Profiling of cell populations
Understanding the heterogeneity within a stem cell population remains a major impediment to the development of clinically effective cell-based therapies. Gene expression patterns exhibited by individual cells are a crucial component of this heterogeneity, yet transcriptional events within a single cell are inherently stochastic and can produce tremendous variability, even among genetically identical cells. It remains unclear how mammalian cellular systems overcome this intrinsic noisiness of gene expression to produce consequential variations in function. To address these questions, we utilized a novel single cell analysis method to characterize transcriptional programs across hundreds of individual murine long-term hematopoietic stem cells (LT-SCs). We demonstrate that multiple subpopulations exist within this putatively homogeneous stem cell population, defined by nonrandom patterns that are distinguishable from noise and can predict functional properties of these cells. This represents a powerful new tool to elucidate the relationship between transcriptional and phenotypic variation within a cell population.
1. A method for comparing gene expression, comprising:
isolating a plurality of single cells in a first cell population and a plurality of single cells in a second cell population;
choosing at least two individual genes to analyze an expression of the genes within the single cells of each of the first cell population and second cell population;
for each of the gene expressions of the first and second cell populations, determining a transcriptional identity of a single cell of the plurality of single cells based on individual gene expression of the at least two genes within the single cell;
for each of the first and second cell populations, using the transcriptional identity of each of the single cells to determine a heterogeneity profile; and
using a logic circuit to perform statistical analysis to compare the heterogeneity profile of the first cell population to the heterogeneity profile of the second cell population, the comparison determining the similarity between the first and second cell populations.
2. A method of comparing gene expression, comprising:
isolating a plurality of single cells in a cell population;
choosing at least two genes to analyze an expression of each gene within each of the single cells;
determining a transcription identity of a single cell of the plurality of single cells based on gene expression of the at least two genes;
using a logic circuit to perform a statistical cluster analysis on the transcriptional identity of each of the single cells of the plurality of single cells to determine whether the cell population includes one or more subpopulations;
determining a heterogeneity profile of the one or more subpopulations; and
using a second logic circuit to perform a divergence-based statistical analysis to compare the heterogeneity profiles of the one or more subpopulations to determine whether the one or more subpopulations are distinct subpopulations.
3. A method for comparing gene expression, comprising:
isolating a plurality of single cells in a first cell population;
choosing at least two genes to analyze an expression of each gene within each of the single cells;
determining a transcription identity of a single cell of the plurality of single cells based on gene expression of the at least two genes;
using the transcriptional identity of the single cells to determine whether the first cell population includes one or more subpopulations;
using a logic circuit to perform cluster analysis based on the transcriptional identity of each of the single cells of the plurality of single cells to determine whether the first cell population includes one or more subpopulations;
determining a heterogeneity profile of the one or more subpopulations;
using the logic circuit to perform a divergence-based statistical analysis to compare the heterogeneity profiles of the one or more subpopulations to determine whether the one or more subpopulations are distinct subpopulations;
determining a total heterogeneity profile for the first cell population; and
using the logic circuit to perform statistical analysis to compare the total heterogeneity profile of the first cell population to a previously determined heterogeneity profile of a second cell population.
4. A method for comparing gene expression of a first cell population, comprising:
using a logic circuit to:
determine a transcriptional identity for each individual cell in the first cell population, the transcriptional identity based on a level of gene expression of at least two genes;
perform cluster analysis, based on the transcriptional identity of each individual cell, to determine whether the first cell population includes one or more subpopulations;
determine a heterogeneity profile of the one or more subpopulations and compare the heterogeneity profiles of the one or more subpopulations to determine whether the one or more subpopulations are distinct subpopulations;
determine a total heterogeneity profile for the first cell population; and perform a divergence-based statistical analysis to compare the total heterogeneity profile of the first cell population to a previously determined heterogeneity profile of a second cell population.
5. The method of claim 1 , wherein using the logic circuit to perform statistical analysis includes utilizing a divergence analysis to determine an amount of divergence between probability distributions of the heterogeneity profile of the first cell population to the heterogeneity profile of the second cell population.
6. The method of claim 5 , wherein utilizing the divergence analysis includes determining whether the heterogeneity profile of the first cell population to the heterogeneity profile of the second cell population are highly divergent thereby indicating distinctness of the first cell population and the second cell population.
7. The method of claim 1 , wherein using the logic circuit to perform statistical analysis includes utilizing a Kullback-Leibler divergence analysis to determine an amount of divergence between probability distributions of the heterogeneity profile of the first cell population and the heterogeneity profile of the second cell population.
8. The method of claim 1 , further including utilizing a phi-entropy based divergence analysis to determine the transcriptional identity of each of the single cells to determine the heterogeneity profile.
9. The method of claim 1 , further including utilizing the logic circuit to compare a heterogeneity profile of at least one of the first cell population and the second cell population to a previously determined heterogeneity profile for the population to determine whether at least one of the first cell population and the second cell population has changed over time.
10. The method of claim 2 , wherein determining the heterogeneity profile of the one or more subpopulations includes utilizing a phi-entropy based divergence analysis to determine the transcriptional identity of each of the single cells to determine the heterogeneity profile.
11. The method of claim 2 , wherein utilizing the divergence-based statistical analysis includes determining an amount of divergence between probability distributions of a heterogeneity profile of a first cell population and a heterogeneity profile of a second cell population.
12. The method of claim 11 , wherein utilizing the divergence-based statistical analysis includes determining whether the heterogeneity profile of the first cell population and the heterogeneity profile of the second cell population are highly divergent, thereby indicating distinctness of the first cell population and the second cell population.
13. The method of claim 3 , wherein using the logic circuit to perform the divergence-based statistical analysis includes determining an amount of divergence between probability distributions of the heterogeneity profile of the first cell population and the heterogeneity profile of the second cell population.
14. The method of claim 13 , wherein utilizing the divergence-based statistical analysis includes determining whether the heterogeneity profile of the first cell population and the heterogeneity profile of the second cell population are highly divergent thereby indicating distinctness of the first cell population and the second cell population.
15. The method of claim 4 , wherein using the logic circuit to perform the divergence-based statistical analysis includes utilizing a divergence analysis to determine an amount of divergence between probability distributions of the heterogeneity profile of a first cell population to the heterogeneity profile of a second cell population.