I trying to see the relative expression of certain genes in sperm RNA. since sperm does not have 18s. I saw ACTB and GAPDH but was not consistant between individuals or assay. In this regard i would like to know suggestions. thank you
Bear in mind that reference genes are not supposed to be totally stable across all samples, unless your samples are all identically isolated, prepared and handled, and thus contain identical amounts of cDNA in each well. And if your samples are all identically isolated, prepared and handled, and contain identical amounts of cDNA....you don't really need to bother with a reference.
So "my reference genes vary" is not an unexpected result, the question is whether they vary in response to your treatment, or just vary because of the cDNA amount in each well. If you consistently see lower Ct values in a treatment group but not in a control, then this is probably an indication of a poor reference gene, whereas highly varied Ct values distributed between treatment and control groups is less damning.
My advice is always to try a fairly large panel of candidate genes (6-10) with a subset of your samples (including treated and control samples), and then run the results through geNorm, normfinder or bestkeeper (or better yet, all three). Include a gene you know (or suspect) to be treatment-sensitive as a benchmark for 'poor reference standard'.
Personally I've tended to use the genorm kits from primerdesign, since these provide primers for 10-12 genes selected to be 'mostly stable under most situations'. You use those on your samples, geNorm/Normfinder/bestkeeper those, and then pick the top two or three. They're reasonably expensive for what amounts to "a box of primers", but they do work very well.
sorry, i had to edit my answer as I misunderstood your question... Maureen's answer is very good.
Why does the inconsistency of housekeeping genes between samples and assays worry you? Are you also seeing inconsistencies in the replicates?
Bear in mind that reference genes are not supposed to be totally stable across all samples, unless your samples are all identically isolated, prepared and handled, and thus contain identical amounts of cDNA in each well. And if your samples are all identically isolated, prepared and handled, and contain identical amounts of cDNA....you don't really need to bother with a reference.
So "my reference genes vary" is not an unexpected result, the question is whether they vary in response to your treatment, or just vary because of the cDNA amount in each well. If you consistently see lower Ct values in a treatment group but not in a control, then this is probably an indication of a poor reference gene, whereas highly varied Ct values distributed between treatment and control groups is less damning.
My advice is always to try a fairly large panel of candidate genes (6-10) with a subset of your samples (including treated and control samples), and then run the results through geNorm, normfinder or bestkeeper (or better yet, all three). Include a gene you know (or suspect) to be treatment-sensitive as a benchmark for 'poor reference standard'.
Personally I've tended to use the genorm kits from primerdesign, since these provide primers for 10-12 genes selected to be 'mostly stable under most situations'. You use those on your samples, geNorm/Normfinder/bestkeeper those, and then pick the top two or three. They're reasonably expensive for what amounts to "a box of primers", but they do work very well.
thanks for valuable suggestions ....i got cleared my doubt.
I did a cross-comparison of 35 different genes representing several in vitro conditions comparing vaccinated and control animals (n=10 animals, n=>1200 overall reactions). Overall comparison was not possible without the use of 2-3 housekeeping genes, as sometimes one is not enough or consistent. I have noticed that no two preps are the same, even if the sample is consistent and assumed identical. Your major mRNA responders may seem similar with similar/identical preps, but the minor mRNAs might not be as consistent. There are too many variables if you do not have a standard, and it is difficult to draw valid conclusions.
If any of the (very good) suggestions above do not work, you could try looking for a housekeeping gene using the Genevestigator program. You have a free trial for some days, and it allows you to look for genes that are expressed near to your genes of interest, in the specific species and tissue... It sorted out my problem when looking for a good housekeeping gene for bone mRNA analysis.
- Badges
- Science topic