Assessing characteristics of RNA amplification methods for single cell RNA sequencing

Abstract

BackgroundRecently, measurement of RNA at single cell resolution has yielded surprising insights. Methods for single-cell RNA sequencing (scRNA-seq) have received considerable attention, but the broad reliability of single cell methods and the factors governing their performance are still poorly known.ResultsHere, we conducted a large-scale control experiment to assess the transfer function of three scRNA-seq methods and factors modulating the function. All three methods detected greater than 70% of the expected number of genes and had a 50% probability of detecting genes with abundance greater than 2 to 4 molecules. Despite the small number of molecules, sequencing depth significantly affected gene detection. While biases in detection and quantification were qualitatively similar across methods, the degree of bias differed, consistent with differences in molecular protocol. Measurement reliability increased with expression level for all methods and we conservatively estimate measurements to be quantitative at an expression level greater than ~5–10 molecules.ConclusionsBased on these extensive control studies, we propose that RNA-seq of single cells has come of age, yielding quantitative biological information.Electronic supplementary materialThe online version of this article (doi:10.1186/s12864-016-3300-3) contains supplementary material, which is available to authorized users.

Systematic comparative analysis of single cell RNA-sequencing methods

“…There is thus an urgent need to provide comparison and benchmarking information to help guide users make informed choices based on each method's capabilities and limitations, compare newly proposed methods to existing ones, and identify shared weaknesses as targets for experimental improvement. Prior comparisons of scRNA-seq methods [16][17][18][19][20][21] , though useful, have several shortcomings. Many are outdated given the fast-paced field, or are incomplete, inapplicable (e.g., not actually performed with single cells), or insufficiently controlled (e.g., performed using different samples for comparisons); others limit their assessment to basic technical factors, but do not assess the key benchmark of ability to recover meaningful biological information, such as population heterogeneity and structure.…”

The successes and future prospects of the linear antisense RNA amplification methodology

“…The study of cancer has benefited preferential amplification of certain RNA species (i.e., it shows no sequence preference) nor creates significant differences in the ratios of specific aRNAs in the amplified RNA population compared with those of the initial host cell mRNA abundances 3,9 . It should be noted that if there were bias in amplification, then the bias would be linearly amplified rather than exponentially as in PCR amplification. Furthermore, aRNA has been shown to provide an accurate and precise amplification product [9][10][11][12] . aRNA has been used extensively to generate probes for northern and Southern blotting, in tandem with PCR 13,14 and microarrays 4,12,15,16 , and for sequencing of tissue-isolated RNA 7,10,17,18 .…”

scBFA: modeling detection patterns to mitigate technical noise in large-scale single cell genomics data

“…1), due to both choice of technology and cost trade-off. We reasoned that when the number of unique molecules drops too low, the signal-noise ratio of the data may be too low to make gene quantification informative 7 , and therefore downstream analyses should be adapted to primarily consider gene detection patterns only.…”