Feature selection revisited in the single-cell era

Considerations for building and using integrated single-cell atlases

The rapid adoption of single-cell technologies has created an opportunity to build single-cell 'atlases' integrating diverse datasets across many laboratories. Such atlases can serve as a reference for analyzing and interpreting current and future data. However, it has become apparent that atlasing approaches differ, and the impact of these differences are often unclear. Here we review the current atlasing literature and present considerations for building and using atlases. Importantly, we find that no one-size-fits-all protocol for atlas building exists, but rather we discuss context-specific considerations and workflows, including atlas conceptualization, data collection, curation and integration, atlas evaluation and atlas sharing. We further highlight the benefits of integrated atlases for analyses of new datasets and deriving biological insights beyond what is possible from individual datasets. Our overview of current practices and associated recommendations will improve the quality of atlases to come, facilitating the shift to a unified, reference-based understanding of single-cell biology.

CosGeneGate selects multi-functional and credible biomarkers for single-cell analysis

MOTIVATION

Selecting representative genes or marker genes to distinguish cell types is an important task in single-cell sequencing analysis. Although many methods have been proposed to select marker genes, the genes selected may have redundancy and/or do not show cell-type-specific expression patterns to distinguish cell types.

RESULTS

Here, we present a novel model, named CosGeneGate, to select marker genes for more effective marker selections. CosGeneGate is inspired by combining the advantages of selecting marker genes based on both cell-type classification accuracy and marker gene specific expression patterns. We demonstrate the better performance of the marker genes selected by CosGeneGate for various downstream analyses than the existing methods with both public datasets and newly sequenced datasets. The non-redundant marker genes identified by CosGeneGate for major cell types and tissues in human can be found at the website as follows: https://github.com/VivLon/CosGeneGate/blob/main/marker gene list.xlsx.

scPair: Boosting single cell multimodal analysis by leveraging implicit feature selection and single cell atlases

Multimodal single-cell assays profile multiple sets of features in the same cells and are widely used for identifying and mapping cell states between chromatin and mRNA and linking regulatory elements to target genes. However, the high dimensionality of input features and shallow sequencing depth compared to unimodal assays pose challenges in data analysis. Here we present scPair, a multimodal single-cell data framework that overcomes these challenges by employing an implicit feature selection approach. scPair uses dual encoder-decoder structures trained on paired data to align cell states across modalities and predict features from one modality to another. We demonstrate that scPair outperforms existing methods in accuracy and execution time, and facilitates downstream tasks such as trajectory inference. We further show scPair can augment smaller multimodal datasets with larger unimodal atlases to increase statistical power to identify groups of transcription factors active during different stages of neural differentiation.

Inorganic pyrophosphatase 1: a key player in immune and metabolic reprogramming in ankylosing spondylitis

The relationships among immune cells, metabolites, and AS events were analyzed via Mendelian randomization (MR), and potential immune cells and metabolites were identified as risk factors for AS. Their relationships were subjected to intermediary MR analysis to identify the final immune cells and metabolites. The vertebral bone marrow blood samples from three patients with and without AS were subjected to 10× single-cell sequencing to further elucidate the role of immune cells in AS. The key genes were screened via expression quantitative trait loci (eQTLs) and MR analyses. The metabolic differences between the two groups were compared through single-cell metabolism analysis. Two subgroups of differentiated (CD)8+ memory T cells and naive B cells were obtained from the combined results of intermediary MR analysis and AS single-cell analysis. After the verification of key genes, inorganic pyrophosphatase 1 (PPA1) was identified as the hub gene, as it is differentially expressed in CD8+ memory T cells and can affect the metabolism of T cells in AS by affecting the expression of ferulic acid (FA)4 sulfate, which participates in the cellular immunity in AS.

Inference of single-cell network using mutual information for scRNA-seq data analysis

BACKGROUND

With the advance in single-cell RNA sequencing (scRNA-seq) technology, deriving inherent biological system information from expression profiles at a single-cell resolution has become possible. It has been known that network modeling by estimating the associations between genes could better reveal dynamic changes in biological systems. However, accurately constructing a single-cell network (SCN) to capture the network architecture of each cell and further explore cell-to-cell heterogeneity remains challenging.

RESULTS

We introduce SINUM, a method for constructing the SIngle-cell Network Using Mutual information, which estimates mutual information between any two genes from scRNA-seq data to determine whether they are dependent or independent in a specific cell. Experiments on various scRNA-seq datasets with different cell numbers based on eight performance indexes (e.g., adjusted rand index and F-measure index) validated the accuracy and robustness of SINUM in cell type identification, superior to the state-of-the-art SCN inference method. Additionally, the SINUM SCNs exhibit high overlap with the human interactome and possess the scale-free property.

CONCLUSIONS

SINUM presents a view of biological systems at the network level to detect cell-type marker genes/gene pairs and investigate time-dependent changes in gene associations during embryo development. Codes for SINUM are freely available at https://github.com/SysMednet/SINUM .

scCAD: Cluster decomposition-based anomaly detection for rare cell identification in single-cell expression data

Single-cell RNA sequencing (scRNA-seq) technologies have become essential tools for characterizing cellular landscapes within complex tissues. Large-scale single-cell transcriptomics holds great potential for identifying rare cell types critical to the pathogenesis of diseases and biological processes. Existing methods for identifying rare cell types often rely on one-time clustering using partial or global gene expression. However, these rare cell types may be overlooked during the clustering phase, posing challenges for their accurate identification. In this paper, we propose a Cluster decomposition-based Anomaly Detection method (scCAD), which iteratively decomposes clusters based on the most differential signals in each cluster to effectively separate rare cell types and achieve accurate identification. We benchmark scCAD on 25 real-world scRNA-seq datasets, demonstrating its superior performance compared to 10 state-of-the-art methods. In-depth case studies across diverse datasets, including mouse airway, brain, intestine, human pancreas, immunology data, and clear cell renal cell carcinoma, showcase scCAD's efficiency in identifying rare cell types in complex biological scenarios. Furthermore, scCAD can correct the annotation of rare cell types and identify immune cell subtypes associated with disease, thereby offering valuable insights into disease progression.

The role of artificial intelligence in disease prediction: using ensemble model to predict disease mellitus

The traditional complications of diabetes are well known and continue to pose a considerable burden to millions of people with diabetes mellitus (DM). With the continuous accumulation of medical data and technological advances, artificial intelligence has shown great potential and advantages in the prediction, diagnosis, and treatment of DM. When DM is diagnosed, some subjective factors and diagnostic methods of doctors will have an impact on the diagnostic results, so the use of artificial intelligence for fast and effective early prediction of DM patients can provide decision-making support to doctors and give more accurate treatment services to patients in time, which is of great clinical medical significance and practical significance. In this paper, an adaptive Stacking ensemble model is proposed based on the theory of "error-ambiguity decomposition," which can adaptively select the base classifiers from the pre-selected models. The adaptive Stacking ensemble model proposed in this paper is compared with KNN, SVM, RF, LR, DT, GBDT, XGBoost, LightGBM, CatBoost, MLP and traditional Stacking ensemble models. The results showed that the adaptive Stacking ensemble model achieved the best performance in five evaluation metrics: accuracy, precision, recall, F1 value and AUC value, which were 0.7559, 0.7286, 0.8132, 0.7686 and 0.8436. The model can effectively predict DM patients and provide a reference value for the screening and diagnosis of clinical DM.

Applications of single-cell multi-omics in liver cancer

Primary liver cancer, more specifically hepatocellular carcinoma (HCC), remains a significant global health problem associated with increasing incidence and mortality. Clinical, biological, and molecular heterogeneity are well-known hallmarks of cancer and HCC is considered one of the most heterogeneous tumour types, displaying substantial inter-patient, intertumoural and intratumoural variability. This heterogeneity plays a pivotal role in hepatocarcinogenesis, metastasis, relapse and drug response or resistance. Unimodal single-cell sequencing techniques have already revolutionised our understanding of the different layers of molecular hierarchy in the tumour microenvironment of HCC. By highlighting the cellular heterogeneity and the intricate interactions among cancer, immune and stromal cells before and during treatment, these techniques have contributed to a deeper comprehension of tumour clonality, hematogenous spreading and the mechanisms of action of immune checkpoint inhibitors. However, major questions remain to be elucidated, with the identification of biomarkers predicting response or resistance to immunotherapy-based regimens representing an important unmet clinical need. Although the application of single-cell multi-omics in liver cancer research has been limited thus far, a revolution of individualised care for patients with HCC will only be possible by integrating various unimodal methods into multi-omics methodologies at the single-cell resolution. In this review, we will highlight the different established single-cell sequencing techniques and explore their biological and clinical impact on liver cancer research, while casting a glance at the future role of multi-omics in this dynamic and rapidly evolving field.

Demultiplexing of single-cell RNA-sequencing data using interindividual variation in gene expression

Motivation

Pooled designs for single-cell RNA sequencing, where many cells from distinct samples are processed jointly, offer increased throughput and reduced batch variation. This study describes expression-aware demultiplexing (EAD), a computational method that employs differential co-expression patterns between individuals to demultiplex pooled samples without any extra experimental steps.

Results

We use synthetic sample pools and show that the top interindividual differentially co-expressed genes provide a distinct cluster of cells per individual, significantly enriching the regulation of metabolism. Our application of EAD to samples of six isogenic inbred mice demonstrated that controlling genetic and environmental effects can solve interindividual variations related to metabolic pathways. We utilized 30 samples from both sepsis and healthy individuals in six batches to assess the performance of classification approaches. The results indicate that combining genetic and EAD results can enhance the accuracy of assignments (Min. 0.94, Mean 0.98, Max. 1). The results were enhanced by an average of 1.4% when EAD and barcoding techniques were combined (Min. 1.25%, Median 1.33%, Max. 1.74%). Furthermore, we demonstrate that interindividual differential co-expression analysis within the same cell type can be used to identify cells from the same donor in different activation states. By analysing single-nuclei transcriptome profiles from the brain, we demonstrate that our method can be applied to nonimmune cells.

Availability and implementation

EAD workflow is available at https://isarnassiri.github.io/scDIV/ as an R package called scDIV (acronym for single-cell RNA-sequencing data demultiplexing using interindividual variations).

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.

scCompressSA: dual-channel self-attention based deep autoencoder model for single-cell clustering by compressing gene-gene interactions

BACKGROUND

Single-cell clustering has played an important role in exploring the molecular mechanisms about cell differentiation and human diseases. Due to highly-stochastic transcriptomics data, accurate detection of cell types is still challenged, especially for RNA-sequencing data from human beings. In this case, deep neural networks have been increasingly employed to mine cell type specific patterns and have outperformed statistic approaches in cell clustering.

RESULTS

Using cross-correlation to capture gene-gene interactions, this study proposes the scCompressSA method to integrate topological patterns from scRNA-seq data, with support of self-attention (SA) based coefficient compression (CC) block. This SA-based CC block is able to extract and employ static gene-gene interactions from scRNA-seq data. This proposed scCompressSA method has enhanced clustering accuracy in multiple benchmark scRNA-seq datasets by integrating topological and temporal features.

CONCLUSION

Static gene-gene interactions have been extracted as temporal features to boost clustering performance in single-cell clustering For the scCompressSA method, dual-channel SA based CC block is able to integrate topological features and has exhibited extraordinary detection accuracy compared with previous clustering approaches that only employ temporal patterns.

DELVE: feature selection for preserving biological trajectories in single-cell data

Single-cell technologies can measure the expression of thousands of molecular features in individual cells undergoing dynamic biological processes. While examining cells along a computationally-ordered pseudotime trajectory can reveal how changes in gene or protein expression impact cell fate, identifying such dynamic features is challenging due to the inherent noise in single-cell data. Here, we present DELVE, an unsupervised feature selection method for identifying a representative subset of molecular features which robustly recapitulate cellular trajectories. In contrast to previous work, DELVE uses a bottom-up approach to mitigate the effects of confounding sources of variation, and instead models cell states from dynamic gene or protein modules based on core regulatory complexes. Using simulations, single-cell RNA sequencing, and iterative immunofluorescence imaging data in the context of cell cycle and cellular differentiation, we demonstrate how DELVE selects features that better define cell-types and cell-type transitions. DELVE is available as an open-source python package: https://github.com/jranek/delve .

MarkerMap: nonlinear marker selection for single-cell studies

Single-cell RNA-seq data allow the quantification of cell type differences across a growing set of biological contexts. However, pinpointing a small subset of genomic features explaining this variability can be ill-defined and computationally intractable. Here we introduce MarkerMap, a generative model for selecting minimal gene sets which are maximally informative of cell type origin and enable whole transcriptome reconstruction. MarkerMap provides a scalable framework for both supervised marker selection, aimed at identifying specific cell type populations, and unsupervised marker selection, aimed at gene expression imputation and reconstruction. We benchmark MarkerMap's competitive performance against previously published approaches on real single cell gene expression data sets. MarkerMap is available as a pip installable package, as a community resource aimed at developing explainable machine learning techniques for enhancing interpretability in single-cell studies.

Data integration and inference of gene regulation using single-cell temporal multimodal data with scTIE

Single-cell technologies offer unprecedented opportunities to dissect gene regulatory mechanisms in context-specific ways. Although there are computational methods for extracting gene regulatory relationships from scRNA-seq and scATAC-seq data, the data integration problem, essential for accurate cell type identification, has been mostly treated as a standalone challenge. Here we present scTIE, a unified method that integrates temporal multimodal data and infers regulatory relationships predictive of cellular state changes. scTIE uses an autoencoder to embed cells from all time points into a common space by using iterative optimal transport, followed by extracting interpretable information to predict cell trajectories. Using a variety of synthetic and real temporal multimodal data sets, we show scTIE achieves effective data integration while preserving more biological signals than existing methods, particularly in the presence of batch effects and noise. Furthermore, on the exemplar multiome data set we generated from differentiating mouse embryonic stem cells over time, we show scTIE captures regulatory elements highly predictive of cell transition probabilities, providing new potentials to understand the regulatory landscape driving developmental processes.

Anti-correlated feature selection prevents false discovery of subpopulations in scRNAseq

While sub-clustering cell-populations has become popular in single cell-omics, negative controls for this process are lacking. Popular feature-selection/clustering algorithms fail the null-dataset problem, allowing erroneous subdivisions of homogenous clusters until nearly each cell is called its own cluster. Using real and synthetic datasets, we find that anti-correlated gene selection reduces or eliminates erroneous subdivisions, increases marker-gene selection efficacy, and efficiently scales to millions of cells.

scHybridBERT: integrating gene regulation and cell graph for spatiotemporal dynamics in single-cell clustering

Graph learning models have received increasing attention in the computational analysis of single-cell RNA sequencing (scRNA-seq) data. Compared with conventional deep neural networks, graph neural networks and language models have exhibited superior performance by extracting graph-structured data from raw gene count matrices. Established deep neural network-based clustering approaches generally focus on temporal expression patterns while ignoring inherent interactions at gene-level as well as cell-level, which could be regarded as spatial dynamics in single-cell data. Both gene-gene and cell-cell interactions are able to boost the performance of cell type detection, under the framework of multi-view modeling. In this study, spatiotemporal embedding and cell graphs are extracted to capture spatial dynamics at the molecular level. In order to enhance the accuracy of cell type detection, this study proposes the scHybridBERT architecture to conduct multi-view modeling of scRNA-seq data using extracted spatiotemporal patterns. In this scHybridBERT method, graph learning models are employed to deal with cell graphs and the Performer model employs spatiotemporal embeddings. Experimental outcomes about benchmark scRNA-seq datasets indicate that the proposed scHybridBERT method is able to enhance the accuracy of single-cell clustering tasks by integrating spatiotemporal embeddings and cell graphs.

Cell Cycle Mapping Using Multiplexed Immunofluorescence

The development of technologies that allow measurement of the cell cycle at the single-cell level has revealed novel insights into the mechanisms that regulate cell cycle commitment and progression through DNA replication and cell division. These studies have also provided evidence of heterogeneity in cell cycle regulation among individual cells, even within a genetically identical population. Cell cycle mapping combines highly multiplexed imaging with manifold learning to visualize the diversity of "paths" that cells can take through the proliferative cell cycle or into various states of cell cycle arrest. In this chapter, we describe a general protocol of the experimental and computational components of cell cycle mapping. We also provide a comprehensive guide for the design and analysis of experiments, discussing key considerations in detail (e.g., antibody library preparation, analysis strategies, etc.) that may vary depending on the research question being addressed.

scFseCluster: a feature selection-enhanced clustering for single-cell RNA-seq data

Single-cell RNA sequencing (scRNA-seq) enables researchers to reveal previously unknown cell heterogeneity and functional diversity, which is impossible with bulk RNA sequencing. Clustering approaches are widely used for analyzing scRNA-seq data and identifying cell types and states. In the past few years, various advanced computational strategies emerged. However, the low generalization and high computational cost are the main bottlenecks of existing methods. In this study, we established a novel computational framework, scFseCluster, for scRNA-seq clustering analysis. scFseCluster incorporates a metaheuristic algorithm (Feature Selection based on Quantum Squirrel Search Algorithm) to extract the optimal gene set, which largely guarantees the performance of cell clustering. We conducted simulation experiments in several aspects to verify the performance of the proposed approach. scFseCluster performed very well on eight benchmark scRNA-seq datasets because of the optimal gene sets obtained using the Feature Selection based on Quantum Squirrel Search Algorithm. The comparative study demonstrated the significant advantages of scFseCluster over seven State-of-the-Art algorithms. In addition, our analysis shows that feature selection on high-variable genes can significantly improve clustering performance. In conclusion, our study demonstrates that scFseCluster is a highly versatile tool for enhancing scRNA-seq data clustering analysis.

Evaluation of deep learning-based feature selection for single-cell RNA sequencing data analysis

BACKGROUND

Feature selection is an essential task in single-cell RNA-seq (scRNA-seq) data analysis and can be critical for gene dimension reduction and downstream analyses, such as gene marker identification and cell type classification. Most popular methods for feature selection from scRNA-seq data are based on the concept of differential distribution wherein a statistical model is used to detect changes in gene expression among cell types. Recent development of deep learning-based feature selection methods provides an alternative approach compared to traditional differential distribution-based methods in that the importance of a gene is determined by neural networks.

RESULTS

In this work, we explore the utility of various deep learning-based feature selection methods for scRNA-seq data analysis. We sample from Tabula Muris and Tabula Sapiens atlases to create scRNA-seq datasets with a range of data properties and evaluate the performance of traditional and deep learning-based feature selection methods for cell type classification, feature selection reproducibility and diversity, and computational time.

CONCLUSIONS

Our study provides a reference for future development and application of deep learning-based feature selection methods for single-cell omics data analyses.

On the use of QDE-SVM for gene feature selection and cell type classification from scRNA-seq data

Cell type identification is one of the fundamental tasks in single-cell RNA sequencing (scRNA-seq) studies. It is a key step to facilitate downstream interpretations such as differential expression, trajectory inference, etc. scRNA-seq data contains technical variations that could affect the interpretation of the cell types. Therefore, gene selection, also known as feature selection in data science, plays an important role in selecting informative genes for scRNA-seq cell type identification. Generally speaking, feature selection methods are categorized into filter-, wrapper-, and embedded-based approaches. From the existing literature, methods from filter- and embedded-based approaches are widely applied in scRNA-seq gene selection tasks. The wrapper-based method that gives promising results in other fields has yet been extensively utilized for selecting gene features from scRNA-seq data; in addition, most of the existing wrapper methods used in this field are clustering instead of classification-based. With a large number of annotated data available today, this study applied a classification-based approach as an alternative to the clustering-based wrapper method. In our work, a quantum-inspired differential evolution (QDE) wrapped with a classification method was introduced to select a subset of genes from twelve well-known scRNA-seq transcriptomic datasets to identify cell types. In particular, the QDE was combined with different machine-learning (ML) classifiers namely logistic regression, decision tree, support vector machine (SVM) with linear and radial basis function kernels, as well as extreme learning machine. The linear SVM wrapped with QDE, namely QDE-SVM, was chosen by referring to the feature selection results from the experiment. QDE-SVM showed a superior cell type classification performance among QDE wrapping with other ML classifiers as well as the recent wrapper methods (i.e., FSCAM, SSD-LAHC, MA-HS, and BSF). QDE-SVM achieved an average accuracy of 0.9559, while the other wrapper methods achieved average accuracies in the range of 0.8292 to 0.8872.

Gene regulatory network reconstruction: harnessing the power of single-cell multi-omic data

Inferring gene regulatory networks (GRNs) is a fundamental challenge in biology that aims to unravel the complex relationships between genes and their regulators. Deciphering these networks plays a critical role in understanding the underlying regulatory crosstalk that drives many cellular processes and diseases. Recent advances in sequencing technology have led to the development of state-of-the-art GRN inference methods that exploit matched single-cell multi-omic data. By employing diverse mathematical and statistical methodologies, these methods aim to reconstruct more comprehensive and precise gene regulatory networks. In this review, we give a brief overview on the statistical and methodological foundations commonly used in GRN inference methods. We then compare and contrast the latest state-of-the-art GRN inference methods for single-cell matched multi-omics data, and discuss their assumptions, limitations and opportunities. Finally, we discuss the challenges and future directions that hold promise for further advancements in this rapidly developing field.

Integrative multi-omics and systems bioinformatics in translational neuroscience: A data mining perspective

Bioinformatic analysis of large and complex omics datasets has become increasingly useful in modern day biology by providing a great depth of information, with its application to neuroscience termed neuroinformatics. Data mining of omics datasets has enabled the generation of new hypotheses based on differentially regulated biological molecules associated with disease mechanisms, which can be tested experimentally for improved diagnostic and therapeutic targeting of neurodegenerative diseases. Importantly, integrating multi-omics data using a systems bioinformatics approach will advance the understanding of the layered and interactive network of biological regulation that exchanges systemic knowledge to facilitate the development of a comprehensive human brain profile. In this review, we first summarize data mining studies utilizing datasets from the individual type of omics analysis, including epigenetics/epigenomics, transcriptomics, proteomics, metabolomics, lipidomics, and spatial omics, pertaining to Alzheimer's disease, Parkinson's disease, and multiple sclerosis. We then discuss multi-omics integration approaches, including independent biological integration and unsupervised integration methods, for more intuitive and informative interpretation of the biological data obtained across different omics layers. We further assess studies that integrate multi-omics in data mining which provide convoluted biological insights and offer proof-of-concept proposition towards systems bioinformatics in the reconstruction of brain networks. Finally, we recommend a combination of high dimensional bioinformatics analysis with experimental validation to achieve translational neuroscience applications including biomarker discovery, therapeutic development, and elucidation of disease mechanisms. We conclude by providing future perspectives and opportunities in applying integrative multi-omics and systems bioinformatics to achieve precision phenotyping of neurodegenerative diseases and towards personalized medicine.

Recent Approaches to Design and Analysis of Electrical Impedance Systems for Single Cells Using Machine Learning

Individual cells have many unique properties that can be quantified to develop a holistic understanding of a population. This can include understanding population characteristics, identifying subpopulations, or elucidating outlier characteristics that may be indicators of disease. Electrical impedance measurements are rapid and label-free for the monitoring of single cells and generate large datasets of many cells at single or multiple frequencies. To increase the accuracy and sensitivity of measurements and define the relationships between impedance and biological features, many electrical measurement systems have incorporated machine learning (ML) paradigms for control and analysis. Considering the difficulty capturing complex relationships using traditional modelling and statistical methods due to population heterogeneity, ML offers an exciting approach to the systemic collection and analysis of electrical properties in a data-driven way. In this work, we discuss incorporation of ML to improve the field of electrical single cell analysis by addressing the design challenges to manipulate single cells and sophisticated analysis of electrical properties that distinguish cellular changes. Looking forward, we emphasize the opportunity to build on integrated systems to address common challenges in data quality and generalizability to save time and resources at every step in electrical measurement of single cells.

Ensemble deep learning of embeddings for clustering multimodal single-cell omics data

MOTIVATION

Recent advances in multimodal single-cell omics technologies enable multiple modalities of molecular attributes, such as gene expression, chromatin accessibility, and protein abundance, to be profiled simultaneously at a global level in individual cells. While the increasing availability of multiple data modalities is expected to provide a more accurate clustering and characterization of cells, the development of computational methods that are capable of extracting information embedded across data modalities is still in its infancy.

RESULTS

We propose SnapCCESS for clustering cells by integrating data modalities in multimodal single-cell omics data using an unsupervised ensemble deep learning framework. By creating snapshots of embeddings of multimodality using variational autoencoders, SnapCCESS can be coupled with various clustering algorithms for generating consensus clustering of cells. We applied SnapCCESS with several clustering algorithms to various datasets generated from popular multimodal single-cell omics technologies. Our results demonstrate that SnapCCESS is effective and more efficient than conventional ensemble deep learning-based clustering methods and outperforms other state-of-the-art multimodal embedding generation methods in integrating data modalities for clustering cells. The improved clustering of cells from SnapCCESS will pave the way for more accurate characterization of cell identity and types, an essential step for various downstream analyses of multimodal single-cell omics data.

AVAILABILITY AND IMPLEMENTATION

SnapCCESS is implemented as a Python package and is freely available from https://github.com/PYangLab/SnapCCESS under the open-source license of GPL-3. The data used in this study are publicly available (see section 'Data availability').

scTIE: data integration and inference of gene regulation using single-cell temporal multimodal data

Single-cell technologies offer unprecedented opportunities to dissect gene regulatory mechanisms in context-specific ways. Although there are computational methods for extracting gene regulatory relationships from scRNA-seq and scATAC-seq data, the data integration problem, essential for accurate cell type identification, has been mostly treated as a standalone challenge. Here we present scTIE, a unified method that integrates temporal multimodal data and infers regulatory relationships predictive of cellular state changes. scTIE uses an autoencoder to embed cells from all time points into a common space using iterative optimal transport, followed by extracting interpretable information to predict cell trajectories. Using a variety of synthetic and real temporal multimodal datasets, we demonstrate scTIE achieves effective data integration while preserving more biological signals than existing methods, particularly in the presence of batch effects and noise. Furthermore, on the exemplar multiome dataset we generated from differentiating mouse embryonic stem cells over time, we demonstrate scTIE captures regulatory elements highly predictive of cell transition probabilities, providing new potentials to understand the regulatory landscape driving developmental processes.

Feature selection for preserving biological trajectories in single-cell data

Single-cell technologies can readily measure the expression of thousands of molecular features from individual cells undergoing dynamic biological processes, such as cellular differentiation, immune response, and disease progression. While examining cells along a computationally ordered pseudotime offers the potential to study how subtle changes in gene or protein expression impact cell fate decision-making, identifying characteristic features that drive continuous biological processes remains difficult to detect from unenriched and noisy single-cell data. Given that all profiled sources of feature variation contribute to the cell-to-cell distances that define an inferred cellular trajectory, including confounding sources of biological variation (e.g. cell cycle or metabolic state) or noisy and irrelevant features (e.g. measurements with low signal-to-noise ratio) can mask the underlying trajectory of study and hinder inference. Here, we present DELVE (dynamic selection of locally covarying features), an unsupervised feature selection method for identifying a representative subset of dynamically-expressed molecular features that recapitulates cellular trajectories. In contrast to previous work, DELVE uses a bottom-up approach to mitigate the effect of unwanted sources of variation confounding inference, and instead models cell states from dynamic feature modules that constitute core regulatory complexes. Using simulations, single-cell RNA sequencing data, and iterative immunofluorescence imaging data in the context of the cell cycle and cellular differentiation, we demonstrate that DELVE selects features that more accurately characterize cell populations and improve the recovery of cell type transitions. This feature selection framework provides an alternative approach for improving trajectory inference and uncovering co-variation amongst features along a biological trajectory. DELVE is implemented as an open-source python package and is publicly available at: https://github.com/jranek/delve.

Benchmarking of analytical combinations for COVID-19 outcome prediction using single-cell RNA sequencing data

The advances of single-cell transcriptomic technologies have led to increasing use of single-cell RNA sequencing (scRNA-seq) data in large-scale patient cohort studies. The resulting high-dimensional data can be summarized and incorporated into patient outcome prediction models in several ways; however, there is a pressing need to understand the impact of analytical decisions on such model quality. In this study, we evaluate the impact of analytical choices on model choices, ensemble learning strategies and integrate approaches on patient outcome prediction using five scRNA-seq COVID-19 datasets. First, we examine the difference in performance between using single-view feature space versus multi-view feature space. Next, we survey multiple learning platforms from classical machine learning to modern deep learning methods. Lastly, we compare different integration approaches when combining datasets is necessary. Through benchmarking such analytical combinations, our study highlights the power of ensemble learning, consistency among different learning methods and robustness to dataset normalization when using multiple datasets as the model input.

A cofunctional grouping-based approach for non-redundant feature gene selection in unannotated single-cell RNA-seq analysis

Feature gene selection has significant impact on the performance of cell clustering in single-cell RNA sequencing (scRNA-seq) analysis. A well-rounded feature selection (FS) method should consider relevance, redundancy and complementarity of the features. Yet most existing FS methods focus on gene relevance to the cell types but neglect redundancy and complementarity, which undermines the cell clustering performance. We develop a novel computational method GeneClust to select feature genes for scRNA-seq cell clustering. GeneClust groups genes based on their expression profiles, then selects genes with the aim of maximizing relevance, minimizing redundancy and preserving complementarity. It can work as a plug-in tool for FS with any existing cell clustering method. Extensive benchmark results demonstrate that GeneClust significantly improve the clustering performance. Moreover, GeneClust can group cofunctional genes in biological process and pathway into clusters, thus providing a means of investigating gene interactions and identifying potential genes relevant to biological characteristics of the dataset. GeneClust is freely available at https://github.com/ToryDeng/scGeneClust.

Omics Data Preprocessing for Machine Learning: A Case Study in Childhood Obesity

The use of machine learning techniques for the construction of predictive models of disease outcomes (based on omics and other types of molecular data) has gained enormous relevance in the last few years in the biomedical field. Nonetheless, the virtuosity of omics studies and machine learning tools are subject to the proper application of algorithms as well as the appropriate pre-processing and management of input omics and molecular data. Currently, many of the available approaches that use machine learning on omics data for predictive purposes make mistakes in several of the following key steps: experimental design, feature selection, data pre-processing, and algorithm selection. For this reason, we propose the current work as a guideline on how to confront the main challenges inherent to multi-omics human data. As such, a series of best practices and recommendations are also presented for each of the steps defined. In particular, the main particularities of each omics data layer, the most suitable preprocessing approaches for each source, and a compilation of best practices and tips for the study of disease development prediction using machine learning are described. Using examples of real data, we show how to address the key problems mentioned in multi-omics research (e.g., biological heterogeneity, technical noise, high dimensionality, presence of missing values, and class imbalance). Finally, we define the proposals for model improvement based on the results found, which serve as the bases for future work.

Comprehensive characterization of fetal and mature retinal cell identity to assess the fidelity of retinal organoids

Characterizing cell identity in complex tissues such as the human retina is essential for studying its development and disease. While retinal organoids derived from pluripotent stem cells have been widely used to model development and disease of the human retina, there is a lack of studies that have systematically evaluated the molecular and cellular fidelity of the organoids derived from various culture protocols in recapitulating their in vivo counterpart. To this end, we performed an extensive meta-atlas characterization of cellular identities of the human eye, covering a wide range of developmental stages. The resulting map uncovered previously unknown biomarkers of major retinal cell types and those associated with cell-type-specific maturation. Using our retinal-cell-identity map from the fetal and adult tissues, we systematically assessed the fidelity of the retinal organoids in mimicking the human eye, enabling us to comprehensively benchmark the current protocols for retinal organoid generation.

A Comparison of the Various Methods for Selecting Features for Single-Cell RNA Sequencing Data in Alzheimer's Disease

The high-throughput sequencing method known as RNA-Seq records the whole transcriptome of individual cells. Single-cell RNA sequencing, also known as scRNA-Seq, is widely utilized in the field of biomedical research and has resulted in the generation of huge quantities and types of data. The noise and artifacts that are present in the raw data require extensive cleaning before they can be used. When applied to applications for machine learning or pattern recognition, feature selection methods offer a method to reduce the amount of time spent on calculation while simultaneously improving predictions and offering a better knowledge of the data. The process of discovering biomarkers is analogous to feature selection methods used in machine learning and is especially helpful for applications in the medical field. An attempt is made by a feature selection algorithm to cut down on the total number of features by eliminating those that are unnecessary or redundant while retaining those that are the most helpful.We apply FS algorithms designed for scRNA-Seq to Alzheimer's disease, which is the most prevalent neurodegenerative disease in the western world and causes cognitive and behavioral impairment. AD is clinically and pathologically varied, and genetic studies imply a diversity of biological mechanisms and pathways. Over 20 new Alzheimer's disease susceptibility loci have been discovered through linkage, genome-wide association, and next-generation sequencing (Tosto G, Reitz C, Mol Cell Probes 30:397-403, 2016). In this study, we focus on the performance of three different approaches to marker gene selection methods and compare them using the support vector machine (SVM), k-nearest neighbors' algorithm (k-NN), and linear discriminant analysis (LDA), which are mainly supervised classification algorithms.

Data analysis guidelines for single-cell RNA-seq in biomedical studies and clinical applications

The application of single-cell RNA sequencing (scRNA-seq) in biomedical research has advanced our understanding of the pathogenesis of disease and provided valuable insights into new diagnostic and therapeutic strategies. With the expansion of capacity for high-throughput scRNA-seq, including clinical samples, the analysis of these huge volumes of data has become a daunting prospect for researchers entering this field. Here, we review the workflow for typical scRNA-seq data analysis, covering raw data processing and quality control, basic data analysis applicable for almost all scRNA-seq data sets, and advanced data analysis that should be tailored to specific scientific questions. While summarizing the current methods for each analysis step, we also provide an online repository of software and wrapped-up scripts to support the implementation. Recommendations and caveats are pointed out for some specific analysis tasks and approaches. We hope this resource will be helpful to researchers engaging with scRNA-seq, in particular for emerging clinical applications.

scFeatures: multi-view representations of single-cell and spatial data for disease outcome prediction

MOTIVATION

With the recent surge of large-cohort scale single cell research, it is of critical importance that analytical methods can fully utilize the comprehensive characterization of cellular systems that single cell technologies produce to provide insights into samples from individuals. Currently, there is little consensus on the best ways to compress information from the complex data structures of these technologies to summary statistics that represent each sample (e.g. individuals).

RESULTS

Here, we present scFeatures, an approach that creates interpretable cellular and molecular representations of single-cell and spatial data at the sample level. We demonstrate that summarizing a broad collection of features at the sample level is both important for understanding underlying disease mechanisms in different experimental studies and for accurately classifying disease status of individuals.

AVAILABILITY AND IMPLEMENTATION

scFeatures is publicly available as an R package at https://github.com/SydneyBioX/scFeatures. All data used in this study are publicly available with accession ID reported in the Section 2.

SUPPLEMENTARY INFORMATION

Supplementary data are available at Bioinformatics online.

Detection of cell markers from single cell RNA-seq with sc2marker

Background

Single-cell RNA sequencing (scRNA-seq) allows the detection of rare cell types in complex tissues. The detection of markers for rare cell types is useful for further biological analysis of, for example, flow cytometry and imaging data sets for either physical isolation or spatial characterization of these cells. However, only a few computational approaches consider the problem of selecting specific marker genes from scRNA-seq data.

Results

Here, we propose sc2marker, which is based on the maximum margin index and a database of proteins with antibodies, to select markers for flow cytometry or imaging. We evaluated the performances of sc2marker and competing methods in ranking known markers in scRNA-seq data of immune and stromal cells. The results showed that sc2marker performed better than the competing methods in accuracy, while having a competitive running time.

Supplementary Information

The online version contains supplementary material available at 10.1186/s12859-022-04817-5.

A Combined Feature Screening Approach of Random Forest and Filter-based Methods for Ultra-high Dimensional Data

Background:

Various feature (variable) screening approaches have been proposed in the past decade to mitigate the impact of ultra-high dimensionality in classification and regression problems, including filter based methods such as sure indepen¬dence screening, and wrapper based methods such random forest. However, the former type of methods rely heavily on strong modelling assumptions while the latter ones requires an adequate sample size to make the data speak for themselves. These require¬ments can seldom be met in biochemical studies in cases where we have only access to ultra-high dimensional data with a complex structure and a small number of observations.

Objective:

In this research, we want to investigate the possibility of combing both filter based screening methods and random forest based screening methods in the regression context.

Method:

We have combined four state-of-art filter approaches, namely, sure independence screening (SIS) , robust rank corre¬lation based screening (RRCS), high dimensional ordinary least squares projection (HOLP) and a model free sure independence screening procedure based on the distance correlation (DCSIS) from the statistical community with a random forest based Boruta screening method from the machine learning community for regression problems.

Result:

Among all combined methods, RF-DCSIS performs better than the other methods in terms of screening accuracy and prediction capability on the simulated scenarios and real benchmark datasets.

Conclusion:

By empirical study from both extensive simulation and real data, we have shown that both filter based screening and random forest based screening have their pros and cons while a combination of both may lead to a better feature screening result and prediction capability

Keywords:

feature screening, filter-based method, ultra-high dimensional data, variable selection, random forest,RF-DCSIS

Graph Based Feature Selection for Reduction of Dimensionality in Next-Generation RNA Sequencing Datasets

Analysis of high-dimensional data, with more features (p) than observations (N) (p>N), places significant demand in cost and memory computational usage attributes. Feature selection can be used to reduce the dimensionality of the data. We used a graph-based approach, principal component analysis (PCA) and recursive feature elimination to select features for classification from RNAseq datasets from two lung cancer datasets. The selected features were discretized for association rule mining where support and lift were used to generate informative rules. Our results show that the graph-based feature selection improved the performance of sequential minimal optimization (SMO) and multilayer perceptron classifiers (MLP) in both datasets. In association rule mining, features selected using the graph-based approach outperformed the other two feature-selection techniques at a support of 0.5 and lift of 2. The non-redundant rules reflect the inherent relationships between features. Biological features are usually related to functions in living systems, a relationship that cannot be deduced by feature selection and classification alone. Therefore, the graph-based feature-selection approach combined with rule mining is a suitable way of selecting and finding associations between features in high-dimensional RNAseq data.