Expression patterns from rna-seq data

[ngs06]

Hi,
I would like to analyse single cell rna-seq data. The dataset contains 21000 genes and 98 samples. I would like to look at the samples and genes behaviours, like which group of samples have similar expression and also which group of genes have similar expression levels. I thought making the gene-wise and samplewise clustering and viewing them in a heatmap could give better picture to start with initial data analysis.
But the dimensions of the dataset are too huge, I am not able to plot the heatmaps in R.
Do anyone have suggestions regarding how to start with initial data analysis to look for the expression patterns in dataset.


MAny thanks!

[mikep]

There's alot of noise in single cell RNA-seq. You are unlikely to get anything decent if you use the entire dataset, I'd recommend at least a count cut off, or as alot of people do use the genes associated with the first 2 principal components.

Anyway, for viz try TM4

http://www.tm4.org/

[ngs06]

There's alot of noise in single cell RNA-seq. You are unlikely to get anything decent if you use the entire dataset, I'd recommend at least a count cut off, or as alot of people do use the genes associated with the first 2 principal components.

Anyway, for viz try TM4
http://www.tm4.org/

Thanks for your suggestion. I do PCA, which seems to be much followed method for data reduction.