Although there is no absolute threshold under which a gene must be considered as "not expressed" for Affymetrix mRNA expression arrays the vlaue of 5 (log scale) is commonly considered as threshold for expression levels that likely have biological meaning. Is there a similar conventional threshold for Affy miRNA arrays?
The threshold serves to exclude "non expressed" genes from further analysis.
The issue with general rule-of-thumb background thresholds for miRNAs--leaving technical variability aside--is related to the sequence constraints of nucleic acid bits that are only 18-23 nt in length and have widely varying GC content. For mRNAs, many probes can be designed; not so for miRNAs. If you look up this paper https://www.researchgate.net/publication/236940377 and go to the section entitled "Normexp background correction with quantile normalization," you will find a good overview of the problem and how Affy tries to address it...plus perhaps some additional direction.
Article The use of miRNA microarrays for the analysis of cancer samp...
- Badges
- Science topic
- Similar topics
- Biological Science
- Biochemistry
- Biomolecules
- Nucleic Acids
- RNA
- Small RNA
- microRNA