Interpretable dimensionality reduction of single cell transcriptome data with deep generative models

Unmixing Autoencoder for Image Reconstruction from Hyperspectral Data

Due to the complexity of samples and the limitations in spatial resolution, the spectra in hyperspectral imaging (HSI) are generally contributed to by multiple components, making univariate analysis ineffective. Although feature extraction methods have been applied, the chemical meaning of the compressed variables is difficult to interpret, limiting their further applications. An unmixing autoencoder (UAE) was developed in this work for the separation of the mixed spectra in HSI. The proposed model is composed of an encoder and a fully connected (FC) layer. The former is used to compress the input spectrum into several variables, and the latter is employed to reconstruct the spectrum. Combining reconstruction loss and sparse regularization, the weights and the spectral profiles of the components will be encoded in the compressed variables and the connection weights of FC, respectively. A simulated and three experimental HSI data sets were adopted to investigate the performance of the UAE model. The spectral components were successfully obtained, from which the handwriting under papers was revealed from the image of near-infrared (NIR) diffusive reflectance spectroscopy, and the images of lipids, proteins, and nucleic acids were reconstructed from the Raman and stimulated Raman scattering (SRS) images.

A comprehensive survey of dimensionality reduction and clustering methods for single-cell and spatial transcriptomics data

In recent years, the application of single-cell transcriptomics and spatial transcriptomics analysis techniques has become increasingly widespread. Whether dealing with single-cell transcriptomic or spatial transcriptomic data, dimensionality reduction and clustering are indispensable. Both single-cell and spatial transcriptomic data are often high-dimensional, making the analysis and visualization of such data challenging. Through dimensionality reduction, it becomes possible to visualize the data in a lower-dimensional space, allowing for the observation of relationships and differences between cell subpopulations. Clustering enables the grouping of similar cells into the same cluster, aiding in the identification of distinct cell subpopulations and revealing cellular diversity, providing guidance for downstream analyses. In this review, we systematically summarized the most widely recognized algorithms employed for the dimensionality reduction and clustering analysis of single-cell transcriptomic and spatial transcriptomic data. This endeavor provides valuable insights and ideas that can contribute to the development of novel tools in this rapidly evolving field.

scIALM: A method for sparse scRNA-seq expression matrix imputation using the Inexact Augmented Lagrange Multiplier with low error

Single-cell RNA sequencing (scRNA-seq) is a high-throughput sequencing technology that quantifies gene expression profiles of specific cell populations at the single-cell level, providing a foundation for studying cellular heterogeneity and patient pathological characteristics. It is effective for developmental, fertility, and disease studies. However, the cell-gene expression matrix of single-cell sequencing data is often sparse and contains numerous zero values. Some of the zero values derive from noise, where dropout noise has a large impact on downstream analysis. In this paper, we propose a method named scIALM for imputation recovery of sparse single-cell RNA data expression matrices, which employs the Inexact Augmented Lagrange Multiplier method to use sparse but clean (accurate) data to recover unknown entries in the matrix. We perform experimental analysis on four datasets, calling the expression matrix after Quality Control (QC) as the original matrix, and comparing the performance of scIALM with six other methods using mean squared error (MSE), mean absolute error (MAE), Pearson correlation coefficient (PCC), and cosine similarity (CS). Our results demonstrate that scIALM accurately recovers the original data of the matrix with an error of 10e-4, and the mean value of the four metrics reaches 4.5072 (MSE), 0.765 (MAE), 0.8701 (PCC), 0.8896 (CS). In addition, at 10%-50% random masking noise, scIALM is the least sensitive to the masking ratio. For downstream analysis, this study uses adjusted rand index (ARI) and normalized mutual information (NMI) to evaluate the clustering effect, and the results are improved on three datasets containing real cluster labels.

Addressing scalability and managing sparsity and dropout events in single-cell representation identification with ZIGACL

Despite significant advancements in single-cell representation learning, scalability and managing sparsity and dropout events continue to challenge the field as scRNA-seq datasets expand. While current computational tools struggle to maintain both efficiency and accuracy, the accurate connection of these dropout events to specific biological functions usually requires additional, complex experiments, often hampered by potential inaccuracies in cell-type annotation. To tackle these challenges, the Zero-Inflated Graph Attention Collaborative Learning (ZIGACL) method has been developed. This innovative approach combines a Zero-Inflated Negative Binomial model with a Graph Attention Network, leveraging mutual information from neighboring cells to enhance dimensionality reduction and apply dynamic adjustments to the learning process through a co-supervised deep graph clustering model. ZIGACL's integration of denoising and topological embedding significantly improves clustering accuracy and ensures similar cells are grouped closely in the latent space. Comparative analyses across nine real scRNA-seq datasets have shown that ZIGACL significantly enhances single-cell data analysis by offering superior clustering performance and improved stability in cell representations, effectively addressing scalability and managing sparsity and dropout events, thereby advancing our understanding of cellular heterogeneity.

Strategic Multi-Omics Data Integration via Multi-Level Feature Contrasting and Matching

The analysis and comprehension of multi-omics data has emerged as a prominent topic in the field of bioinformatics and data science. However, the sparsity characteristics and high dimensionality of omics data pose difficulties in terms of extracting meaningful information. Moreover, the heterogeneity inherent in multiple omics sources makes the effective integration of multi-omics data challenging To tackle these challenges, we propose MFCC-SAtt, a multi-level feature contrast clustering model based on self-attention to extract informative features from multi-omics data. MFCC-SAtt treats each omics type as a distinct modality and employs autoencoders with self-attention for each modality to integrate and compress their respective features into a shared feature space. By utilizing a multi-level feature extraction framework along with incorporating a semantic information extractor, we mitigate optimization conflicts arising from different learning objectives. Additionally, MFCC-SAtt guides deep clustering based on multi-level features which further enhances the quality of output labels. By conducting extensive experiments on multi-omics data, we have validated the exceptional performance of MFCC-SAtt. For instance, in a pan-cancer clustering task, MFCC-SAtt achieved an accuracy of over 80.38%.

Machine learning and related approaches in transcriptomics

Data acquisition for transcriptomic studies used to be the bottleneck in the transcriptomic analytical pipeline. However, recent developments in transcriptome profiling technologies have increased researchers' ability to obtain data, resulting in a shift in focus to data analysis. Incorporating machine learning to traditional analytical methods allows the possibility of handling larger volumes of complex data more efficiently. Many bioinformaticians, especially those unfamiliar with ML in the study of human transcriptomics and complex biological systems, face a significant barrier stemming from their limited awareness of the current landscape of ML utilisation in this field. To address this gap, this review endeavours to introduce those individuals to the general types of ML, followed by a comprehensive range of more specific techniques, demonstrated through examples of their incorporation into analytical pipelines for human transcriptome investigations. Important computational aspects such as data pre-processing, task formulation, results (performance of ML models), and validation methods are encompassed. In hope of better practical relevance, there is a strong focus on studies published within the last five years, almost exclusively examining human transcriptomes, with outcomes compared with standard non-ML tools.

Joint inference of discrete cell types and continuous type-specific variability in single-cell datasets with MMIDAS

Reproducible definition and identification of cell types is essential to enable investigations into their biological function and to understand their relevance in the context of development, disease and evolution. Current approaches model variability in data as continuous latent factors, followed by clustering as a separate step, or immediately apply clustering on the data. We show that such approaches can suffer from qualitative mistakes in identifying cell types robustly, particularly when the number of such cell types is in the hundreds or even thousands. Here we propose an unsupervised method, Mixture Model Inference with Discrete-coupled AutoencoderS (MMIDAS), which combines a generalized mixture model with a multi-armed deep neural network to jointly infer the discrete type and continuous type-specific variability. Using four recent datasets of brain cells spanning different technologies, species and conditions, we demonstrate that MMIDAS can identify reproducible cell types and infer cell type-dependent continuous variability in both unimodal and multimodal datasets.

scNODE : generative model for temporal single cell transcriptomic data prediction

SUMMARY

Measurement of single-cell gene expression at different timepoints enables the study of cell development. However, due to the resource constraints and technical challenges associated with the single-cell experiments, researchers can only profile gene expression at discrete and sparsely sampled timepoints. This missing timepoint information impedes downstream cell developmental analyses. We propose scNODE, an end-to-end deep learning model that can predict in silico single-cell gene expression at unobserved timepoints. scNODE integrates a variational autoencoder with neural ordinary differential equations to predict gene expression using a continuous and nonlinear latent space. Importantly, we incorporate a dynamic regularization term to learn a latent space that is robust against distribution shifts when predicting single-cell gene expression at unobserved timepoints. Our evaluations on three real-world scRNA-seq datasets show that scNODE achieves higher predictive performance than state-of-the-art methods. We further demonstrate that scNODE's predictions help cell trajectory inference under the missing timepoint paradigm and the learned latent space is useful for in silico perturbation analysis of relevant genes along a developmental cell path.

AVAILABILITY AND IMPLEMENTATION

The data and code are publicly available at https://github.com/rsinghlab/scNODE.

Machine learning for the advancement of genome-scale metabolic modeling

Constraint-based modeling (CBM) has evolved as the core systems biology tool to map the interrelations between genotype, phenotype, and external environment. The recent advancement of high-throughput experimental approaches and multi-omics strategies has generated a plethora of new and precise information from wide-ranging biological domains. On the other hand, the continuously growing field of machine learning (ML) and its specialized branch of deep learning (DL) provide essential computational architectures for decoding complex and heterogeneous biological data. In recent years, both multi-omics and ML have assisted in the escalation of CBM. Condition-specific omics data, such as transcriptomics and proteomics, helped contextualize the model prediction while analyzing a particular phenotypic signature. At the same time, the advanced ML tools have eased the model reconstruction and analysis to increase the accuracy and prediction power. However, the development of these multi-disciplinary methodological frameworks mainly occurs independently, which limits the concatenation of biological knowledge from different domains. Hence, we have reviewed the potential of integrating multi-disciplinary tools and strategies from various fields, such as synthetic biology, CBM, omics, and ML, to explore the biochemical phenomenon beyond the conventional biological dogma. How the integrative knowledge of these intersected domains has improved bioengineering and biomedical applications has also been highlighted. We categorically explained the conventional genome-scale metabolic model (GEM) reconstruction tools and their improvement strategies through ML paradigms. Further, the crucial role of ML and DL in omics data restructuring for GEM development has also been briefly discussed. Finally, the case-study-based assessment of the state-of-the-art method for improving biomedical and metabolic engineering strategies has been elaborated. Therefore, this review demonstrates how integrating experimental and in silico strategies can help map the ever-expanding knowledge of biological systems driven by condition-specific cellular information. This multiview approach will elevate the application of ML-based CBM in the biomedical and bioengineering fields for the betterment of society and the environment.

Variational inference of single cell time series

Time course single-cell RNA sequencing (scRNA-seq) enables researchers to probe genome-wide expression dynamics at the the single cell scale. However, when gene expression is affected jointly by time and cellular identity, analyzing such data - including conducting cell type annotation and modeling cell type-dependent dynamics - becomes challenging. To address this problem, we propose SNOW (SiNgle cell flOW map), a deep learning algorithm to deconvolve single cell time series data into time-dependent and time-independent contributions. SNOW has a number of advantages. First, it enables cell type annotation based on the time-independent dimensions. Second, it yields a probabilistic model that can be used to discriminate between biological temporal variation and batch effects contaminating individual timepoints, and provides an approach to mitigate batch effects. Finally, it is capable of projecting cells forward and backward in time, yielding time series at the individual cell level. This enables gene expression dynamics to be studied without the need for clustering or pseudobulking, which can be error prone and result in information loss. We describe our probabilistic framework in detail and demonstrate SNOW using data from three distinct time course scRNA-seq studies. Our results show that SNOW is able to construct biologically meaningful latent spaces, remove batch effects, and generate realistic time-series at the single-cell level. By way of example, we illustrate how the latter may be used to enhance the detection of cell type-specific circadian gene expression rhythms, and may be readily extended to other time-series analyses.

Metal-detection based techniques and their applications in metallobiology

Metals are essential for human health and play a crucial role in numerous biological processes and pathways. Gaining a deeper insight into these biological events will facilitate novel strategies for disease prevention, early detection, and personalized treatment. In recent years, there has been significant progress in the development of metal-detection based techniques from single cell metallome and proteome profiling to multiplex imaging, which greatly enhance our comprehension of the intricate roles played by metals in complex biological systems. This perspective summarizes the recent progress in advanced metal-detection based techniques and highlights successful applications in elucidating the roles of metals in biology and medicine. Technologies including machine learning that couple with single-cell analysis such as mass cytometry and their application in metallobiology, cancer biology and immunology are also emphasized. Finally, we provide insights into future prospects and challenges involved in metal-detection based techniques, with the aim of inspiring further methodological advancements and applications that are accessible to chemists, biologists, and clinicians.

A variational autoencoder trained with priors from canonical pathways increases the interpretability of transcriptome data

Interpreting transcriptome data is an important yet challenging aspect of bioinformatic analysis. While gene set enrichment analysis is a standard tool for interpreting regulatory changes, we utilize deep learning techniques, specifically autoencoder architectures, to learn latent variables that drive transcriptome signals. We investigate whether simple, variational autoencoder (VAE), and beta-weighted VAE are capable of learning reduced representations of transcriptomes that retain critical biological information. We propose a novel VAE that utilizes priors from biological data to direct the network to learn a representation of the transcriptome that is based on understandable biological concepts. After benchmarking five different autoencoder architectures, we found that each succeeded in reducing the transcriptomes to 50 latent dimensions, which captured enough variation for accurate reconstruction. The simple, fully connected autoencoder, performs best across the benchmarks, but lacks the characteristic of having directly interpretable latent dimensions. The beta-weighted, prior-informed VAE implementation is able to solve the benchmarking tasks, and provide semantically accurate latent features equating to biological pathways. This study opens a new direction for differential pathway analysis in transcriptomics with increased transparency and interpretability.

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.

K-nearest-neighbors induced topological PCA for single cell RNA-sequence data analysis

Single-cell RNA sequencing (scRNA-seq) is widely used to reveal heterogeneity in cells, which has given us insights into cell-cell communication, cell differentiation, and differential gene expression. However, analyzing scRNA-seq data is a challenge due to sparsity and the large number of genes involved. Therefore, dimensionality reduction and feature selection are important for removing spurious signals and enhancing downstream analysis. Traditional PCA, a main workhorse in dimensionality reduction, lacks the ability to capture geometrical structure information embedded in the data, and previous graph Laplacian regularizations are limited by the analysis of only a single scale. We propose a topological Principal Components Analysis (tPCA) method by the combination of persistent Laplacian (PL) technique and L2,1 norm regularization to address multiscale and multiclass heterogeneity issues in data. We further introduce a k-Nearest-Neighbor (kNN) persistent Laplacian technique to improve the robustness of our persistent Laplacian method. The proposed kNN-PL is a new algebraic topology technique which addresses the many limitations of the traditional persistent homology. Rather than inducing filtration via the varying of a distance threshold, we introduced kNN-tPCA, where filtrations are achieved by varying the number of neighbors in a kNN network at each step, and find that this framework has significant implications for hyper-parameter tuning. We validate the efficacy of our proposed tPCA and kNN-tPCA methods on 11 diverse benchmark scRNA-seq datasets, and showcase that our methods outperform other unsupervised PCA enhancements from the literature, as well as popular Uniform Manifold Approximation (UMAP), t-Distributed Stochastic Neighbor Embedding (tSNE), and Projection Non-Negative Matrix Factorization (NMF) by significant margins. For example, tPCA provides up to 628%, 78%, and 149% improvements to UMAP, tSNE, and NMF, respectively on classification in the F1 metric, and kNN-tPCA offers 53%, 63%, and 32% improvements to UMAP, tSNE, and NMF, respectively on clustering in the ARI metric.

Comprehensive single-cell RNA-seq analysis using deep interpretable generative modeling guided by biological hierarchy knowledge

Recent advances in microfluidics and sequencing technologies allow researchers to explore cellular heterogeneity at single-cell resolution. In recent years, deep learning frameworks, such as generative models, have brought great changes to the analysis of transcriptomic data. Nevertheless, relying on the potential space of these generative models alone is insufficient to generate biological explanations. In addition, most of the previous work based on generative models is limited to shallow neural networks with one to three layers of latent variables, which may limit the capabilities of the models. Here, we propose a deep interpretable generative model called d-scIGM for single-cell data analysis. d-scIGM combines sawtooth connectivity techniques and residual networks, thereby constructing a deep generative framework. In addition, d-scIGM incorporates hierarchical prior knowledge of biological domains to enhance the interpretability of the model. We show that d-scIGM achieves excellent performance in a variety of fundamental tasks, including clustering, visualization, and pseudo-temporal inference. Through topic pathway studies, we found that d-scIGM-learned topics are better enriched for biologically meaningful pathways compared to the baseline models. Furthermore, the analysis of drug response data shows that d-scIGM can capture drug response patterns in large-scale experiments, which provides a promising way to elucidate the underlying biological mechanisms. Lastly, in the melanoma dataset, d-scIGM accurately identified different cell types and revealed multiple melanin-related driver genes and key pathways, which are critical for understanding disease mechanisms and drug development.

Single-cell colocalization analysis using a deep generative model

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

Analyzing Molecular Dynamics Trajectories Thermodynamically through Artificial Intelligence

Molecular dynamics simulations produce trajectories that correspond to vast amounts of structure when exploring biochemical processes. Extracting valuable information, e.g., important intermediate states and collective variables (CVs) that describe the major movement modes, from molecular trajectories to understand the underlying mechanisms of biological processes presents a significant challenge. To achieve this goal, we introduce a deep learning approach, coined DIKI (deep identification of key intermediates), to determine low-dimensional CVs distinguishing key intermediate conformations without a-priori assumptions. DIKI dynamically plans the distribution of latent space and groups together similar conformations within the same cluster. Moreover, by incorporating two user-defined parameters, namely, coarse focus knob and fine focus knob, to help identify conformations with low free energy and differentiate the subtle distinctions among these conformations, resolution-tunable clustering was achieved. Furthermore, the integration of DIKI with a path-finding algorithm contributes to the identification of crucial intermediates along the lowest free-energy pathway. We postulate that DIKI is a robust and flexible tool that can find widespread applications in the analysis of complex biochemical processes.

Intelligent phenotype-detection and gene expression profile generation with generative adversarial networks

Gene expression analysis is valuable for cancer type classification and identifying diverse cancer phenotypes. The latest high-throughput RNA sequencing devices have enabled access to large volumes of gene expression data. However, we face several challenges, such as data security and privacy, when we develop machine learning-based classifiers for categorizing cancer types with these datasets. To address these issues, we propose IP3G (Intelligent Phenotype-detection and Gene expression profile Generation with Generative adversarial network), a model based on Generative Adversarial Networks. IP3G tackles two major problems: augmenting gene expression data and unsupervised phenotype discovery. By converting gene expression profiles into 2-Dimensional images and leveraging IP3G, we generate new profiles for specific phenotypes. IP3G learns disentangled representations of gene expression patterns and identifies phenotypes without labeled data. We improve the objective function of the GAN used in IP3G by employing the earth mover distance and a novel mutual information function. IP3G outperforms clustering methods like k-Means, DBSCAN, and GMM in unsupervised phenotype discovery, while also surpassing SVM and CNN classification accuracy by up to 6% through gene expression profile augmentation. The source code for the developed IP3G is accessible to the public on GitHub.

scCAN: Clustering With Adaptive Neighbor-Based Imputation Method for Single-Cell RNA-Seq Data

Single-cell RNA sequencing (scRNA-seq) is widely used to study cellular heterogeneity in different samples. However, due to technical deficiencies, dropout events often result in zero gene expression values in the gene expression matrix. In this paper, we propose a new imputation method called scCAN, based on adaptive neighborhood clustering, to estimate the zero value of dropouts. Our method continuously updates cell-cell similarity information by simultaneously learning similarity relationships, clustering structures, and imposing new rank constraints on the Laplacian matrix of the similarity matrix, improving the imputation of dropout zero values. To evaluate the performance of this method, we used four simulated and eight real scRNA-seq data for downstream analyses, including cell clustering, recovered gene expression, and reconstructed cell trajectories. Our method improves the performance of the downstream analysis and is better than other imputation methods.

Variable selection for nonlinear dimensionality reduction of biological datasets through bootstrapping of correlation networks

Identifying the most relevant variables or features in massive datasets for dimensionality reduction can lead to improved and more informative display, faster computation times, and more explainable models of complex systems. Despite significant advances and available algorithms, this task generally remains challenging, especially in unsupervised settings. In this work, we propose a method that constructs correlation networks using all intervening variables and then selects the most informative ones based on network bootstrapping. The method can be applied in both supervised and unsupervised scenarios. We demonstrate its functionality by applying Uniform Manifold Approximation and Projection for dimensionality reduction to several high-dimensional biological datasets, derived from 4D live imaging recordings of hundreds of morpho-kinetic variables, describing the dynamics of thousands of individual leukocytes at sites of prominent inflammation. We compare our method with other standard ones in the field, such as Principal Component Analysis and Elastic Net, showing that it outperforms them. The proposed method can be employed in a wide range of applications, encompassing data analysis and machine learning.

scNAT: a deep learning method for integrating paired single-cell RNA and T cell receptor sequencing profiles

Many deep learning-based methods have been proposed to handle complex single-cell data. Deep learning approaches may also prove useful to jointly analyze single-cell RNA sequencing (scRNA-seq) and single-cell T cell receptor sequencing (scTCR-seq) data for novel discoveries. We developed scNAT, a deep learning method that integrates paired scRNA-seq and scTCR-seq data to represent data in a unified latent space for downstream analysis. We demonstrate that scNAT is capable of removing batch effects, and identifying cell clusters and a T cell migration trajectory from blood to cerebrospinal fluid in multiple sclerosis.

Cytomulate: accurate and efficient simulation of CyTOF data

Recently, many analysis tools have been devised to offer insights into data generated via cytometry by time-of-flight (CyTOF). However, objective evaluations of these methods remain absent as most evaluations are conducted against real data where the ground truth is generally unknown. In this paper, we develop Cytomulate, a reproducible and accurate simulation algorithm of CyTOF data, which could serve as a foundation for future method development and evaluation. We demonstrate that Cytomulate can capture various characteristics of CyTOF data and is superior in learning overall data distributions than single-cell RNA-seq-oriented methods such as scDesign2, Splatter, and generative models like LAMBDA.

Similarity-assisted variational autoencoder for nonlinear dimension reduction with application to single-cell RNA sequencing data

BACKGROUND

Deep generative models naturally become nonlinear dimension reduction tools to visualize large-scale datasets such as single-cell RNA sequencing datasets for revealing latent grouping patterns or identifying outliers. The variational autoencoder (VAE) is a popular deep generative method equipped with encoder/decoder structures. The encoder and decoder are useful when a new sample is mapped to the latent space and a data point is generated from a point in a latent space. However, the VAE tends not to show grouping pattern clearly without additional annotation information. On the other hand, similarity-based dimension reduction methods such as t-SNE or UMAP present clear grouping patterns even though these methods do not have encoder/decoder structures.

RESULTS

To bridge this gap, we propose a new approach that adopts similarity information in the VAE framework. In addition, for biological applications, we extend our approach to a conditional VAE to account for covariate effects in the dimension reduction step. In the simulation study and real single-cell RNA sequencing data analyses, our method shows great performance compared to existing state-of-the-art methods by producing clear grouping structures using an inferred encoder and decoder. Our method also successfully adjusts for covariate effects, resulting in more useful dimension reduction.

CONCLUSIONS

Our method is able to produce clearer grouping patterns than those of other regularized VAE methods by utilizing similarity information encoded in the data via the highly celebrated UMAP loss function.

Dimensionality reduction methods for extracting functional networks from large-scale CRISPR screens

CRISPR-Cas9 screens facilitate the discovery of gene functional relationships and phenotype-specific dependencies. The Cancer Dependency Map (DepMap) is the largest compendium of whole-genome CRISPR screens aimed at identifying cancer-specific genetic dependencies across human cell lines. A mitochondria-associated bias has been previously reported to mask signals for genes involved in other functions, and thus, methods for normalizing this dominant signal to improve co-essentiality networks are of interest. In this study, we explore three unsupervised dimensionality reduction methods-autoencoders, robust, and classical principal component analyses (PCA)-for normalizing the DepMap to improve functional networks extracted from these data. We propose a novel "onion" normalization technique to combine several normalized data layers into a single network. Benchmarking analyses reveal that robust PCA combined with onion normalization outperforms existing methods for normalizing the DepMap. Our work demonstrates the value of removing low-dimensional signals from the DepMap before constructing functional gene networks and provides generalizable dimensionality reduction-based normalization tools.

Collaborative analysis for drug discovery by federated learning on non-IID data

With the increasing availability of large-scale QSAR (Quantitative Structure-Activity Relationship) datasets, collaborative analysis has become a promising approach for drug discovery. Traditional centralized analysis which typically concentrates data on a central server for training faces challenges such as data privacy and security. Distributed analysis such as federated learning offers a solution by enabling collaborative model training without sharing raw data. However, it may fail when the training data in the local devices are non-independent and identically distributed (non-IID). In this paper, we propose a novel framework for collaborative drug discovery using federated learning on non-IID datasets. We address the difficulty of training on non-IID data by globally sharing a small subset of data among all institutions. Our framework allows multiple institutions to jointly train a robust predictive model while preserving the privacy of their individual data. We leverage the federated learning paradigm to distribute the model training process across local devices, eliminating the need for data exchange. The experimental results on 15 benchmark datasets demonstrate that the proposed method achieves competitive predictive accuracy to centralized analysis while respecting data privacy. Moreover, our framework offers benefits such as reduced data transmission and enhanced scalability, making it suitable for large-scale collaborative drug discovery efforts.

Identifying dynamical persistent biomarker structures for rare events using modern integrative machine learning approach

The evolution of omics and computational competency has accelerated discoveries of the underlying biological processes in an unprecedented way. High throughput methodologies, such as flow cytometry, can reveal deeper insights into cell processes, thereby allowing opportunities for scientific discoveries related to health and diseases. However, working with cytometry data often imposes complex computational challenges due to high-dimensionality, large size, and nonlinearity of the data structure. In addition, cytometry data frequently exhibit diverse patterns across biomarkers and suffer from substantial class imbalances which can further complicate the problem. The existing methods of cytometry data analysis either predict cell population or perform feature selection. Through this study, we propose a "wisdom of the crowd" approach to simultaneously predict rare cell populations and perform feature selection by integrating a pool of modern machine learning (ML) algorithms. Given that our approach integrates superior performing ML models across different normalization techniques based on entropy and rank, our method can detect diverse patterns existing across the model features. Furthermore, the method identifies a dynamic biomarker structure that divides the features into persistently selected, unselected, and fluctuating assemblies indicating the role of each biomarker in rare cell prediction, which can subsequently aid in studies of disease progression.

Single-cell analysis technologies for cancer research: from tumor-specific single cell discovery to cancer therapy

Single-cell sequencing (SCS) technology is changing our understanding of cellular components, functions, and interactions across organisms, because of its inherent advantage of avoiding noise resulting from genotypic and phenotypic heterogeneity across numerous samples. By directly and individually measuring multiple molecular characteristics of thousands to millions of single cells, SCS technology can characterize multiple cell types and uncover the mechanisms of gene regulatory networks, the dynamics of transcription, and the functional state of proteomic profiling. In this context, we conducted systematic research on SCS techniques, including the fundamental concepts, procedural steps, and applications of scDNA, scRNA, scATAC, scCITE, and scSNARE methods, focusing on the unique clinical advantages of SCS, particularly in cancer therapy. We have explored challenging but critical areas such as circulating tumor cells (CTCs), lineage tracing, tumor heterogeneity, drug resistance, and tumor immunotherapy. Despite challenges in managing and analyzing the large amounts of data that result from SCS, this technique is expected to reveal new horizons in cancer research. This review aims to emphasize the key role of SCS in cancer research and promote the application of single-cell technologies to cancer therapy.

Essential procedures of single-cell RNA sequencing in multiple myeloma and its translational value

Multiple myeloma (MM) is a malignant neoplasm characterized by clonal proliferation of abnormal plasma cells. In many countries, it ranks as the second most prevalent malignant neoplasm of the hematopoietic system. Although treatment methods for MM have been continuously improved and the survival of patients has been dramatically prolonged, MM remains an incurable disease with a high probability of recurrence. As such, there are still many challenges to be addressed. One promising approach is single-cell RNA sequencing (scRNA-seq), which can elucidate the transcriptome heterogeneity of individual cells and reveal previously unknown cell types or states in complex tissues. In this review, we outlined the experimental workflow of scRNA-seq in MM, listed some commonly used scRNA-seq platforms and analytical tools. In addition, with the advent of scRNA-seq, many studies have made new progress in the key molecular mechanisms during MM clonal evolution, cell interactions and molecular regulation in the microenvironment, and drug resistance mechanisms in target therapy. We summarized the main findings and sequencing platforms for applying scRNA-seq to MM research and proposed broad directions for targeted therapies based on these findings.

Vis-SPLIT: Interactive Hierarchical Modeling for mRNA Expression Classification

We propose an interactive visual analytics tool, Vis-SPLIT, for partitioning a population of individuals into groups with similar gene signatures. Vis-SPLIT allows users to interactively explore a dataset and exploit visual separations to build a classification model for specific cancers. The visualization components reveal gene expression and correlation to assist specific partitioning decisions, while also providing overviews for the decision model and clustered genetic signatures. We demonstrate the effectiveness of our framework through a case study and evaluate its usability with domain experts. Our results show that Vis-SPLIT can classify patients based on their genetic signatures to effectively gain insights into RNA sequencing data, as compared to an existing classification system.

Deep learning applications in single-cell genomics and transcriptomics data analysis

Traditional bulk sequencing methods are limited to measuring the average signal in a group of cells, potentially masking heterogeneity, and rare populations. The single-cell resolution, however, enhances our understanding of complex biological systems and diseases, such as cancer, the immune system, and chronic diseases. However, the single-cell technologies generate massive amounts of data that are often high-dimensional, sparse, and complex, thus making analysis with traditional computational approaches difficult and unfeasible. To tackle these challenges, many are turning to deep learning (DL) methods as potential alternatives to the conventional machine learning (ML) algorithms for single-cell studies. DL is a branch of ML capable of extracting high-level features from raw inputs in multiple stages. Compared to traditional ML, DL models have provided significant improvements across many domains and applications. In this work, we examine DL applications in genomics, transcriptomics, spatial transcriptomics, and multi-omics integration, and address whether DL techniques will prove to be advantageous or if the single-cell omics domain poses unique challenges. Through a systematic literature review, we have found that DL has not yet revolutionized the most pressing challenges of the single-cell omics field. However, using DL models for single-cell omics has shown promising results (in many cases outperforming the previous state-of-the-art models) in data preprocessing and downstream analysis. Although developments of DL algorithms for single-cell omics have generally been gradual, recent advances reveal that DL can offer valuable resources in fast-tracking and advancing research in single-cell.

Alignment of spatial genomics data using deep Gaussian processes

Spatially resolved genomic technologies have allowed us to study the physical organization of cells and tissues, and promise an understanding of local interactions between cells. However, it remains difficult to precisely align spatial observations across slices, samples, scales, individuals and technologies. Here, we propose a probabilistic model that aligns spatially-resolved samples onto a known or unknown common coordinate system (CCS) with respect to phenotypic readouts (for example, gene expression). Our method, Gaussian Process Spatial Alignment (GPSA), consists of a two-layer Gaussian process: the first layer maps observed samples' spatial locations onto a CCS, and the second layer maps from the CCS to the observed readouts. Our approach enables complex downstream spatially aware analyses that are impossible or inaccurate with unaligned data, including an analysis of variance, creation of a dense three-dimensional (3D) atlas from sparse two-dimensional (2D) slices or association tests across data modalities.

Accurate noise-robust classification of Bacillus species from MALDI-TOF MS spectra using a denoising autoencoder

Bacillus strains are ubiquitous in the environment and are widely used in the microbiological industry as valuable enzyme sources, as well as in agriculture to stimulate plant growth. The Bacillus genus comprises several closely related groups of species. The rapid classification of these remains challenging using existing methods. Techniques based on MALDI-TOF MS data analysis hold significant promise for fast and precise microbial strains classification at both the genus and species levels. In previous work, we proposed a geometric approach to Bacillus strain classification based on mass spectra analysis via the centroid method (CM). One limitation of such methods is the noise in MS spectra. In this study, we used a denoising autoencoder (DAE) to improve bacteria classification accuracy under noisy MS spectra conditions. We employed a denoising autoencoder approach to convert noisy MS spectra into latent variables representing molecular patterns in the original MS data, and the Random Forest method to classify bacterial strains by latent variables. Comparison of the DAE-RF with the CM method using the artificially noisy test samples showed that DAE-RF offers higher noise robustness. Hence, the DAE-RF method could be utilized for noise-robust, fast, and neat classification of Bacillus species according to MALDI-TOF MS data.

An introduction to representation learning for single-cell data analysis

Single-cell-resolved systems biology methods, including omics- and imaging-based measurement modalities, generate a wealth of high-dimensional data characterizing the heterogeneity of cell populations. Representation learning methods are routinely used to analyze these complex, high-dimensional data by projecting them into lower-dimensional embeddings. This facilitates the interpretation and interrogation of the structures, dynamics, and regulation of cell heterogeneity. Reflecting their central role in analyzing diverse single-cell data types, a myriad of representation learning methods exist, with new approaches continually emerging. Here, we contrast general features of representation learning methods spanning statistical, manifold learning, and neural network approaches. We consider key steps involved in representation learning with single-cell data, including data pre-processing, hyperparameter optimization, downstream analysis, and biological validation. Interdependencies and contingencies linking these steps are also highlighted. This overview is intended to guide researchers in the selection, application, and optimization of representation learning strategies for current and future single-cell research applications.

Generative modeling of single-cell gene expression for dose-dependent chemical perturbations

Single-cell sequencing reveals the heterogeneity of cellular response to chemical perturbations. However, testing all relevant combinations of cell types, chemicals, and doses is a daunting task. A deep generative learning formalism called variational autoencoders (VAEs) has been effective in predicting single-cell gene expression perturbations for single doses. Here, we introduce single-cell variational inference of dose-response (scVIDR), a VAE-based model that predicts both single-dose and multiple-dose cellular responses better than existing models. We show that scVIDR can predict dose-dependent gene expression across mouse hepatocytes, human blood cells, and cancer cell lines. We biologically interpret the latent space of scVIDR using a regression model and use scVIDR to order individual cells based on their sensitivity to chemical perturbation by assigning each cell a "pseudo-dose" value. We envision that scVIDR can help reduce the need for repeated animal testing across tissues, chemicals, and doses.

Computational Methods for Single-Cell Proteomics

Advances in single-cell proteomics technologies have resulted in high-dimensional datasets comprising millions of cells that are capable of answering key questions about biology and disease. The advent of these technologies has prompted the development of computational tools to process and visualize the complex data. In this review, we outline the steps of single-cell and spatial proteomics analysis pipelines. In addition to describing available methods, we highlight benchmarking studies that have identified advantages and pitfalls of the currently available computational toolkits. As these technologies continue to advance, robust analysis tools should be developed in tandem to take full advantage of the potential biological insights provided by these data.

Deciphering the Biology of Circulating Tumor Cells through Single-Cell RNA Sequencing: Implications for Precision Medicine in Cancer

Circulating tumor cells (CTCs) hold unique biological characteristics that directly involve them in hematogenous dissemination. Studying CTCs systematically is technically challenging due to their extreme rarity and heterogeneity and the lack of specific markers to specify metastasis-initiating CTCs. With cutting-edge technology, single-cell RNA sequencing (scRNA-seq) provides insights into the biology of metastatic processes driven by CTCs. Transcriptomics analysis of single CTCs can decipher tumor heterogeneity and phenotypic plasticity for exploring promising novel therapeutic targets. The integrated approach provides a perspective on the mechanisms underlying tumor development and interrogates CTCs interactions with other blood cell types, particularly those of the immune system. This review aims to comprehensively describe the current study on CTC transcriptomic analysis through scRNA-seq technology. We emphasize the workflow for scRNA-seq analysis of CTCs, including enrichment, single cell isolation, and bioinformatic tools applied for this purpose. Furthermore, we elucidated the translational knowledge from the transcriptomic profile of individual CTCs and the biology of cancer metastasis for developing effective therapeutics through targeting key pathways in CTCs.

Feature importance-based interpretation of UMAP-visualized polymer space

Dimensionality reduction (DR) techniques are used for various purposes such as exploratory data analysis. A commonly employed linear DR technique is principal component analysis (PCA), which is one of the most popular methods for DR. Owing to its linear nature, PCA enables the determination of axes in a low-dimensional space and the calculation of corresponding loading vectors. However, PCA cannot necessarily extract important features of non-linearly distributed data. This study presents a technique aimed at aiding the interpretation of data reduced through non-linear DR methods. In the proposed method, non-linear dimensionally reduced data was clustered via a density-based clustering method. Thereafter, the obtained cluster labels were classified by random forest (RF) classifiers. Further, feature importance (FI) of RF classifiers and Spearman's rank correlation coefficients between predictive probabilities to obtained clusters and original feature values were utilized for characterizing the visualized dimensionally reduced data. The results revealed that the proposed method can provide the interpretable FI-based images of the handwritten digits dataset. Moreover, the proposed method was also applied to the polymer dataset. The study found that incorporating signed FI was advantageous in achieving a meaningful interpretation. Furthermore, Gaussian process regression was utilized to produce intuitive FI-based heatmaps on a 2-dimensional space for greater ease of understanding. Additionally, to enhance the interpretability of the obtained clusters, a feature selection technique called Boruta was applied. The Boruta feature selection method worked effectively to interpret the obtained clusters with limited and commonly important features. Additionally, the study suggested that computing FI solely from substructure-based descriptors could further enhance the interpretability of the results. Finally, the automation of the proposed method was investigated, and through maximizing the target score based on the quality of both the DR and clustering, indicative results were automatically obtained for both the handwritten digits and polymer datasets.

Greedy design space construction based on regression and latent space extraction for pharmaceutical development

Implementation of the design space (DS) is a scientific concept for ensuring quality to be submitted as a part of the regulatory filing of a drug product for approval to market. An empirical approach is constructing the DS based on the regression model whose inputs are process parameters and material attributes over the different unit operations, i.e., a high-dimensional statistical model. While the high-dimensional model assures quality and process flexibility through a comprehensive process understanding, it has difficulty visualizing the feasible range of input parameters, i.e., DS. Therefore, this study proposes a greedy approach to constructing the extensive and flexible low-dimensional DS based on the high-dimensional statistical model and the observed internal representations that satisfies both comprehensive process understanding and the DS visualization capability. Introducing the observed correlation structure enabled the dimensionality reduction of the DS. The non-critical controllable parameters were fixed to the target values in visualizing the low-dimensional DS as a function of critical parameters. The expected variation of non-critical non-controllable parameters was considered the source of variation in prediction. The case study demonstrated the proposed approach's usefulness for developing the pharmaceutical manufacturing process.

ScInfoVAE: interpretable dimensional reduction of single cell transcription data with variational autoencoders and extended mutual information regularization

Single-cell RNA-sequencing (scRNA-seq) data can serve as a good indicator of cell-to-cell heterogeneity and can aid in the study of cell growth by identifying cell types. Recently, advances in Variational Autoencoder (VAE) have demonstrated their ability to learn robust feature representations for scRNA-seq. However, it has been observed that VAEs tend to ignore the latent variables when combined with a decoding distribution that is too flexible. In this paper, we introduce ScInfoVAE, a dimensional reduction method based on the mutual information variational autoencoder (InfoVAE), which can more effectively identify various cell types in scRNA-seq data of complex tissues. A joint InfoVAE deep model and zero-inflated negative binomial distributed model design based on ScInfoVAE reconstructs the objective function to noise scRNA-seq data and learn an efficient low-dimensional representation of it. We use ScInfoVAE to analyze the clustering performance of 15 real scRNA-seq datasets and demonstrate that our method provides high clustering performance. In addition, we use simulated data to investigate the interpretability of feature extraction, and visualization results show that the low-dimensional representation learned by ScInfoVAE retains local and global neighborhood structure data well. In addition, our model can significantly improve the quality of the variational posterior.

Review of single-cell RNA-seq data clustering for cell-type identification and characterization

In recent years, the advances in single-cell RNA-seq techniques have enabled us to perform large-scale transcriptomic profiling at single-cell resolution in a high-throughput manner. Unsupervised learning such as data clustering has become the central component to identify and characterize novel cell types and gene expression patterns. In this study, we review the existing single-cell RNA-seq data clustering methods with critical insights into the related advantages and limitations. In addition, we also review the upstream single-cell RNA-seq data processing techniques such as quality control, normalization, and dimension reduction. We conduct performance comparison experiments to evaluate several popular single-cell RNA-seq clustering approaches on simulated and multiple single-cell transcriptomic data sets.

Engineering an AI-based forward-reverse platform for the design of cross-ribosome binding sites of a transcription factor biosensor

A cross-ribosome binding site (cRBS) adjusts the dynamic range of transcription factor-based biosensors (TFBs) by controlling protein expression and folding. The rational design of a cRBS with desired TFB dynamic range remains an important issue in TFB forward and reverse engineering. Here, we report a novel artificial intelligence (AI)-based forward-reverse engineering platform for TFB dynamic range prediction and de novo cRBS design with selected TFB dynamic ranges. The platform demonstrated superior in processing unbalanced minority-class datasets and was guided by sequence characteristics from trained cRBSs. The platform identified correlations between cRBSs and dynamic ranges to mimic bidirectional design between these factors based on Wasserstein generative adversarial network (GAN) with a gradient penalty (GP) (WGAN-GP) and balancing GAN with GP (BAGAN-GP). For forward and reverse engineering, the predictive accuracy was up to 98% and 82%, respectively. Collectively, we generated an AI-based method for the rational design of TFBs with desired dynamic ranges.

Structure-preserving visualization for single-cell RNA-Seq profiles using deep manifold transformation with batch-correction

Dimensionality reduction and visualization play an important role in biological data analysis, such as data interpretation of single-cell RNA sequences (scRNA-seq). It is desired to have a visualization method that can not only be applicable to various application scenarios, including cell clustering and trajectory inference, but also satisfy a variety of technical requirements, especially the ability to preserve inherent structure of data and handle with batch effects. However, no existing methods can accommodate these requirements in a unified framework. In this paper, we propose a general visualization method, deep visualization (DV), that possesses the ability to preserve inherent structure of data and handle batch effects and is applicable to a variety of datasets from different application domains and dataset scales. The method embeds a given dataset into a 2- or 3-dimensional visualization space, with either a Euclidean or hyperbolic metric depending on a specified task type with type static (at a time point) or dynamic (at a sequence of time points) scRNA-seq data, respectively. Specifically, DV learns a structure graph to describe the relationships between data samples, transforms the data into visualization space while preserving the geometric structure of the data and correcting batch effects in an end-to-end manner. The experimental results on nine datasets in complex tissue from human patients or animal development demonstrate the competitiveness of DV in discovering complex cellular relations, uncovering temporal trajectories, and addressing complex batch factors. We also provide a preliminary attempt to pre-train a DV model for visualization of new incoming data.

Comparative analysis of dimension reduction methods for cytometry by time-of-flight data

While experimental and informatic techniques around single cell sequencing (scRNA-seq) are advanced, research around mass cytometry (CyTOF) data analysis has severely lagged behind. CyTOF data are notably different from scRNA-seq data in many aspects. This calls for the evaluation and development of computational methods specific for CyTOF data. Dimension reduction (DR) is one of the critical steps of single cell data analysis. Here, we benchmark the performances of 21 DR methods on 110 real and 425 synthetic CyTOF samples. We find that less well-known methods like SAUCIE, SQuaD-MDS, and scvis are the overall best performers. In particular, SAUCIE and scvis are well balanced, SQuaD-MDS excels at structure preservation, whereas UMAP has great downstream analysis performance. We also find that t-SNE (along with SQuad-MDS/t-SNE Hybrid) possesses the best local structure preservation. Nevertheless, there is a high level of complementarity between these tools, so the choice of method should depend on the underlying data structure and the analytical needs.

Dimensionality reduction methods for extracting functional networks from large-scale CRISPR screens

CRISPR-Cas9 screens facilitate the discovery of gene functional relationships and phenotype-specific dependencies. The Cancer Dependency Map (DepMap) is the largest compendium of whole-genome CRISPR screens aimed at identifying cancer-specific genetic dependencies across human cell lines. A mitochondria-associated bias has been previously reported to mask signals for genes involved in other functions, and thus, methods for normalizing this dominant signal to improve co-essentiality networks are of interest. In this study, we explore three unsupervised dimensionality reduction methods - autoencoders, robust, and classical principal component analyses (PCA) - for normalizing the DepMap to improve functional networks extracted from these data. We propose a novel "onion" normalization technique to combine several normalized data layers into a single network. Benchmarking analyses reveal that robust PCA combined with onion normalization outperforms existing methods for normalizing the DepMap. Our work demonstrates the value of removing low-dimensional signals from the DepMap before constructing functional gene networks and provides generalizable dimensionality reduction-based normalization tools.

Deep Nonnegative Matrix Factorization Using a Variational Autoencoder With Application to Single-Cell RNA Sequencing Data

Single-cell RNA sequencing is used to analyze the gene expression data of individual cells, thereby adding to existing knowledge of biological phenomena. Accordingly, this technology is widely used in numerous biomedical studies. Recently, the variational autoencoder has emerged and has been adopted for the analysis of single-cell data owing to its high capacity to manage large-scale data. Many different variants of the variational autoencoder have been applied, and have yielded superior results. However, because it is nonlinear, the model does not provide parameters that can be used to explain the underlying biological patterns. In this paper, we propose an interpretable nonnegative matrix factorization method that decomposes parameters into those shared across cells and those that are cell-specific. Effective nonlinear dimension reduction was achieved via a variational autoencoder applied to the cell-specific parameters. In addition to achieving nonlinear dimension reduction, our model could estimate the cell-type-specific gene expression. To improve the estimation accuracy, we introduced log-regularization, which reflects the single-cell property. Overall, our approach displayed excellent performance in a simulation study and in real data analyses, while maintaining good biological interpretability.

Discovery of drug-omics associations in type 2 diabetes with generative deep-learning models

The application of multiple omics technologies in biomedical cohorts has the potential to reveal patient-level disease characteristics and individualized response to treatment. However, the scale and heterogeneous nature of multi-modal data makes integration and inference a non-trivial task. We developed a deep-learning-based framework, multi-omics variational autoencoders (MOVE), to integrate such data and applied it to a cohort of 789 people with newly diagnosed type 2 diabetes with deep multi-omics phenotyping from the DIRECT consortium. Using in silico perturbations, we identified drug-omics associations across the multi-modal datasets for the 20 most prevalent drugs given to people with type 2 diabetes with substantially higher sensitivity than univariate statistical tests. From these, we among others, identified novel associations between metformin and the gut microbiota as well as opposite molecular responses for the two statins, simvastatin and atorvastatin. We used the associations to quantify drug-drug similarities, assess the degree of polypharmacy and conclude that drug effects are distributed across the multi-omics modalities.

Semi-Supervised Deep Learning for Cell Type Identification From Single-Cell Transcriptomic Data

Cell type identification from single-cell transcriptomic data is a common goal of single-cell RNA sequencing (scRNAseq) data analysis. Deep neural networks have been employed to identify cell types from scRNAseq data with high performance. However, it requires a large mount of individual cells with accurate and unbiased annotated types to train the identification models. Unfortunately, labeling the scRNAseq data is cumbersome and time-consuming as it involves manual inspection of marker genes. To overcome this challenge, we propose a semi-supervised learning model "SemiRNet" to use unlabeled scRNAseq cells and a limited amount of labeled scRNAseq cells to implement cell identification. The proposed model is based on recurrent convolutional neural networks (RCNN), which includes a shared network, a supervised network and an unsupervised network. The proposed model is evaluated on two large scale single-cell transcriptomic datasets. It is observed that the proposed model is able to achieve encouraging performance by learning on the very limited amount of labeled scRNAseq cells together with a large number of unlabeled scRNAseq cells.

Complex hierarchical structures in single-cell genomics data unveiled by deep hyperbolic manifold learning

With the advances in single-cell sequencing techniques, numerous analytical methods have been developed for delineating cell development. However, most are based on Euclidean space, which would distort the complex hierarchical structure of cell differentiation. Recently, methods acting on hyperbolic space have been proposed to visualize hierarchical structures in single-cell RNA-seq (scRNA-seq) data and have been proven to be superior to methods acting on Euclidean space. However, these methods have fundamental limitations and are not optimized for the highly sparse single-cell count data. To address these limitations, we propose scDHMap, a model-based deep learning approach to visualize the complex hierarchical structures of scRNA-seq data in low-dimensional hyperbolic space. The evaluations on extensive simulation and real experiments show that scDHMap outperforms existing dimensionality-reduction methods in various common analytical tasks as needed for scRNA-seq data, including revealing trajectory branches, batch correction, and denoising the count matrix with high dropout rates. In addition, we extend scDHMap to visualize single-cell ATAC-seq data.

Improved downstream functional analysis of single-cell RNA-sequence data using DGAN

The dramatic increase in the number of single-cell RNA-sequence (scRNA-seq) investigations is indeed an endorsement of the new-fangled proficiencies of next generation sequencing technologies that facilitate the accurate measurement of tens of thousands of RNA expression levels at the cellular resolution. Nevertheless, missing values of RNA amplification persist and remain as a significant computational challenge, as these data omission induce further noise in their respective cellular data and ultimately impede downstream functional analysis of scRNA-seq data. Consequently, it turns imperative to develop robust and efficient scRNA-seq data imputation methods for improved downstream functional analysis outcomes. To overcome this adversity, we have designed an imputation framework namely deep generative autoencoder network [DGAN]. In essence, DGAN is an evolved variational autoencoder designed to robustly impute data dropouts in scRNA-seq data manifested as a sparse gene expression matrix. DGAN principally reckons count distribution, besides data sparsity utilizing a gaussian model whereby, cell dependencies are capitalized to detect and exclude outlier cells via imputation. When tested on five publicly available scRNA-seq data, DGAN outperformed every single baseline method paralleled, with respect to downstream functional analysis including cell data visualization, clustering, classification and differential expression analysis. DGAN is executed in Python and is accessible at https://github.com/dikshap11/DGAN .

Structure-preserved dimension reduction using joint triplets sampling for multi-batch integration of single-cell transcriptomic data

Dimension reduction (DR) plays an important role in single-cell RNA sequencing (scRNA-seq), such as data interpretation, visualization and other downstream analysis. A desired DR method should be applicable to various application scenarios, including identifying cell types, preserving the inherent structure of data and handling with batch effects. However, most of the existing DR methods fail to accommodate these requirements simultaneously, especially removing batch effects. In this paper, we develop a novel structure-preserved dimension reduction (SPDR) method using intra- and inter-batch triplets sampling. The constructed triplets jointly consider each anchor's mutual nearest neighbors from inter-batch, k-nearest neighbors from intra-batch and randomly selected cells from the whole data, which capture higher order structure information and meanwhile account for batch information of the data. Then we minimize a robust loss function for the chosen triplets to obtain a structure-preserved and batch-corrected low-dimensional representation. Comprehensive evaluations show that SPDR outperforms other competing DR methods, such as INSCT, IVIS, Trimap, Scanorama, scVI and UMAP, in removing batch effects, preserving biological variation, facilitating visualization and improving clustering accuracy. Besides, the two-dimensional (2D) embedding of SPDR presents a clear and authentic expression pattern, and can guide researchers to determine how many cell types should be identified. Furthermore, SPDR is robust to complex data characteristics (such as down-sampling, duplicates and outliers) and varying hyperparameter settings. We believe that SPDR will be a valuable tool for characterizing complex cellular heterogeneity.