38 ms·

Creepy! It makes me wonder what other ailments people may currently have for which their isn't yet technology to discover. Had they done an Xray of Nellis they

by spraak 9y ago

Creepy! It makes me wonder what other ailments people may currently have for which their isn't yet technology to discover. Had they done an Xray of Nellis they surely would have seen the needle, I'd imagine.

ars 9y ago
> for which their isn't yet technology to discover Without a doubt there will be a ton of DNA based issues. Right now we know just a tiny number of them, and usually only when the issue is a single change. At some point we'll have the computing power to detect issues caused by two mutations at once, then 3, and so on. We would need massive quantities of data: Full genome analysis, and a full medical description in a computer readable format. So expect a DNA drive at some point in your future: Hopefully anonymized, and even more hopefully optional. (Anonymized is pretty likely - but optional, well, that one we'll see - I suspect at some point you will be required to submit data in order to receive health insurance, so technically optional, but not practically.) 2: Chemically induced issues. (What chemical specifically I obviously can not say.) I suspect we more or less know all the single-chemical issues. But issues caused by chemicals that act synergistically are much less likely to be known. 3: Various parasites and other types of mutualism (Commensalism, Amensalism).
- lewispb 9y ago
  We’re having such a drive in the UK, and my wife is working on the project. It’s called 100,000 genomes and aims to sequence the whole genome of 70,000 patients with rare disease or cancer (the other 30,000 genomes are from tumors), and their families. The ethics of consent and the possibility of incidental findings are all being taken into account.
- mnw21cam 9y ago
  In terms of finding issues caused by multiple mutations, it isn't so much a case of not having enough computing power - instead, it is about having enough patients with the same condition that gives enough statistical power to say for sure that these variants are the cause. Humans are extremely variable. If I do whole-genome sequencing on a random person, I'll find probably 3 million places in their DNA where they differ from the average Joe. 90% of those are common variants, and we can filter those out because we know they are common, but that still leaves on the order of 300,000 variants that are rare, the vast majority of which cause absolutely nothing. We already diagnose conditions caused by two mutations - they are called compound heterozygous, because there are two heterozygous mutations, one on each copy of a particular gene, therefore clobbering both copies and causing disease. We have very very few conditions where we know that mutations on multiple genes are required for disease - they are simply too hard to look for, as we don't have enough people in the world to get the statistical power.

by spraak 9y ago

ars 9y ago

> for which their isn't yet technology to discover Without a doubt there will be a ton of DNA based issues. Right now we know just a tiny number of them, and usually only when the issue is a single change. At some point we'll have the computing power to detect issues caused by two mutations at once, then 3, and so on. We would need massive quantities of data: Full genome analysis, and a full medical description in a computer readable format. So expect a DNA drive at some point in your future: Hopefully anonymized, and even more hopefully optional. (Anonymized is pretty likely - but optional, well, that one we'll see - I suspect at some point you will be required to submit data in order to receive health insurance, so technically optional, but not practically.) 2: Chemically induced issues. (What chemical specifically I obviously can not say.) I suspect we more or less know all the single-chemical issues. But issues caused by chemicals that act synergistically are much less likely to be known. 3: Various parasites and other types of mutualism (Commensalism, Amensalism).

lewispb 9y ago
We’re having such a drive in the UK, and my wife is working on the project. It’s called 100,000 genomes and aims to sequence the whole genome of 70,000 patients with rare disease or cancer (the other 30,000 genomes are from tumors), and their families. The ethics of consent and the possibility of incidental findings are all being taken into account.
mnw21cam 9y ago
In terms of finding issues caused by multiple mutations, it isn't so much a case of not having enough computing power - instead, it is about having enough patients with the same condition that gives enough statistical power to say for sure that these variants are the cause. Humans are extremely variable. If I do whole-genome sequencing on a random person, I'll find probably 3 million places in their DNA where they differ from the average Joe. 90% of those are common variants, and we can filter those out because we know they are common, but that still leaves on the order of 300,000 variants that are rare, the vast majority of which cause absolutely nothing. We already diagnose conditions caused by two mutations - they are called compound heterozygous, because there are two heterozygous mutations, one on each copy of a particular gene, therefore clobbering both copies and causing disease. We have very very few conditions where we know that mutations on multiple genes are required for disease - they are simply too hard to look for, as we don't have enough people in the world to get the statistical power.