Usually we do differential analysis between two states of interest (control vs. disease) or two cell types (MCF7 vs. HL60) to find gene which are up and down regulated 'differentially'.
What we should do when we are interested in knowing genes which are expressed (or not-expressed) in only 1 cell line of interest. How to define the threshold score for gene being expressed and not expressed?
Use-case: . Suppose I have a 'transcription profiling by array assay of GM12878'. I want to find out what genes are expressed/not-expressed in it, how I can find out this?
look at this link, you will find your answer.
https://www.biostars.org/p/166952/
It may be difficult to say that a gene is not (i.e. never) expressed, because some genes are expressed depending on context (in rich media, serum-free media, temperature, growth factors, number of passages etc etc etc). Let's assume you are talking about happily dividing cell culture in a similar state of confluency (i.e. not to sparse, not too crowded).
If you are working with microarray data, you could take the bottom 5-10% fluorescence (scaled to some external reference RNA, perhaps) as "low expression" and the maybe top 10-20% as "high expression" in each cell type. This is assuming you have put in place controls such as having the same input concentration of RNA for both experiments, a good reference for scaling the fluorescence values etc.
For a more macro view, see if you can plot the scaled expression from both cells and simply make a scatterplot to see where the outliers are.
It would be good to have several biological replicates (cells harvested from separate cell cultures, processed and hybridized on separate chips). Affymetrix expression microarrays allow to use MAS5.0 processing that detects Present/Marginal/Absent calls through statistical significance of the differences between perfect match and mismatch probesets. You can take all probesets which show A (absent) call for at least 90% of specimens and calculate 90-percentile of their value, which can be subsequently used as the threshold signal intensity to distinguish "expressed" genes from "non-expressed". This procedure has been reported by Saviozzi et al. (Methods Mol Med. 2004;94:67-90; PMID: 14959823).
- Badges
- Science topic
- Similar topics
- Biological Science
- Genetics
- Gene Expression