At the Genomes Unzipped blog, Caroline Wright suggests that the UK government may need to reevaluate its plans to sequence 100,000 patients' genomes for clinical use.
Last December, the UK government announced that it had earmarked £100 million, or about $160.9 million, for the initiative, which would be led by the National Health Service. The project would train British geneticists and the healthcare community in using genome-based medicine and also support research and therapy development for cancer and rare, inherited diseases.
In her piece, Wright argues that the NHS funds can't cover current costs of whole genome sequencing for each patient let alone the "substantial costs of labor, informatics, and training."
She points out that genomic data interpretation is "still in its infancy" with as much as 98 percent of the genome still shrouded in mystery. She also notes that "the reduced analytical accuracy of current next-generation sequencing machines for whole genomes versus targeted multigene or exome sequencing" poses a problem "as the depth of coverage may be substantially lower in coding regions."
"Opting for genomes instead of gene-targeted approaches may actually reduce the diagnostic yield!" she writes. "This is major problem for rare disease sequencing, as the ability to accurately call novel variants in key pathogenic genes may be sacrificed, thus preventing a diagnosis. It is even more of a problem for cancer genome sequencing due to sample heterogeneity."
Finally, Wright adds that researchers need to ask what the most appropriate test is to use for each population. Here, she suggests that alternatives such as exome, targeted gene, and microbial sequencing are better suited for diagnostic tasks — and are cheaper then whole-genome sequencing.