RNA RIN scores and RNA-Seq

[Nick2782]

Hello All,
This is my first post and I would like to start with asking a very basic question. What would you expect to see if you performed a RNA-seq experiment with heavily degraded RNA (RIN 2-8), with wide distribution? The library prep was done using a kit that does PolyA enrichment.
After spending thousands of dollars, would the data be useful enough to draw inferences?

[DonDolowy]

You would probably see a strong 3´ bias. We have been sequencing samples down to 5.5 and it works out fine, though we do see the 3´bias is stronger compared to samples having RIN of 7-8.

But some times when working with primary cells, it is not always possible to get super high quality RNA from precious animals, so we have to use what we can get.

[Nick2782]

Hi DonDolowy,
Thanks for responding. How does that affect the cuffdiff/link pipeline? Since, you have a wide distribution of samples with differing RNA integrity, will this affect normalization in anyway? If yes, how?

[DonDolowy]

Since I always have at least triplicates, I doubt that it will have a huge impact on downstream applications. With that said, we are only interested in RNA-seq as better (and still cheap) alternative to microarray, so we are not looking at splicing events etc. Here I could imagine a 3´bias would have greater impact on the results.

[TonyBrooks]

Why not switch the mRNA enrichment step for something that doesn't rely on polyA (RiboZero)? The downside is you need more RNA than for most polyA methods.