Challenges of Analyzing Single Cell Gene Expression Data

3,560 views

|

July 20, 2012

Kenneth Livak, Senior Scientific Fellow, Fluidigm

Abstract
Single cell researchers are currently using the Fluidigm BioMark System to measure gene expression levels for 10s to 100s of genes in 100s to 1000s of samples. Data from single eukaryotic cells show cell-to-cell variation in mRNA amounts that ranges from 10- to 1000-fold depending on the gene and type of cell. These results fit the stochastic model that eukaryotic transcripts are produced in short but intense bursts interspersed with intervals of inactivity. This noise inherent in single cell gene expression challenges conventional methods for obtaining and analyzing qPCR data. Factors such as replicates, limit of detection, normalization, data display, and univariate versus multivariate analysis need to be re-evaluated. Another important consideration is that data needs to be collected from a statistically significant number of single cells and for at least tens to hundreds of genes. By acknowledging and addressing the intrinsic noise, single-cell gene expression profiling can be used to gain biological insights that are just not possible when measuring average expression levels from hundreds or thousands of cells.

Drug DiscoveryGenomicsInformatics

Keep up to date with all your favourite videos and channels.

Get personalised notifications on new releases and channel content by subscribing to the LabTube eNewsletter.