3 ms·

There are reasons to sequence an individual multiple times: tracking their microbiome, tracking the genetic evolution of their tumor, looking at gene expression

There are reasons to sequence an individual multiple times: tracking their microbiome, tracking the genetic evolution of their tumor, looking at gene expression, etc. There's not much value in sequencing someone's genome over and over again though. Additionally, there's a lot more to sequence besides just human. Illumina does genotyping for agriculture, they supply 23andme and Ancestry with their SNP chips, sequencing for forensic applications, etc.

killjoywashere 8y ago
> There's not much value in sequencing someone's genome over and over again though. There's plenty of value to each custodian that gets their own copy though, just like every cloud company wants their own copy of your name, email, and physical address.
- adenadel 8y ago
  Sure, but IMO it would be silly for another player to sequence you again rather than negotiate access to the data.
  - killjoywashere 8y ago
    Ah, I see you're not in medicine. Physicians will repeat studies of arbitrary cost at the drop of a hat.
    - adenadel 8y ago
      But in those cases, they are actually looking for changes. For germline WGS there are no differences.
      - killjoywashere 8y ago
        a) we generally don't have access to the records of any other health systems in the area. b) we generally assume the older data is stale. In the case of omics: did they keep the unaligned reads? What version of the genome was it aligned against? Was it a graph alignment? What depth? Did my geneticist approve of this methodology? The SNP evaluation? The variants? This goes all the way down. They'll question what generation of instrument produced the data, the duty cycles on the instrument, and on, and on, and on.
peter303 8y ago
Agreed. The somatic genome changes a tiny amount every cell division (1 miscopy per 100 million bases) due to spontaneous mutations, gene expression and occasional recombination. These may be useful for diagnosing disease and determining effective drug treatments.

killjoywashere 8y ago

> There's not much value in sequencing someone's genome over and over again though. There's plenty of value to each custodian that gets their own copy though, just like every cloud company wants their own copy of your name, email, and physical address.

adenadel 8y ago
Sure, but IMO it would be silly for another player to sequence you again rather than negotiate access to the data.
- killjoywashere 8y ago
  Ah, I see you're not in medicine. Physicians will repeat studies of arbitrary cost at the drop of a hat.
  - adenadel 8y ago
    But in those cases, they are actually looking for changes. For germline WGS there are no differences.
    - killjoywashere 8y ago
      a) we generally don't have access to the records of any other health systems in the area. b) we generally assume the older data is stale. In the case of omics: did they keep the unaligned reads? What version of the genome was it aligned against? Was it a graph alignment? What depth? Did my geneticist approve of this methodology? The SNP evaluation? The variants? This goes all the way down. They'll question what generation of instrument produced the data, the duty cycles on the instrument, and on, and on, and on.

peter303 8y ago

Agreed. The somatic genome changes a tiny amount every cell division (1 miscopy per 100 million bases) due to spontaneous mutations, gene expression and occasional recombination. These may be useful for diagnosing disease and determining effective drug treatments.