This is a test version of Biostars. For the public version, visit https://www.biostars.org.
How To Minimize The Bias Caused By Rna Amplification In Rna-Seq?

Hi, All,

In the RNA sequencing on a small amount of tissue material, RNA amplification is inevitable step. Since RNA amplification is variably dependent on GC content, it is not absolutely linear. There is highly possibility that transcriptome data cannot reflect true transcript distribution in cell. Are bioinformatic normalization methods available to minimize such variabilities and bias?

One poster abstract shows that some amplification kits are better than others. http://www.ncbi.nlm.nih.gov/pmc/articles/PMC3635348/ Why will there be such difference?

Thanks

rnaseq rna
• 5,522 views • link
updated by Ryan Dale Ryan Dale • written by sckinta sckinta

2 answers

I realize you're asking about the bioinformatics side, but Library preparation methods for next-generation sequencing: Tone down the bias (Exp Cell Res 2014) reviews some ideas for avoiding bias in the library prep.

• 346 views • link
written by Ryan Dale Ryan Dale

Generally you'll find these sorts of things to be kit or batch dependent, so they end up not affecting most analyses. When that's not the case, you can use things like CQN to help correct for at least GC and length bias.

• 103 views • link
written by Devon Ryan Devon Ryan

I agree - RNA-Seq coverage is typically very uneven (even when working with high quantities of RNA). I think the only way to really do a good job of correcting this is through wet lab sample preparation.

Unfortunately, I haven't seen any data from any such preparations first-hand, but I have heard of developments like this. Hopefully, some of these links can be helpful:

Illumina FAQ (I think TruSeq helps provide more even coverage?): http://www.illumina.com/applications/sequencing/rna.ilmn

http://www.pnas.org/content/111/5/1891.full

• 95 views • link
written by Charles Warden Charles Warden

Yeah, you can only rescue things to a certain extent with algorithms.

• 0 views • link
written by Devon Ryan Devon Ryan

Log in to answer this question.

More posts like this