A step-by-step workflow for low-level analysis of single-cell RNA-seq data with Bioconductor

Lun, Aaron T. L.; McCarthy, Davis J.; Marioni, John C.

doi:10.12688/f1000research.9501.2

Cited by 1,222 publications

(1,312 citation statements)

References 49 publications

Supporting

Mentioning

1,192

Contrasting

Unclassified

Order By: Relevance

“…This choice depends on whether total RNA content in each cell is of interest (Lun et al 2016b). Spike-in normalization will preserve changes in total RNA content between cells, whereas non-DE methods will treat such changes as bias (as a majority of genes are affected) and remove them.…”

Section: Discussionmentioning

confidence: 99%

“…For HVG detection, we used methods based on the coefficient of variation (Brennecke et al 2013) or the variance of log-expression values (Lun et al 2016b). Again, only minor changes were observed in most data sets (Fig.…”

Section: Wwwgenomeorgmentioning

confidence: 99%

“…The first approach is based on the method of Brennecke et al (2013), in which the squared coefficient of variation for each gene is tested for a significant increase above technical noise. The second approach is based on the variance of the log-normalized expression values (Lun et al 2016b), which provides some more robustness against outlier expression patterns. Each method was applied on the original and simulated data, and a set of significant HVGs was detected at a FDR of 5%.…”

Section: Evaluating the Robustness Of Hvg Detectionmentioning

confidence: 99%

“…One particular advantage of this strategy is that it does not make any assumptions about the endogenous expression profile, unlike the non-DE approach described above. This means that spike-in normalization can be applied in situations in which large-scale changes in expression (e.g., related to changes in total RNA content, or involving highly heterogeneous populations containing many cell types) are expected and of interest (Lun et al 2016b;Nestorowa et al 2016).…”

mentioning

confidence: 99%

See 3 more Smart Citations

Assessing the reliability of spike-in normalization for analyses of single-cell RNA sequencing data

et al. 2017

Self Cite

View full text Add to dashboard Cite

By profiling the transcriptomes of individual cells, single-cell RNA sequencing provides unparalleled resolution to study cellular heterogeneity. However, this comes at the cost of high technical noise, including cell-specific biases in capture efficiency and library generation. One strategy for removing these biases is to add a constant amount of spike-in RNA to each cell and to scale the observed expression values so that the coverage of spike-in transcripts is constant across cells. This approach has previously been criticized as its accuracy depends on the precise addition of spike-in RNA to each sample. Here, we perform mixture experiments using two different sets of spike-in RNA to quantify the variance in the amount of spike-in RNA added to each well in a plate-based protocol. We also obtain an upper bound on the variance due to differences in behavior between the two spike-in sets. We demonstrate that both factors are small contributors to the total technical variance and have only minor effects on downstream analyses, such as detection of highly variable genes and clustering. Our results suggest that scaling normalization using spike-in transcripts is reliable enough for routine use in single-cell RNA sequencing data analyses.

show abstract

Section: Discussionmentioning

confidence: 99%

Section: Wwwgenomeorgmentioning

confidence: 99%

Section: Evaluating the Robustness Of Hvg Detectionmentioning

confidence: 99%

mentioning

confidence: 99%

See 2 more Smart Citations

Assessing the reliability of spike-in normalization for analyses of single-cell RNA sequencing data

et al. 2017

Self Cite

View full text Add to dashboard Cite

show abstract

“…To demonstrate the utility of beachmat for faciliting analyses of large data sets, we 301 converted several functions in the scater [17] and scran packges [14] to use the beachmat 302 API in their C++ code. We applied these functions to the 1 million neuron data set 303 from 10X Genomics (see Methods).…”

mentioning

confidence: 99%

beachmat: a Bioconductor C++ API for accessing single-cell genomics data from a variety of R matrix types

Lun

Pagès

Smith³

2017

Preprint

View full text Add to dashboard Cite

Recent advances in single-cell RNA sequencing have dramatically increased the number of cells that can be profiled in a single experiment. This provides unparalleled resolution to study cellular heterogeneity within biological processes such as differentiation. However, the explosion of data that are generated from such experiments poses a challenge to the existing computational infrastructure for statistical data analysis. In particular, large matrices holding expression values for each gene in each cell require sparse or file-backed representations for manipulation with the popular R programming language. Here, we describe a C++ interface named beachmat, which enables agnostic data access from various matrix representations. This allows package developers to write efficient C++ code that is interoperable with simple, sparse and HDF5-backed matrices, amongst others. We perform simulations to examine the performance of beachmat on each matrix representation, and we demonstrate how beachmat can be incorporated into the code of other packages to drive analyses of a very large single-cell data set.

show abstract

Single‐cell RNA‐seq reveals the invasive trajectory and molecular cascades underlying glioblastoma progression

Pang

et al. 2019

Molecular Oncology

View full text Add to dashboard Cite

Glioblastoma (GBM) is the most common and aggressive primary brain tumor, in which GBM stem cells (GSCs) were identified to contribute to aggressive phenotypes and poor prognosis. Yet, how GSCs progress to invasive cells remains largely unexplored. Here, we revealed the cell subpopulations with distinct functional status and the existence of cells with high invasive potential within heterogeneous primary GBM tumors. We reconstructed a branched trajectory by pseudotemporal ordering of single tumor cells, in which the root showed GSC‐like phenotype while the end displayed high invasive activity. Thus, we further determined a path along which GSCs gradually transformed to invasive cells, called the ‘stem‐to‐invasion path’. Along this path, cells showed incremental expression of GBM invasion‐associated signatures and diminishing expression of GBM stem cell markers. These findings were validated in an independent single‐cell data set of GBM. Through analyzing the molecular cascades underlying the path, we identify crucial factors controlling the attainment of invasive potential of tumor cells, including transcription factors and long noncoding RNAs. Our work provides novel insights into GBM progression, especially the attainment of invasive potential in primary tumor cells, and supports the cancer stem cell model, with valuable implications for GBM therapy.

show abstract

A step-by-step workflow for low-level analysis of single-cell RNA-seq data with Bioconductor

Cited by 1,222 publications

References 49 publications

Assessing the reliability of spike-in normalization for analyses of single-cell RNA sequencing data

Assessing the reliability of spike-in normalization for analyses of single-cell RNA sequencing data

beachmat: a Bioconductor C++ API for accessing single-cell genomics data from a variety of R matrix types

Single‐cell RNA‐seq reveals the invasive trajectory and molecular cascades underlying glioblastoma progression

Contact Info

Product

Resources

About