Data denoising with transfer learning in single-cell transcriptomics.
Ontology highlight
ABSTRACT:
SUBMITTER: Wang J
PROVIDER: S-EPMC7781045 | biostudies-literature | 2019 Sep
REPOSITORIES: biostudies-literature
Publications
Data denoising with transfer learning in single-cell transcriptomics.
Wang Jingshu J Agarwal Divyansh D Huang Mo M Hu Gang G Zhou Zilu Z Ye Chengzhong C Zhang Nancy R NR
Nature methods 20190830 9
Single-cell RNA sequencing (scRNA-seq) data are noisy and sparse. Here, we show that transfer learning across datasets remarkably improves data quality. By coupling a deep autoencoder with a Bayesian model, SAVER-X extracts transferable gene-gene relationships across data from different labs, varying conditions and divergent species, to denoise new target datasets. ...[more]
Similar Datasets
Project description:Recent developments of single-cell RNA-seq (scRNA-seq) technologies have led to enormous biological discoveries. As the scale of scRNA-seq studies increases, a major challenge in analysis is batch effects, which are inevitable in studies involving human tissues. Most existing methods remove batch effects in a low-dimensional embedding space. Although useful for clustering, batch effects are still present in the gene expression space, leaving downstream gene-level analysis susceptible to batch effects. Recent studies have shown that batch effect correction in the gene expression space is much harder than in the embedding space. Methods such as Seurat 3.0 rely on the mutual nearest neighbor (MNN) approach to remove batch effects in gene expression, but MNN can only analyze two batches at a time, and it becomes computationally infeasible when the number of batches is large. Here, we present CarDEC, a joint deep learning model that simultaneously clusters and denoises scRNA-seq data while correcting batch effects both in the embedding and the gene expression space. Comprehensive evaluations spanning different species and tissues showed that CarDEC outperforms Scanorama, DCA + Combat, scVI, and MNN. With CarDEC denoising, non-highly variable genes offer as much signal for clustering as the highly variable genes (HVGs), suggesting that CarDEC substantially boosted information content in scRNA-seq. We also showed that trajectory analysis using CarDEC's denoised and batch-corrected expression as input revealed marker genes and transcription factors that are otherwise obscured in the presence of batch effects. CarDEC is computationally fast, making it a desirable tool for large-scale scRNA-seq studies.
| S-EPMC8494213 | biostudies-literature
Project description:Large single-cell atlases are now routinely generated to serve as references for analysis of smaller-scale studies. Yet learning from reference data is complicated by batch effects between datasets, limited availability of computational resources and sharing restrictions on raw data. Here we introduce a deep learning strategy for mapping query datasets on top of a reference called single-cell architectural surgery (scArches). scArches uses transfer learning and parameter optimization to enable efficient, decentralized, iterative reference building and contextualization of new datasets with existing references without sharing raw data. Using examples from mouse brain, pancreas, immune and whole-organism atlases, we show that scArches preserves biological state information while removing batch effects, despite using four orders of magnitude fewer parameters than de novo integration. scArches generalizes to multimodal reference mapping, allowing imputation of missing modalities. Finally, scArches retains coronavirus disease 2019 (COVID-19) disease variation when mapping to a healthy reference, enabling the discovery of disease-specific cell states. scArches will facilitate collaborative projects by enabling iterative construction, updating, sharing and efficient use of reference atlases.
| S-EPMC8763644 | biostudies-literature
Project description:The development of single-cell sequencing technologies has allowed researchers to gain important new knowledge about the expression profile of genes in thousands of individual cells of a model organism or tissue. A common disadvantage of this technology is the loss of the three-dimensional (3-D) structure of the cells. Consequently, the Dialogue on Reverse Engineering Assessment and Methods (DREAM) organized the Single-Cell Transcriptomics Challenge, in which we participated, with the aim to address the following two problems: (a) to identify the top 60, 40, and 20 genes of the Drosophila melanogaster embryo that contain the most spatial information and (b) to reconstruct the 3-D arrangement of the embryo using information from those genes. We developed two independent techniques, leveraging machine learning models from least absolute shrinkage and selection operator (Lasso) and deep neural networks (NNs), which are applied to high-dimensional single-cell sequencing data in order to accurately identify genes that contain spatial information. Our first technique, Lasso.TopX, utilizes the Lasso and ranking statistics and allows a user to define a specific number of features they are interested in. The NN approach utilizes weak supervision for linear regression to accommodate for uncertain or probabilistic training labels. We show, individually for both techniques, that we are able to identify important, stable, and a user-defined number of genes containing the most spatial information. The results from both techniques achieve high performance when reconstructing spatial information in D. melanogaster and also generalize to zebrafish (Danio rerio). Furthermore, we identified novel D. melanogaster genes that carry important positional information and were not previously suspected. We also show how the indirect use of the full datasets' information can lead to data leakage and generate bias in overestimating the model's performance. Lastly, we discuss the applicability of our approaches to other feature selection problems outside the realm of single-cell sequencing and the importance of being able to handle probabilistic training labels. Our source code and detailed documentation are available at https://github.com/TJU-CMC-Org/SingleCell-DREAM/.
| S-EPMC7902049 | biostudies-literature
Project description:Reference-based cell-type annotation can significantly reduce time and effort in single-cell analysis by transferring labels from a previously-annotated dataset to a new dataset. However, label transfer by end-to-end computational methods is challenging due to the entanglement of technical (e.g., from different sequencing batches or techniques) and biological (e.g., from different cellular microenvironments) variations, only the first of which must be removed. To address this issue, we propose Polyphony, an interactive transfer learning (ITL) framework, to complement biologists' knowledge with advanced computational methods. Polyphony is motivated and guided by domain experts' needs for a controllable, interactive, and algorithm-assisted annotation process, identified through interviews with seven biologists. We introduce anchors, i.e., analogous cell populations across datasets, as a paradigm to explain the computational process and collect user feedback for model improvement. We further design a set of visualizations and interactions to empower users to add, delete, or modify anchors, resulting in refined cell type annotations. The effectiveness of this approach is demonstrated through quantitative experiments, two hypothetical use cases, and interviews with two biologists. The results show that our anchor-based ITL method takes advantage of both human and machine intelligence in annotating massive single-cell datasets.
| S-EPMC10039961 | biostudies-literature
Project description:Single-cell RNA sequencing (scRNA-seq) data provides unprecedented information on cell fate decisions; however, the spatial arrangement of cells is often lost. Several recent computational methods have been developed to impute spatial information onto a scRNA-seq dataset through analyzing known spatial expression patterns of a small subset of genes known as a reference atlas. However, there is a lack of comprehensive analysis of the accuracy, precision, and robustness of the mappings, along with the generalizability of these methods, which are often designed for specific systems. We present a system-adaptive deep learning-based method (DEEPsc) to impute spatial information onto a scRNA-seq dataset from a given spatial reference atlas. By introducing a comprehensive set of metrics that evaluate the spatial mapping methods, we compare DEEPsc with four existing methods on four biological systems. We find that while DEEPsc has comparable accuracy to other methods, an improved balance between precision and robustness is achieved. DEEPsc provides a data-adaptive tool to connect scRNA-seq datasets and spatial imaging datasets to analyze cell fate decisions. Our implementation with a uniform API can serve as a portal with access to all the methods investigated in this work for spatial exploration of cell fate decisions in scRNA-seq data. All methods evaluated in this work are implemented as an open-source software with a uniform interface.
| S-EPMC8021700 | biostudies-literature
Project description:Hematopoietic stem cells (HSCs) are an essential source and reservoir for normal hematopoiesis, and their function is compromised in many blood disorders. HSC research has benefitted from the recent development of single-cell molecular profiling technologies, where single-cell RNA sequencing (scRNA-seq) in particular has rapidly become an established method to profile HSCs and related hematopoietic populations. The classic definition of HSCs relies on transplantation assays, which have been used to validate HSC function for cell populations defined by flow cytometry. Flow cytometry information for single cells, however, is not available for many new high-throughput scRNA-seq methods, thus highlighting an urgent need for the establishment of alternative ways to pinpoint the likely HSCs within large scRNA-seq data sets. To address this, we tested a range of machine learning approaches and developed a tool, hscScore, to score single-cell transcriptomes from murine bone marrow based on their similarity to gene expression profiles of validated HSCs. We evaluated hscScore across scRNA-seq data from different laboratories, which allowed us to establish a robust method that functions across different technologies. To facilitate broad adoption of hscScore by the wider hematopoiesis community, we have made the trained model and example code freely available online. In summary, our method hscScore provides fast identification of mouse bone marrow HSCs from scRNA-seq measurements and represents a broadly useful tool for analysis of single-cell gene expression data.
| S-EPMC6900257 | biostudies-literature
Project description:BackgroundSingle-cell multiplex imaging data have provided new insights into disease subtypes and prognoses recently. However, quantitative models that explicitly capture single-cell resolution cell-cell interaction features to predict patient survival at a population scale are currently missing.MethodsWe quantified hundreds of single-cell resolution cell-cell interaction features through neighborhood calculation, in addition to cellular phenotypes. We applied these features to a neural-network-based Cox-nnet survival model to identify survival-associated features. We used non-negative matrix factorization (NMF) to identify patient survival subtypes. We identified atypical subpopulations of triple-negative breast cancer (TNBC) patients with moderate prognosis and Luminal A patients with poor prognosis and validated these subpopulations by label transferring using the UNION-COM method.ResultsThe neural-network-based Cox-nnet survival model using all cellular phenotype and cell-cell interaction features is highly predictive of patient survival in the test data (Concordance Index > 0.8). We identify seven survival subtypes using the top survival features, presenting distinct profiles of epithelial, immune, and fibroblast cells and their interactions. We reveal atypical subpopulations of TNBC patients with moderate prognosis (marked by GATA3 over-expression) and Luminal A patients with poor prognosis (marked by KRT6 and ACTA2 over-expression and CDH1 under-expression). These atypical subpopulations are validated in TCGA-BRCA and METABRIC datasets.ConclusionsThis work provides an approach to bridge single-cell level information toward population-level survival prediction.
| S-EPMC10730890 | biostudies-literature
Project description:Single-cell RNA-seq (scRNA-seq) is quite prevalent in studying transcriptomes, but it suffers from excessive zeros, some of which are true, but others are false. False zeros, which can be seen as missing data, obstruct the downstream analysis of single-cell RNA-seq data. How to distinguish true zeros from false ones is the key point of this problem. Here, we propose sparsity-penalized stacked denoising autoencoders (scSDAEs) to impute scRNA-seq data. scSDAEs adopt stacked denoising autoencoders with a sparsity penalty, as well as a layer-wise pretraining procedure to improve model fitting. scSDAEs can capture nonlinear relationships among the data and incorporate information about the observed zeros. We tested the imputation efficiency of scSDAEs on recovering the true values of gene expression and helping downstream analysis. First, we show that scSDAE can recover the true values and the sample-sample correlations of bulk sequencing data with simulated noise. Next, we demonstrate that scSDAEs accurately impute RNA mixture dataset with different dilutions, spike-in RNA concentrations affected by technical zeros, and improves the consistency of RNA and protein levels in CITE-seq data. Finally, we show that scSDAEs can help downstream clustering analysis. In this study, we develop a deep learning-based method, scSDAE, to impute single-cell RNA-seq affected by technical zeros. Furthermore, we show that scSDAEs can recover the true values, to some extent, and help downstream analysis.
| S-EPMC7291078 | biostudies-literature
Project description:Multimodal single-cell profiling methods that measure protein expression with oligo-conjugated antibodies hold promise for comprehensive dissection of cellular heterogeneity, yet the resulting protein counts have substantial technical noise that can mask biological variations. Here we integrate experiments and computational analyses to reveal two major noise sources and develop a method called "dsb" (denoised and scaled by background) to normalize and denoise droplet-based protein expression data. We discover that protein-specific noise originates from unbound antibodies encapsulated during droplet generation; this noise can thus be accurately estimated and corrected by utilizing protein levels in empty droplets. We also find that isotype control antibodies and the background protein population average in each cell exhibit significant correlations across single cells, we thus use their shared variance to correct for cell-to-cell technical noise in each cell. We validate these findings by analyzing the performance of dsb in eight independent datasets spanning multiple technologies, including CITE-seq, ASAP-seq, and TEA-seq. Compared to existing normalization methods, our approach improves downstream analyses by better unmasking biologically meaningful cell populations. Our method is available as an open-source R package that interfaces easily with existing single cell software platforms such as Seurat, Bioconductor, and Scanpy and can be accessed at "dsb [ https://cran.r-project.org/package=dsb ]".
| S-EPMC9018908 | biostudies-literature
Project description:Batch effects in single-cell RNA-seq data pose a significant challenge for comparative analyses across samples, individuals, and conditions. Although batch effect correction methods are routinely applied, data integration often leads to overcorrection and can result in the loss of biological variability. In this work we present STACAS, a batch correction method for scRNA-seq that leverages prior knowledge on cell types to preserve biological variability upon integration. Through an open-source benchmark, we show that semi-supervised STACAS outperforms state-of-the-art unsupervised methods, as well as supervised methods such as scANVI and scGen. STACAS scales well to large datasets and is robust to incomplete and imprecise input cell type labels, which are commonly encountered in real-life integration tasks. We argue that the incorporation of prior cell type information should be a common practice in single-cell data integration, and we provide a flexible framework for semi-supervised batch effect correction.
| S-EPMC10825117 | biostudies-literature