Information provided to users by sequencing sites

[epistatic]

Hello all,

Data generation at our site is now a non-issue and I see a huge need to standardize the QC of sequencing data before handing off to customers. We have several different options for baseline QC and specialized QC information based upon library source: RNA-seq, Exome-seq, ChIP-seq, etc. I also believe it is hard to know the true quality without some initial alignment. This can get computationally expensive with the sample #s we are starting to have from multiplexed runs on the HiSeqs but I still think should be done.

Does anyone have recommendations on what information is essential? Are there any posted standards more savvy than Illumina's 80% of bases >Q30?

[NextGenSeq]

The HiSeq aligns the PhiX control to PhiX. If you spike in 1 to 2% that's enough to know if the flowcell or sequencing reagents were good.

If those look good and the library doesn't it's the library (which is almost always the case) or the library was misquantified.

[epistatic]

Yes, PhiX tells you the instrument is working...

I apologize if my question was vague. I am interested in knowing what information deliverables would be the most useful for users to receive with their sequencing data. Providing users with FASTQ files and the CASAVA demultiplexing summaries for the data (yield, read count, PF, %>/= Q30) seems reasonable but also not completely informative.

[kopi-o]

Mapping rate to the genome, % duplicates, fraction mRNA (for RNA-seq), % reads on target (exome / seq cap) seem useful.