Tuesday, July 17, 2012

Information about RNAseq

In the Genomics Core Lab we define two different types of RNAseq
  • Transcriptome Analysis
  •  Transcriptome Discovery

For a Transcriptome analysis, we will generate an amount of data similar to an microarray data, between 20 and 30 million reads on a Paired-end Hiseq run (Paired end means that each fragment will be sequenced in both direction, 50bp from the reverse side, and 50bp from the forward side). The dynamic range will be similar as the one observed on an Affymetrix or an Illumina Gene Expression array.

Cost of transcriptome analysis : ~ $500/sample  (from total RNA to raw data)

In a Transcriptome discovery, you will get exon level information, you should be able to monitor expression of rare transcript and hopefully you should be able to identify known as well as new fusion events. We will generate between 80 and 100 million reads per sample, on a Hiseq Paired end 75bp/75bp run.

Cost of transcriptome discovery: ~$900/sample (from total RNA to raw data)

Last update

20-30 M (6 samples/lane)
80-100 M( 2samples/lane)

Some frequently asked questions...

Amount of material needed?   
The standard protocol calls for 500ng of total RNA.
Less is doable, but would have to be treated on a case by case based on the amount.

What is the lowest amount of RNA I can submit?  
 We have processed samples that are below the threshold of detection of ribogreen or Agilent BioAnalyzer, coming from as low as 7 cells. We are actively working on a valid single cell protocol.

If your samples fall in this category, keep in mind that the amplifications always introduce bias in the data, no matter what method of  amplification we will use –PCR, IVT, Rolling amplification-. Another consequence is that instead of 70 to 80% of the data being informative, probably only 30 to 40% will be informative. On the bright side of things, 30 to 40% is better than 0% (which is what you would get if you don’t try!).

Will my samples be barcoded? 

Yes. For transcriptome analysis, we will put  up to 5 samples per hiseq 2000 lane. For transcriptome discovery, we will sequence 2 samples per lane. This difference is what is dictating the difference in cost.
Quality of RNA?
A RIN (RNA Integrity Number) of 8 is ideal, however, as for a microarray experiment, what is essential is homogeneity among a sample set .

Biological replicates
Absolutely YES. You will need some of these if you want to reduce the number of false positive targets.

Turnaround time? 
3 weeks at the minimum.  A typical break down of the time is described below:
Week 1: RNA QC, OD, library making
Week 2: preparation of the flow cell for sequencing (Monday)
Week 2: Tuesday : start the run (12 days)
Week 3; Sunday , end of the run
Week 4: Monday : data archiving, mapping, and quality control.
Week 4: Wednesday. Data ready for secondary analysis.

Additional cost? 
If you do not have the expertise to transform raw data into a gene expression matrix or fold change, you will need the assistance of the bioinformatics core.

Can a transcriptome discovery run also be used for transcriptome quantification?
Absolutely yes

Can we go back and get higher read depth with a library after a run, either for the same read length or the other read length?
Absolutely yes


No comments:

Post a Comment