SoylentNews is people
SoylentNews
from the how-do-you-know-how-much-you-don't-know? dept.
The feat made headlines around the world: "Scientists Say Human Genome is Complete," The New York Times announced in 2003. "The Human Genome," the journals Science and Nature said in identical ta-dah cover lines unveiling the historic achievement.
There was one little problem.
"As a matter of truth in advertising, the 'finished' sequence isn't finished," said Eric Lander, who led the lab at the Whitehead Institute that deciphered more of the genome for the government-funded Human Genome Project than any other. "I always say 'finished' is a term of art."
"It's very fair to say the human genome was never fully sequenced," Craig Venter, another genomics luminary, told STAT.
"The human genome has not been completely sequenced and neither has any other mammalian genome as far as I'm aware," said Harvard Medical School bioengineer George Church, who made key early advances in sequencing technology.
[...] FAQs from the National Institutes of Health refer to the sequence's "essential completion," and to the question, "Is the human genome completely sequenced?" they answer, "Yes," with the caveat — that it's "as complete as it can be" given available technology.
[...] Church estimates 4 percent to 9 percent of the human genome hasn't been sequenced. Miga thinks it's 8 percent.
https://www.statnews.com/2017/06/20/human-genome-not-fully-sequenced/
I'm glad this is finally getting some coverage. A few years ago I looked into the human genome to prove to myself it didn't contain a certain sequence, and found this was impossible since ~10% of it was missing. When they talk about "sequencing a genome" it is total false advertising.
(Score: 3, Interesting) by kaszz on Sunday July 30 2017, @02:14PM (17 children)
So what makes the last percents so hard to sequence?
(Score: 3, Informative) by Anonymous Coward on Sunday July 30 2017, @02:53PM (5 children)
DNA sequencing only reads small chunks at a time, TFA says typically 1000 base pairs in the Human Genome Project, but mentions new capabilities of up to 60000 base pairs. You get a ton of these short sequences, then turn a computer loose, looking for overlaps and trying to build one complete sequence from them.
The remaining gaps are where sequences couldn't be spliced because of ambiguity created by repeating sequences longer than the maximum read length.
(Score: 2) by kaszz on Sunday July 30 2017, @03:21PM (1 child)
Seems the sequencing method is incomplete. Just imagine a harddisc that would read random blocks and let the filesystem try to figure out which blocks belonged to what file from the block contents alone without any LBA reference or anything. A better method where the DNA string is pulled out and read in one long sequence is needed.
Reminds of CD-rippers where block reading would not align properly so the ripper program had to align them on the fly.
(Score: 0) by Anonymous Coward on Monday July 31 2017, @05:25PM
This method is what they were using before shot-gun sequencing came about.
In several years they managed to sequence was shot-gun sequencing covered in a matter of months.
The linear read method is really hard, though they should be able to use it to target these few ambiguous regions.
(Score: 2) by takyon on Sunday July 30 2017, @03:37PM (2 children)
If newer machines can read fragments that are 60x longer than before, could some random joe's $1,500 sequencing yield a more complete sequence than the HGP reference genome?
[SIG] 10/28/2017: Soylent Upgrade v14 [soylentnews.org]
(Score: 0) by Anonymous Coward on Sunday July 30 2017, @08:37PM (1 child)
60x longer is a far stretch, there would be quite some engineering (both physical and biochemical) problems to overcome (and you would require a fundamental new technique of doing it). But sequencing is just part of the problem. The genome you want to sequence is often first added to a library, genome is cut up and parts are randomly placed in "stable" vectors, which are transformed to bacteria/yeast. Some genes could be expressed from these vectors and be toxic to the bacteria/yeast (hence the quote marks in stable), hitting a hole in your sequence. Even if the vector survives, the sequences in the library are random, which would mean that, statistically, you'll never get 100% certainty to obtain 100% of the sequences.
Newer sequencing techniques actually use shorter reads, but the problem stays the same, you're always looking at a pool of sequences of the large complete sequence.
(Score: 3, Insightful) by gringer on Sunday July 30 2017, @11:41PM
No, it's not. Nanopore can get quite a lot of sequences over 60kb, and Nick Loman has managed a end-to-end-mappable sequence that is over 700kb in length.
The challenge is now in sample prep, not sequencing. It's quite difficult to keep DNA intact enough for really long reads with most library preparation methods-
Ask me about Sequencing DNA in front of Linus Torvalds [youtube.com]
(Score: 1, Interesting) by Anonymous Coward on Sunday July 30 2017, @03:57PM (2 children)
Progressively worse funding since 2001 is the problem.
Americans don't care about science. Since 9/11 the singular purpose of America has been the targeted murder of towelheads. The year is 2017 and there isn't any money left for anything else.
The last time the human genome was almost completely sequenced, a grad student sacrificed the use of his wrists to finish the project in a month before funding was cut.
Today you would have to find someone naive enough to risk death for the privilege of working on the project.
(Score: 1, Insightful) by Anonymous Coward on Sunday July 30 2017, @06:12PM (1 child)
Not enough funding is why geneticists act like weasels and say they completed the sequencing when they actually were unable to achieve their goals? Sorry, but no.
(Score: 0) by Anonymous Coward on Sunday July 30 2017, @07:44PM
... wait, what?
kaszz: "What's the problem with sequencing?"
AC1: "Lack of funding."
AC2: "That doesn't excuse their lying!"
AC2, please stop responding to nonexistent arguments. Thank you.
(Score: 2) by gringer on Sunday July 30 2017, @11:37PM (6 children)
Highly repetitive sequences that are longer than the template length of Sanger sequencing and Illumina sequencing:
http://i.imgur.com/p5U1YCz.jpg [imgur.com]
https://twitter.com/gringene_bio/status/887822844280709120 [twitter.com]
https://twitter.com/gringene_bio/status/882006992268648449 [twitter.com]
Long non-tandem repeats also cause problems.
Ask me about Sequencing DNA in front of Linus Torvalds [youtube.com]
(Score: 2) by kaszz on Monday July 31 2017, @02:13PM (1 child)
The last tweet is intriguing .. "The missing bits of code in the human genome could be key for understanding cancer.".
(Score: 0) by Anonymous Coward on Monday July 31 2017, @03:13PM
The tweet links to an article that links back to the article in the OP.
(Score: 2) by kaszz on Monday July 31 2017, @02:17PM (3 children)
If repetitive sequences is the problem. Then the solution seems to be to stabilize long enough sequences and read them. The question is then, how long sequences do these need to be? and how long is the total?
Is it possible to read the DNA code in place using say laser interferometry etc?
Or other complete out-of-the-box methods.
(Score: 2) by gringer on Monday July 31 2017, @08:23PM (2 children)
To capture everything, about 3 million bases in length, according to Karen Miga at the London Calling conference this year.
There are a lot of "totals" that I can pull out of human genome statistics; you need to be more specific than that. Concentrating just on centromeric regions, she said that the average centromeric length is about 3 Mb, so that suggests there's a bit over 60Mb of highly-repetitive sequence in the human genome.
Lasers and/or SEM are not yet able to scan at the 0.33 nm resolution required to resolve single bases of DNA. There have been a couple of papers in the last month or so discussing methods of visually resolving the 3D structure of DNA, but nothing about an adapter-free resolution of the sequence.
Nanopore sequencing is pretty good; it's what I used to make those images. Nick Loman has managed to get a read of a bit over 750 kb, and I expect that if ONT can get their accuracy for most 100bp sub-sequences up to Q30, or quadruple the read length, reads that span (or unambiguously cover) centromeres will be possible.
Ask me about Sequencing DNA in front of Linus Torvalds [youtube.com]
(Score: 2) by kaszz on Tuesday August 01 2017, @12:52AM
Just a thought.. 8 kV would accelerate electrons with the energy for a x-ray photon at 0.1 nm wavelength. Possible solution?
I'm thinking if the x-ray photon is shoot at different angles and the output angle is measured. A problem can be if the sample is not still though.
(Score: 2) by gringer on Tuesday August 01 2017, @06:11AM
Karen Miga has just put a preprint paper onto BioRXiv about resolving the centromere of the Y chromosome using nanopore reads:
http://www.biorxiv.org/content/early/2017/07/31/170373 [biorxiv.org]
Ask me about Sequencing DNA in front of Linus Torvalds [youtube.com]
(Score: 0) by Anonymous Coward on Monday July 31 2017, @03:15AM
Well, as everybody knows the first 90% of the work takes 90% of the time. The remaining 10% takes 90% more of the time.