3.2 billion base pairs. And a report that fits on 20 pages.

In whole genome sequencing, or WGS, your complete genome is read: around 3.2 billion base pairs, coding and non-coding regions, plus the mitochondrial DNA. Technically this runs on next-generation sequencing, where the DNA is fragmented, read millions of times in parallel and then assembled against a reference genome. For clinical interpretation a mean coverage of around 30-fold is the usual standard, meaning every position is read thirty times on average. The sequencing itself is the easier part today. The work lies in the interpretation.

Four classes of finding are added by full sequencing. First, structural variants, meaning larger rearrangements such as deletions, duplications or inversions that sit between exons and therefore remain invisible in an exome. Second, variants in regulatory regions that change the activity of a gene without touching the protein itself. Third, the option of reanalysing the findings later without a new sample, because the raw data covers the whole genome. Fourth, pharmacogenetic interpretation in its full breadth, because several clinically important pharmacogenes carry variants in intronic regions, meaning the stretches between the protein-coding parts of a gene. How much of that becomes genuinely actionable in a preventive examination is a separate question.

Coverage states how often a position in the genome was read on average. At 30-fold coverage every base is captured thirty times on average. This is necessary because individual reads make errors: only repeated reading separates a genuine sequence difference from a read error. For detecting heterozygous variants, meaning variants on only one of the two chromosomes, 30-fold is the clinical standard. Some questions, such as mosaic findings or somatic variants in tumour tissue, require considerably more. Lower coverage reduces the price but increases the number of positions on which no reliable statement is possible.

Not 3.2 billion base pairs, but a filtered, physician-curated selection. At YEARS the report has four parts. First, variants in more than 170 clinically curated risk genes, classified from disease-causing to benign under the ACMG criteria, the international standard for this assessment. Second, the pharmacogenetic section with concrete dosing guidance for the drug classes affected. Third, carrier status for recessive conditions, relevant for family planning. Fourth, clearly separated and labelled as research-adjacent, polygenic risk scores. Variants of uncertain significance are named as such and not reinterpreted into a recommendation. You receive your raw data in addition on request.

For most questions in a preventive examination the exome delivers the clinically interpretable findings. Full sequencing pays off mainly where structural variants are in question, where complete pharmacogenetic interpretation is wanted, or where a targeted prior investigation returned no finding. At YEARS the Ultimate programme runs both, because the question cannot be reopened after the blood draw and the marginal cost of the additional interpretation is lower than a second sequencing months later. The details of the comparison are on our page on WGS and WES.

From blood draw to a discussed report takes four to six weeks at YEARS. Most of that time goes not into sequencing, which takes a few days, but into bioinformatic analysis and physician curation. A raw data file from a 30-fold whole genome sequencing run comes to roughly 90 to 120 gigabytes in uncompressed FASTQ depending on format, considerably less as compressed CRAM. Storage is within the EU. On request you receive the raw data in a common format, so a physician continuing your care can reanalyse it without you having to sequence again.