Exomes from a spot of blood

Michael Snyder has a great paper published in AJRCCM: Exome Sequencing of Neonatal Blood Spots Identifies Genes Implicated in Bronchopulmonary Dysplasia (BPD). BPD is a lung disease of premature babies which appears to have a strong genetic component and was investigated using exome sequencing. The neat twist was they used the blood spot taken from a heel-prick to extract DNA for exome library prep. In the paper they report finding over 250 rare nonsynonymous mutations after comparing 50 affected and unaffected twin pairs. Many of these were enriched for, and up-regulated in, a murine model of BPD.

 

Image from : http://en.wikipedia.org/wiki/Robert_Guthrie

 

Guthrie test blood spots: Blood spots from a heel prick are very common with almost all kids born in the UK having one taken, but this paper is one of the first to report their use in next-generation sequencing studies (they have previously been used in gene expression and methylation studies). However the generation of high-quality exome data suggest that these could become a more routine sample collection device. They used a DNA extraction protocol published by a group from Stanford in 2013, the 96 well format protocol generated around 100ng from a 2 mm punch. I'm sure we'll see lots more studies following this most recent publication.

Improving the reliability of genetic testing: With projects like the UK's 100,000 genomes there is an increasing need to verify the provenance of a sample. Simply genotyping the blood used to make a WGS library is unlikely to be good enough. If a mix up happens in the blood bank or DNA extraction facility then WGS and genotyping will match, but the wrong patient might get treated.

 

Using an orthogonal method is common in validation of GX or GT studies. It would be relatively simple to apply this orthogonality to genotyping of patients recruited into genomic studies. If a blood spot can be collected ahead of surgery from a finger prick, this could be genotyped and compared to data from a WGS or WES. And there is little reason patient ID SNPs can't be built into all clinical NGS tests to report not just the results but also confirm patient ID.