Ensemble learning for classifying single-cell data and projection across reference atlases.
Ontology highlight
ABSTRACT: Summary
Single-cell data are being generated at an accelerating pace. How best to project data across single-cell atlases is an open problem. We developed a boosted learner that overcomes the greatest challenge with status quo classifiers: low sensitivity, especially when dealing with rare cell types. By comparing novel and published data from distinct scRNA-seq modalities that were acquired from the same tissues, we show that this approach preserves cell-type labels when mapping across diverse platforms.Availability and implementation
https://github.com/diazlab/ELSA.Contact
aaron.diaz@ucsf.edu.Supplementary information
Supplementary data are available at Bioinformatics online.
SUBMITTER: Wang L
PROVIDER: S-EPMC7267838 | biostudies-literature |
REPOSITORIES: biostudies-literature
Similar Datasets
Project description:Single-cell data are being generated at an accelerating pace. How best to project data across single-cell atlases is an open problem. We developed a boosted learner that overcomes the greatest challenge with status quo classifiers: low sensitivity, especially when dealing with rare cell types. By comparing novel and published data from distinct scRNA-seq modalities that were acquired from the same tissues, we show that this approach preserves cell-type labels when mapping across diverse platforms.
2020-03-01 | GSE141982 | GEO
Project description:Ensemble learning for classifying single-cell data and projection across reference atlases
| PRJNA595499 | ENA
Project description:Single-cell data are being generated at an accelerating pace.How best to project data across single-cell atlases is an open problem. We developed a boosted learner that overcomes the greatest challenge with status quo classifiers: low sensitivity, especially when dealing with rare cell types.
| EGAS00001004283 | EGA
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:Single-cell RNA sequencing (scRNA-seq) has revealed an unprecedented degree of immune cell diversity. However, consistent definition of cell subtypes and cell states across studies and diseases remains a major challenge. Here we generate reference T cell atlases for cancer and viral infection by multi-study integration, and develop ProjecTILs, an algorithm for reference atlas projection. In contrast to other methods, ProjecTILs allows not only accurate embedding of new scRNA-seq data into a reference without altering its structure, but also characterizing previously unknown cell states that "deviate" from the reference. ProjecTILs accurately predicts the effects of cell perturbations and identifies gene programs that are altered in different conditions and tissues. A meta-analysis of tumor-infiltrating T cells from several cohorts reveals a strong conservation of T cell subtypes between human and mouse, providing a consistent basis to describe T cell heterogeneity across studies, diseases, and species.
| S-EPMC8137700 | biostudies-literature
Project description:The definition of neuronal type and how it relates to the transcriptome are open questions. Drosophila olfactory projection neurons (PNs) are among the best-characterized neuronal types: different PN classes target dendrites to distinct olfactory glomeruli, while PNs of the same class exhibit indistinguishable anatomical and physiological properties. Using single-cell RNA sequencing, we comprehensively characterized the transcriptomes of most PN classes and unequivocally mapped transcriptomes to specific olfactory function for six classes. Transcriptomes of closely related PN classes exhibit the largest differences during circuit assembly but become indistinguishable in adults, suggesting that neuronal subtype diversity peaks during development. Transcription factors and cell-surface molecules are the most differentially expressed genes between classes and are highly informative in encoding cell identity, enabling us to identify a new lineage-specific transcription factor that instructs PN dendrite targeting. These findings establish that neuronal transcriptomic identity corresponds with anatomical and physiological identity defined by connectivity and function.
| S-EPMC6095479 | biostudies-literature
Project description:How a neuronal cell type is defined and how this relates to its transcriptome are still open questions. The Drosophila olfactory projection neurons (PNs) are among the bestcharacterized neuronal types: Different PN classes target dendrites to distinct olfactory glomeruli and PNs of the same class exhibit indistinguishable anatomical and physiological properties. Using single-cell RNA-sequencing, we comprehensively characterized the transcriptomes of 40 PN classes and unequivocally identified transcriptomes for 6 classes. We found a new lineage-specific transcription factor that instructs PN dendrite targeting. Transcriptomes of closely-related PN classes exhibit the largest difference during circuit assembly, but become indistinguishable in adults, suggesting that neuronal subtype diversity peaks during development. Genes encoding transcription factors and cell-surface molecules are the most differentially expressed, indicating their central roles in specifying neuronal identity. Finally, we show that PNs use highly redundant combinatorial molecular codes to distinguish subtypes, enabling robust specification of cell identity and circuit assembly.
2017-10-21 | GSE100058 | GEO
Project description:Single-cell sequencing provides novel means to interpret the transcriptomic profiles of individual cells. To obtain in-depth analysis of single-cell sequencing, it requires effective computational methods to accurately predict single-cell clusters because single-cell sequencing techniques only provide the transcriptomic profiles of each cell. Although an accurate estimation of the cell-to-cell similarity is an essential first step to derive reliable single-cell clustering results, it is challenging to obtain the accurate similarity measurement because it highly depends on a selection of genes for similarity evaluations and the optimal set of genes for the accurate similarity estimation is typically unknown. Moreover, due to technical limitations, single-cell sequencing includes a larger number of artificial zeros, and the technical noise makes it difficult to develop effective single-cell clustering algorithms. Here, we describe a novel single-cell clustering algorithm that can accurately predict single-cell clusters in large-scale single-cell sequencing by effectively reducing the zero-inflated noise and accurately estimating the cell-to-cell similarities. First, we construct an ensemble similarity network based on different similarity estimates, and reduce the artificial noise using a random walk with restart framework. Finally, starting from a larger number small size but highly consistent clusters, we iteratively merge a pair of clusters with the maximum similarities until it reaches the predicted number of clusters. Extensive performance evaluation shows that the proposed single-cell clustering algorithm can yield the accurate single-cell clustering results and it can help deciphering the key messages underlying complex biological mechanisms.
| S-EPMC8623803 | biostudies-literature
Project description:(1) Background: Gene-expression data usually contain missing values (MVs). Numerous methods focused on how to estimate MVs have been proposed in the past few years. Recent studies show that those imputation algorithms made little difference in classification. Thus, some scholars believe that how to select the informative genes for downstream classification is more important than how to impute MVs. However, most feature-selection (FS) algorithms need beforehand imputation, and the impact of beforehand MV imputation on downstream FS performance is seldom considered. (2) Method: A modified chi-square test-based FS is introduced for gene-expression data. To deal with the challenge of a small sample size of gene-expression data, a heuristic method called recursive element aggregation is proposed in this study. Our approach can directly handle incomplete data without any imputation methods or missing-data assumptions. The most informative genes can be selected through a threshold. After that, the best-first search strategy is utilized to find optimal feature subsets for classification. (3) Results: We compare our method with several FS algorithms. Evaluation is performed on twelve original incomplete cancer gene-expression datasets. We demonstrate that MV imputation on an incomplete dataset impacts subsequent FS in terms of classification tasks. Through directly conducting FS on incomplete data, our method can avoid potential disturbances on subsequent FS procedures caused by MV imputation. An experiment on small, round blue cell tumor (SRBCT) dataset showed that our method found additional genes besides many common genes with the two compared existing methods.
| S-EPMC6274900 | biostudies-literature
Project description:MotivationAccurately clustering cell types from a mass of heterogeneous cells is a crucial first step for the analysis of single-cell RNA-seq (scRNA-Seq) data. Although several methods have been recently developed, they utilize different characteristics of data and yield varying results in terms of both the number of clusters and actual cluster assignments.ResultsHere, we present SAFE-clustering, single-cell aggregated (From Ensemble) clustering, a flexible, accurate and robust method for clustering scRNA-Seq data. SAFE-clustering takes as input, results from multiple clustering methods, to build one consensus solution. SAFE-clustering currently embeds four state-of-the-art methods, SC3, CIDR, Seurat and t-SNE + k-means; and ensembles solutions from these four methods using three hypergraph-based partitioning algorithms. Extensive assessment across 12 datasets with the number of clusters ranging from 3 to 14, and the number of single cells ranging from 49 to 32, 695 showcases the advantages of SAFE-clustering in terms of both cluster number (18.2-58.1% reduction in absolute deviation to the truth) and cluster assignment (on average 36.0% improvement, and up to 18.5% over the best of the four methods, measured by adjusted rand index). Moreover, SAFE-clustering is computationally efficient to accommodate large datasets, taking <10 min to process 28 733 cells.Availability and implementationSAFEclustering, including source codes and tutorial, is freely available at https://github.com/yycunc/SAFEclustering.Supplementary informationSupplementary data are available at Bioinformatics online.
| S-EPMC6477982 | biostudies-literature