February 24, 2006
On Falling DNA Sequencing Costs
The latest 454 Life Sciences DNA sequencer might lower the cost of sequencing a complete human genome to below $10 million.
454’s Genome Sequencer 20 ($500,000) uses an on-bead sequencing-by-synthesis approach to generate some 40 million bases of raw data per four-hour run, meaning I could squeeze out a human genome in just under a month. With per-run reagent costs of $6,000, my genome would cost a mere $900,000.
But that’s just one sequence pass, and according to vice president for molecular biology Michael Egholm, “It’s simply ludicrous to say you can sequence a human genome with 1x coverage.” He suggests 8x or 15x coverage, which would boost my costs to between $7.2 million and $13.5 million, and increase my sequencing time to about a year.
The whole article is worth reading and covers some of the academic and commercial efforts to drive down the cost of DNA sequencing.
While some rich people could afford to get their DNA sequenced now they'd derive very little benefit from doing so. The meaning of a small number of genetic variations is known. But they can be tested for individually.
Harvard professor George Church may have found a way to drive DNA sequencing costs down by a factor of 5 or more.
Church knew that a key to making gene sequencing fast and affordable lay in miniaturizing the process. He coats a slide with millions of microscopic beads, each impregnated with chemicals that light up when exposed to DNA base pairs. A digital camera fitted to a microscope photographs the pattern, and software decodes the results. His process is more than 250 times faster than conventional technology. In short, rather than take seven years to sequence the human genome, Church's machines can theoretically do it in less than a week. He says "theoretically" because he and his students have only decoded the DNA of E. coli, which is 1/1000th the size of the human genome. Based on his current costs, he thinks he could decode a human genome for about $2.2 million.
On Church's $2.2 million estimate see my August 2005 post "Harvard Group Lowers DNA Sequencing Cost Order Of Magnitude".
The first human genome cost $3 billion to sequence and costs have already fallen by over 2 orders of magnitude since then.
The PGP is an offshoot of the Human Genome Project, the massive government effort to read and put in proper sequence all 3 billion bits of human DNA. The project was completed in 2003 at about $3 billion - about $1 for each of the tiny chemical units, called bases, that make up the human genome.
Since then, better technology and greater efficiency have brought down the cost to $10 million - less than a penny per base - for a complete DNA sequence, according to Jeffery Schloss, the director of technology development at the National Human Genome Research Institute, a federal agency in Bethesda, Md.
The institute is financing a campaign to cut the cost of sequencing a genome to $10,000 by 2009 and drive it all the way down to $1,000 by 2014. An affordable $1,000 genome is biology’s next dream.
The cost of sequencing fell during the first sequencing of the human genome. So most of the $3 billion dollar amount represents a higher cost than had been achieved by the completion of the sequence.
Church is trying to recruit people into his Personal Genome Project at the Harvard Medical School where the recruits would allow comparison of their DNA sequences with lots of other information about them. We need this sort of research and on a massive scale with tens of thousands or even hundreds of thousands of enrollees. People should be enrolled long before DNA sequencing becomes cheap because some connections between genes and other characteristics will require longitudinal studies (i.e. studies that follow people for years and even decades).
Update: I see that we are on the cusp of a big change as a result of dropping DNA sequencing costs. 10 or 15 years from now people will use genetic analyses to formulate custom diet advice (nutritional genomics) to reduce the risk of diseases or choose the best diet for losing weight or putting on muscle. They will surreptitiously get DNA samples from their romantic interests to decide whether the other person has genetic profile good enough to warrant trying to marry (genetic tendency to cheat? intelligence? disease risks? violent tendencies? genetic tendencies toward laziness or conscientiousness?). Surreptitious DNA sample collection will find use for other purposes. Some sharp smaller employers will collect DNA from job interviewees (e.g. from a coffee cup) to analyse their DNA for personality tendencies and approximate level of intelligence.
Drugs will get developed for specific genetic profiles. Some will get preventative genetic therapies based on their genetic profiles to avoid diseases before they get sick.
Of course, mate selection is a pretty slow way to get the genes that you desire for your offspring and most will not be able to secure genetically ideal mates. More women will turn toward using sperm donors to get exactly what they want. Eventually gene therapies on sperm, eggs, and embryos will replace much of the coming increased use of sperm donors. But my guess is there will be a 5 to 15 year period during which use of sperm donors will soar before gene therapies provide better alternatives.
Still seems slow to me. I won't be pleased until I can get my entire genome sequenced for a shiny nickel. Then I'd carry it around on my ipod so I can prove my fitness to potential mates. I'll let you know how that works out for me.
wow... I sure hope Congress does the right thing and proactively regulates the usage of this information. What will prevent the eventual usage of this information for eugenics? I mean, that's the crucial issue here. Who really wants to be the recessive representitive? The nice thing about natural selection was that there was no choice; some animals just had to carry that burden, and hopefully their lion's share of recessive genes wasn't in itself fatal, while the existience of every combination of genome allowed preservation across any number of environmental disasters.
But now, all people can be freed from that responsibility. Heck, what if the apathy gene is the one that also makes us immune to bird flu? Who would willingly choose the apathy gene in their offspring? The conclusion is simple: by not enforcing the diversity of our species, we invite our own eventual genocide.
danfan, if you need your sequence to convince a potential mate, it probably won't help. :)
This is truly awesome work.
My one challenge to Randall and danratherfan is why push this into mate selection. Humans are simply not as rational as you assert when making this inference. I think a more material line of reasoning is how other technologies will get linked to procreation and who other than women will be making the decision.
About the use of sperm donors, don't think that's a big threat guys. Remember women still have to take care of these children and most want a man to help them with the work. Also children are financially a burden so they'll want a man around to help them there too. What are the odds men will want to help raise someone else's child when they are biologically capable of reproducing? What will men think of women who had children sired via a catalogue? Remember the nobel sperm bank went bankrupt.
Why push this into mate selection? When people have additional information they use it. My favorite saying: The streets find their own uses of technology.
I did not assert that humans are all that rational. The average IQ of the human race is well less than 100. Even people with high IQs make irrational choices and damaging choices.
Not all women can get even okay mates let along good or great mates and providers. In America the illegitimacy rate is over 20% for whites and about 68% for blacks with Hispanics somewhere in between. A woman who has a choice between a 90 IQ mate who won't even stick around and a 160 IQ sperm donor whose genetic profile is superior for disease resistance, looks, and personality is going to wonder why she shouldn't use the sperm donor.
Cheap DNA sequencing is going to make the value of the best donors much greater because women who can't catch Mr. Right will be far clearer on what advantages they'd gain by using a great sperm donor rather than Mr. Wrong.
What percent of kids are from sperm donors, and how has that number changed over the last decade or so?
I've gotta say that from the underclass people I've met, no thought at all goes into conception. It just happens. They don't plan ahead at all, and would not even understand that a sperm donor is an option.
I agree that the underclass won't rush to embrace sperm donors. But above maybe 90 IQ I would expect to see a rise in the use of sperm donors once it becomes possible for a woman to get a DNA sample of her potential mate and compare the IQ of resulting kids to what their IQs would be if she used sperm donors.
Imagine a 100 IQ woman has the choice between some guys she knows who are between 95 and 105. Her kids would come out around 100. But suppose a computer analysis of her own DNA could match her up with a sperm donor whose sperm would give her kids 125 IQ minimum. Suddenly the kids are going to grow up to be medical doctors, lawyers, accountants, and engineers rather than truck drivers or construction workers.
The underclass will get left behind as the middle class moves up in cognitive ability.
With cheaper DNA sequencing, men will respond to the increase in sperm bank use by increasing use of paternity testing. However, that's not to say it won't happen. I expect some women to choose a donor by matching most physical attributes to a current or prior mate.
Consider the situation now: Sue breaks up with Dan and hooks up with Bill. It turns out Sue is pregnant and Dan might be the father. In many cases, Bill will hang around and raise the child from birth. Bill doesn't know and doesn't want to know. He will probably resist any suggestion for a paternity test.
Consider the future: Sue really wants a child, but wants a really good child. When she breaks up with Dan, she orders a donor matched to most of Dan's physical attributes, and starts looking for a dad. She quickly hooks up with Bill, and it turns out she is pregnant...
But when it comes right down to it, a lot of men right now raise children they did not conceive. Many and perhaps even most of them love those children. Eventually, as technology improves, the compromise will use a lot of a father's dna with some baddies deleted and some goodies inserted.
Randall - "A woman who has a choice between a 90 IQ mate who won't even stick around and a 160 IQ sperm donor whose genetic profile is superior for disease resistance, looks, and personality is going to wonder why she shouldn't use the sperm donor."
Exactly. Eugenics is a very good thing that got tied up with some psychotic movements and imo wrongly discredited. Today and tomorrow it can used much more effectively then in the past. As you gave an example a woman who ends up with a 90 IQ man(average IQ in the world is around that), either because she didn't have other options or she may love him.. Yet she wants a high end genetic father for her child. The good looking, intelligent, fertile and successful man, many women want to have his genes for their child.. but by definition success is a relative term as are intelligent and good looking. So today only a few women can have those genes for their children.
"The underclass will get left behind as the middle class moves up in cognitive ability."
That is possible in that scenario. Another scenario is the underclass for a small cost can get high value genetic material for their children, then their children have advantages instead of disadvantages as now. If you are born today into a home of 90IQ parents, and get a bad draw having 80 IQ.. no matter how hard you study at school you aren't going to have a successful career.
Now imagine your mother got impregnated by a brilliant, attractive and healthy sperm donor.. and you got lucky having 130 IQ and came out good looking. I would say even though you are born into a poor home, you have big advantages over people in even upper middle class homes who are born 'average'.
Randall, I must have reading comprehension issues, I don't know why I thought you meant sperm donors would be used on the bottom of society. I guess I thought that because the cost of not having a dad would be a lot lower. But the upper middle class tends to have high investment dads, who also tend to be above average on most traits. Given that women make having kids an emotional decision, I'm still sceptical of large-scale sperm donation actually happening.
But if it does happen, it will really change the relationship between the sexes. Keep in mind the sexy son theory. You want children that their cohort will find attractive. That largely means finding the other parent who has the traits that your sex likes. (If I want desireable kids, I need to find a woman who's tall, blonde, thin, and smart.) Reducing paternal investment to almost zero will result in women looking for artists, musicians, bodybuilders, pick-up artists, etc. Looking at low paternal investment cultures, they suck, one and all. Sure adult women will want to have non-reproductive drones to help raise kids and make money, but they won't want to reproduce them. Most men will be condemned to never having children, and those who do, won't know which kids are theirs.
Regards parental investment: But a woman will be faced with an interesting dilemma. She can invest more in the future kid's DNA or in getting a guy who might stick around. If she invests in the kid via primo DNA then that investment can be counted on far more than if she tries for a father who will stick around. Also, the primo DNA will confer a bigger advantage in the long term.
Right now women are not fully considering the DNA investment for a couple of reasons. First off, the liberal press and liberal intellectuals deny the importance of DNA in determining intelligence, personality, and behavioral tendencies. Second, comparison is hard. How good is that donor sperm? Might turn out to be a dud. But cheap DNA sequencing will reveal so many politically incorrect truths that the era of ignorance about the genetic basis for human nature is about to come to an abrupt end.
Really smart women can attract smarter mates than average women. My guess is that the smart sperm donors will be most attractive to women in the 90 to 110 IQ range. Or maybe 100 to 115. Women who have average or below average mate choices who are smart enough to be aware of their situation will be most eager to use donor sperm.
However, the reason smarter women might use donors is that they'll be able to achieve income levels that make them able on their own to more easily raise children by themselves.
Keep in mind that women who select sperm donors will also be able to select for personality characteristics and behavioral tendencies. A woman who wants to have a boy who is genetically programmed for monogamy will be able to select a donor who will let her have that kind of boy.
First off good points Rob, you raise some very interesting thoughts.
Right now in Eastern Europe as an example over 75% of marriages are over within 20 years. So in many parts of the developed world we are already near the point of fathers not raising their biological children. Secondly we are doing more and more investing in children through the state, (wrongly I believe). So kids go to government school, then government run activities like daycare and sports for a large percentage of their waking hours. Add it all up and the children aren't having much time with the biological father as is.
Natural selection will soon make the women who want monogamous sons extinct. The women who choose sons who are confident and attractive to women will have greater reproductive fitness.
Bob, are you implying that men who are monogamous are not "confident and attractive to women"? BTW, is reproductive fitness everything that determines the success of a human? How about emotional stability? Why not ask some serial killer or Marshal Mathers?
If they want my genome sequence, they'll have to pay for it. A thousand dollars might cover it, I accept cheques and cards.