We are trying to enrich for very rare viral transcripts in RNA extracted from a frozen piece of tumor. Due to the low expression level of these genes relative to cellular genes, we have a hard time finding an assay sensitive enough to consistently and quantitatively measure transcript levels.
Has anyone ever tried suppression subtractive hybridization for similar problems? It seems to be a complicated assay, so I am cautious about investing in the necessary materials unless I am confident it will help. Also, in the protocols I have looked at the final subtracted cDNAs, subcloned using the attached tags - has anyone ever tried to use a biotin tag or something similar to purify these cDNAs as opposed to cloning?
I have not used biotin tags. But SSH can be very tricky. I got a lot of genes from one try on a non-model organism but the expression results that led to those genes were wrong after I had to verify over 100 genes (using northerns). I even confirmed the expression results doing protein gels and again the SSH results were not to be trusted. If you are looking at this solely as a source of expression data, and your material is limited, I would be very careful.
If you have any idea about the viral sequence, you are better off with qPCR.
We SSH between tumor tissue or adjacent healthy tissue as the driver The resulting amplicons were used to amplify and label for microarray. Differential gene lists obtained in low abundance that could be validated between 70 and 80% used as the tumor tissue. Although it was a bit tricky, after six months of training we get good results with a subtraction efficiency confirmed by an average decrease of 50% in the expression levels of 100 housekeeping selected probes following the subtraction step in the arrays. And we reconfirmed the decreased expression (between 100 to 10.000 times) of some housekeeping genes by qRT-PCR.
When I started I did not have much hope in technique but finally it worked quite well, especially if you have many biological replicates.
Lisa why not you try for normalized cDNA libraries, as per my experience for finding of any rare transcripts normalized is better option than SSH libraries, bcoz SSH is little bit tricky.
We used SSH in combination with cDNA microarrays in several studies for analysis of differential gene expression. Generally, this technique is working very well. However, if you are not able to quantify your viral transcripts using qPCR (Did you also try Taqman assays?) I would suspect that you won't be successful with SSH applied to your specific problem. You will get many false positives in your subtracted library and won't get quantitative results.
You could try Illumina RNA-Seq. Sensitivity increases with the number of reads. From one HiSeq lane you can get more than 150 Mio reads.
- Badges
- Science topic
- Similar topics
- Biological Science
- Molecular Biology
- Molecular Biological Techniques