Exome sequencing DNA from FFPE tissue

[JER]

Does anyone know how well Illumina Hiseq can deal with DNA from FFPE tissue? We have a method to deal with the Cytosines changing to Uracil but is it feasible to use DNA from paraffin for exome sequencing if it hasn't degraded too much? We'd be using the Nextera rapid capture Enrichment Kit ideally.

Thanks

[nucacidhunter]

Does anyone know how well Illumina Hiseq can deal with DNA from FFPE tissue?

HiSeq and other Illumina sequencing systems sequences prepped libraries, so the origin of DNA does not make much difference on sequencer output.

We have a method to deal with the Cytosines changing to Uracil but is it feasible to use DNA from paraffin for exome sequencing if it hasn't degraded too much? We'd be using the Nextera rapid capture Enrichment Kit ideally.

The issue would be preparing a suitable input Nextera library (300 bp peak) for hybridisation step from FFPE tissue DNA as Nextera requires high molecular weight input DNA. A high degree of customisation may be necessary to obtain such library. You might be better off with kits that use short DNA fragments from shearing for library prep step.

[Bukowski]

We've used SureSelect capture on FFPE samples regularly. Works just fine, but basically there is a trade off, if the DNA is highly fragmented you will run into trouble. You can somewhat ameliorate this by throwing more DNA into the hyb, but the bigger the fragments in your sample, the better it will perform.

We've done exome sequencing on 25 year old FFPE samples. I won't say the data was great, but it worked.

[mellyflow]

Hi Bukowski, you say with highly-fragmented DNA you will run into trouble. Can you elaborate on this? What sorts of troublesome things have you seen? Increased dupes and varying insert sizes?

[dannydesloover]

FFPE - qPCR to check X-linking.

Even if your FFPE DNA is not highly fragmented, you should run some qPCR to check for residual cross-linking via the FFPE process. If doing sureselect or nextera, you'll have a severely less diverse library from FFPE compared to the same amount of non-FFPE DNA.