concept showing that data fusion and ensemble learning techniques can outperform traditional machine learning techniques when provided small and complex multi-omics and clinical datasets. Heterogenous stacking classifiers consisting of single-omic experts/models achieved an accuracy of 86%, F1 score of 0.89, and AUC value of 0.83. Ensemble methods also provided a ranked list of important features that suggests exome single nucleotide polymorphisms found in the thalamus and cerebellum could be potential biomarkers for developmental dyslexia and heavily influenced the classification of DD within our machine learning models.

Multimodal sensor fusion in the latent representation space

A new method for multimodal sensor fusion is introduced. The technique relies on a two-stage process. In the first stage, a multimodal generative model is constructed from unlabelled training data. In the second stage, the generative model serves as a reconstruction prior and the search manifold for the sensor fusion tasks. The method also handles cases where observations are accessed only via subsampling i.e. compressed sensing. We demonstrate the effectiveness and excellent performance on a range of multimodal fusion experiments such as multisensory classification, denoising, and recovery from subsampled observations.

Single-cell pair-wise relationships untangled by composite embedding model

In multicellular organisms, cell identity and functions are primed and refined through interactions with other surrounding cells. Here, we propose a scalable machine learning method, termed SPRUCE, which is designed to systematically ascertain common cell-cell communication patterns embedded in single-cell RNA-seq data. We applied our approach to investigate tumor microenvironments consolidating multiple breast cancer datasets and found seven frequently observed interaction signatures and underlying gene-gene interaction networks. Our results implicate that a part of tumor heterogeneity, especially within the same subtype, is better understood by differential interaction patterns rather than the static expression of known marker genes.

A multi-view latent variable model reveals cellular heterogeneity in complex tissues for paired multimodal single-cell data

MOTIVATION

Single-cell multimodal assays allow us to simultaneously measure two different molecular features of the same cell, enabling new insights into cellular heterogeneity, cell development and diseases. However, most existing methods suffer from inaccurate dimensionality reduction for the joint-modality data, hindering their discovery of novel or rare cell subpopulations.

RESULTS

Here, we present VIMCCA, a computational framework based on variational-assisted multi-view canonical correlation analysis to integrate paired multimodal single-cell data. Our statistical model uses a common latent variable to interpret the common source of variances in two different data modalities. Our approach jointly learns an inference model and two modality-specific non-linear models by leveraging variational inference and deep learning. We perform VIMCCA and compare it with 10 existing state-of-the-art algorithms on four paired multi-modal datasets sequenced by different protocols. Results demonstrate that VIMCCA facilitates integrating various types of joint-modality data, thus leading to more reliable and accurate downstream analysis. VIMCCA improves our ability to identify novel or rare cell subtypes compared to existing widely used methods. Besides, it can also facilitate inferring cell lineage based on joint-modality profiles.

AVAILABILITY AND IMPLEMENTATION

The VIMCCA algorithm has been implemented in our toolkit package scbean (≥0.5.0), and its code has been archived at https://github.com/jhu99/scbean under MIT license.

SUPPLEMENTARY INFORMATION

Supplementary data are available at Bioinformatics online.

Clustering of single-cell multi-omics data with a multimodal deep learning method

Single-cell multimodal sequencing technologies are developed to simultaneously profile different modalities of data in the same cell. It provides a unique opportunity to jointly analyze multimodal data at the single-cell level for the identification of distinct cell types. A correct clustering result is essential for the downstream complex biological functional studies. However, combining different data sources for clustering analysis of single-cell multimodal data remains a statistical and computational challenge. Here, we develop a novel multimodal deep learning method, scMDC, for single-cell multi-omics data clustering analysis. scMDC is an end-to-end deep model that explicitly characterizes different data sources and jointly learns latent features of deep embedding for clustering analysis. Extensive simulation and real-data experiments reveal that scMDC outperforms existing single-cell single-modal and multimodal clustering methods on different single-cell multimodal datasets. The linear scalability of running time makes scMDC a promising method for analyzing large multimodal datasets.

Clustering single-cell multi-omics data with MoClust

MOTIVATION

Single-cell multi-omics sequencing techniques have rapidly developed in the past few years. Clustering analysis with single-cell multi-omics data may give us novel perspectives to dissect cellular heterogeneity. However, multi-omics data have the properties of inherited large dimension, high sparsity and existence of doublets. Moreover, representations of different omics from even the same cell follow diverse distributions. Without proper distribution alignment techniques, clustering methods will encounter less separable clusters easily affected by less informative omics data.

RESULTS

We developed MoClust, a novel joint clustering framework that can be applied to several types of single-cell multi-omics data. A selective automatic doublet detection module that can identify and filter out doublets is introduced in the pretraining stage to improve data quality. Omics-specific autoencoders are introduced to characterize the multi-omics data. A contrastive learning way of distribution alignment is adopted to adaptively fuse omics representations into an omics-invariant representation. This novel way of alignment boosts the compactness and separableness of clusters, while accurately weighting the contribution of each omics to the clustering object. Extensive experiments, over both simulated and real multi-omics datasets, demonstrated the powerful alignment, doublet detection and clustering ability features of MoClust.

AVAILABILITY AND IMPLEMENTATION

An implementation of MoClust is available from https://doi.org/10.5281/zenodo.7306504.

SUPPLEMENTARY INFORMATION

Supplementary data are available at Bioinformatics online.

Multimodal Single-Cell Translation and Alignment with Semi-Supervised Learning

Single-cell multi-omics technologies enable comprehensive interrogation of cellular regulation, yet most single-cell assays measure only one type of activity-such as transcription, chromatin accessibility, DNA methylation, or 3D chromatin architecture-for each cell. To enable a multimodal view for individual cells, we propose Polarbear, a semi-supervised machine learning framework that facilitates missing modality profile prediction and single-cell cross-modality alignment. Polarbear learns to translate between modalities by using data from co-assay measurements coupled with the large quantity of single-assay data available in public databases. This semi-supervised scheme mitigates issues related to low cell quantities and high sparsity in co-assay data. Polarbear first pre-trains a beta-variational autoencoder for each modality using both co-assay and single-assay profiles to learn robust representations of individual cells, and it then uses the co-assay labels to train a translator between these cell representations. This semi-supervised framework enables us to predict missing modality profiles and match single cells across modalities with improved accuracy compared with fully supervised methods, thus facilitating multimodal data integration.

The performance of deep generative models for learning joint embeddings of single-cell multi-omics data

Recent extensions of single-cell studies to multiple data modalities raise new questions regarding experimental design. For example, the challenge of sparsity in single-omics data might be partly resolved by compensating for missing information across modalities. In particular, deep learning approaches, such as deep generative models (DGMs), can potentially uncover complex patterns via a joint embedding. Yet, this also raises the question of sample size requirements for identifying such patterns from single-cell multi-omics data. Here, we empirically examine the quality of DGM-based integrations for varying sample sizes. We first review the existing literature and give a short overview of deep learning methods for multi-omics integration. Next, we consider eight popular tools in more detail and examine their robustness to different cell numbers, covering two of the most common multi-omics types currently favored. Specifically, we use data featuring simultaneous gene expression measurements at the RNA level and protein abundance measurements for cell surface proteins (CITE-seq), as well as data where chromatin accessibility and RNA expression are measured in thousands of cells (10x Multiome). We examine the ability of the methods to learn joint embeddings based on biological and technical metrics. Finally, we provide recommendations for the design of multi-omics experiments and discuss potential future developments.

Machine learning and deep learning applications in microbiome research

The many microbial communities around us form interactive and dynamic ecosystems called microbiomes. Though concealed from the naked eye, microbiomes govern and influence macroscopic systems including human health, plant resilience, and biogeochemical cycling. Such feats have attracted interest from the scientific community, which has recently turned to machine learning and deep learning methods to interrogate the microbiome and elucidate the relationships between its composition and function. Here, we provide an overview of how the latest microbiome studies harness the inductive prowess of artificial intelligence methods. We start by highlighting that microbiome data - being compositional, sparse, and high-dimensional - necessitates special treatment. We then introduce traditional and novel methods and discuss their strengths and applications. Finally, we discuss the outlook of machine and deep learning pipelines, focusing on bottlenecks and considerations to address them.

MIRA: joint regulatory modeling of multimodal expression and chromatin accessibility in single cells

Rigorously comparing gene expression and chromatin accessibility in the same single cells could illuminate the logic of how coupling or decoupling of these mechanisms regulates fate commitment. Here we present MIRA, probabilistic multimodal models for integrated regulatory analysis, a comprehensive methodology that systematically contrasts transcription and accessibility to infer the regulatory circuitry driving cells along cell state trajectories. MIRA leverages topic modeling of cell states and regulatory potential modeling of individual gene loci. MIRA thereby represents cell states in an efficient and interpretable latent space, infers high-fidelity cell state trees, determines key regulators of fate decisions at branch points and exposes the variable influence of local accessibility on transcription at distinct loci. Applied to epidermal differentiation and embryonic brain development from two different multimodal platforms, MIRA revealed that early developmental genes were tightly regulated by local chromatin landscape whereas terminal fate genes were titrated without requiring extensive chromatin remodeling.

Clustering CITE-seq data with a canonical correlation-based deep learning method

Single-cell multiomics sequencing techniques have rapidly developed in the past few years. Among these techniques, single-cell cellular indexing of transcriptomes and epitopes (CITE-seq) allows simultaneous quantification of gene expression and surface proteins. Clustering CITE-seq data have the great potential of providing us with a more comprehensive and in-depth view of cell states and interactions. However, CITE-seq data inherit the properties of scRNA-seq data, being noisy, large-dimensional, and highly sparse. Moreover, representations of RNA and surface protein are sometimes with low correlation and contribute divergently to the clustering object. To overcome these obstacles and find a combined representation well suited for clustering, we proposed scCTClust for multiomics data, especially CITE-seq data, and clustering analysis. Two omics-specific neural networks are introduced to extract cluster information from omics data. A deep canonical correlation method is adopted to find the maximumly correlated representations of two omics. A novel decentralized clustering method is utilized over the linear combination of latent representations of two omics. The fusion weights which can account for contributions of omics to clustering are adaptively updated during training. Extensive experiments over both simulated and real CITE-seq data sets demonstrated the power of scCTClust. We also applied scCTClust on transcriptome–epigenome data to illustrate its potential for generalizing.

Variational autoencoding of gene landscapes during mouse CNS development uncovers layered roles of Polycomb Repressor Complex 2

A prominent aspect of most, if not all, central nervous systems (CNSs) is that anterior regions (brain) are larger than posterior ones (spinal cord). Studies in Drosophila and mouse have revealed that Polycomb Repressor Complex 2 (PRC2), a protein complex responsible for applying key repressive histone modifications, acts by several mechanisms to promote anterior CNS expansion. However, it is unclear what the full spectrum of PRC2 action is during embryonic CNS development and how PRC2 intersects with the epigenetic landscape. We removed PRC2 function from the developing mouse CNS, by mutating the key gene Eed, and generated spatio-temporal transcriptomic data. To decode the role of PRC2, we developed a method that incorporates standard statistical analyses with probabilistic deep learning to integrate the transcriptomic response to PRC2 inactivation with epigenetic data. This multi-variate analysis corroborates the central involvement of PRC2 in anterior CNS expansion, and also identifies several unanticipated cohorts of genes, such as proliferation and immune response genes. Furthermore, the analysis reveals specific profiles of regulation via PRC2 upon these gene cohorts. These findings uncover a differential logic for the role of PRC2 upon functionally distinct gene cohorts that drive CNS anterior expansion. To support the analysis of emerging multi-modal datasets, we provide a novel bioinformatics package that integrates transcriptomic and epigenetic datasets to identify regulatory underpinnings of heterogeneous biological processes.

Cobolt: integrative analysis of multimodal single-cell sequencing data

A growing number of single-cell sequencing platforms enable joint profiling of multiple omics from the same cells. We present Cobolt, a novel method that not only allows for analyzing the data from joint-modality platforms, but provides a coherent framework for the integration of multiple datasets measured on different modalities. We demonstrate its performance on multi-modality data of gene expression and chromatin accessibility and illustrate the integration abilities of Cobolt by jointly analyzing this multi-modality data with single-cell RNA-seq and ATAC-seq datasets.