Sampling strategies to capture single-cell heterogeneity

Abstract

Advances in single-cell technologies have highlighted the prevalence and biological significance of cellular heterogeneity. A critical question is how to design experiments that faithfully capture the true range of heterogeneity from samples of cellular populations. Here, we develop a data-driven approach, illustrated in the context of image data, that estimates sampling depth required for prospective investigations of single-cell heterogeneity from an existing collection of samples.

Collective gradient sensing and chemotaxis: modeling and recent developments

“…Understanding to what extent these properties will alter the ability of cell clusters to follow gradients, and whether this effect will dominate those potential biases identified in [92], is a natural next step. The critical nature of understanding CCV in motility and signaling also highlights the need for new computational tools to extract these parameters from experimental observations [142]. …”

Unmasking tumor heterogeneity and clonal evolution by single-cell analysis

“…However, DNA methylation patterns are heterogeneous and occurs in both large and poorly defined genomic regions [20] , posing a challenge in using CpG methylation as a biomarker. In a recent study by Rajaram et al [38] , a data-driven framework based on single-cell analysis has been reported that provides an estimate of the depth of sampling that may be minimally required to cover the full range of phenotypic heterogeneity for accurate biomarker discovery. Based on the analysis of 215 single-cell features, three replicates were sufficient to capture the heterogeneity for many features if they were defined by clear biomarkers without background noise [38] .…”

“…In a recent study by Rajaram et al [38] , a data-driven framework based on single-cell analysis has been reported that provides an estimate of the depth of sampling that may be minimally required to cover the full range of phenotypic heterogeneity for accurate biomarker discovery. Based on the analysis of 215 single-cell features, three replicates were sufficient to capture the heterogeneity for many features if they were defined by clear biomarkers without background noise [38] . For example, nuclear staining (the number of nuclei staining by DAPI: an easily detectable feature) requires 1-2 cores to capture the heterogeneity in > 90% of the patients, while 10 cores or more are needed to assess the heterogeneity of YAP transcription factor expression (a sparsely detectable feature) [38] .…”

“…Based on the analysis of 215 single-cell features, three replicates were sufficient to capture the heterogeneity for many features if they were defined by clear biomarkers without background noise [38] . For example, nuclear staining (the number of nuclei staining by DAPI: an easily detectable feature) requires 1-2 cores to capture the heterogeneity in > 90% of the patients, while 10 cores or more are needed to assess the heterogeneity of YAP transcription factor expression (a sparsely detectable feature) [38] . Therefore, both the complexity of the feature and the biomarkers that define the feature determine the number of samples required for studying heterogeneity [38] .…”

“…For example, nuclear staining (the number of nuclei staining by DAPI: an easily detectable feature) requires 1-2 cores to capture the heterogeneity in > 90% of the patients, while 10 cores or more are needed to assess the heterogeneity of YAP transcription factor expression (a sparsely detectable feature) [38] . Therefore, both the complexity of the feature and the biomarkers that define the feature determine the number of samples required for studying heterogeneity [38] . Once the tissue is processed, multiple techniques to isolate single cells can be implemented [ Figure 4].…”