dNATE.com

A simple blood test for Down Syndrome

A simple blood test for Down Syndrome is now available for women who are 10 weeks pregnant. It's more accurate than amniocentesis, and safer for the mother and child.

A recent article asks: Do All Women Need Genetic Testing Before Pregnancy?

More than half of pregnant women do some kind of prenatal screening or testing of their fetus ... The vast majority of women undergo [disease] carrier screening based on their ethnic background, although many, regardless of their heritage, are also screened for cystic fibrosis. Traditionally, certain ethnic groups — especially Ashkenazic Jews of Eastern European descent — have been advised to consider carrier screening for various fatal conditions including Tay-Sachs and Canavan diseases.

[GenPath recently announced a test to detect] 160 diseases in a single genetic screening test. Indeed, GenPath is presenting the new test as a way to level the genetic playing field and expand testing to any woman, anywhere.

This is just the latest of a battery of new blood tests that can determine everything from fetal sex to Down syndrome to paternity.

Bacteria and cancer

Scientists have found a link between bacteria and colorectal cancers.

"Tumors and their surroundings contain complex mixtures of cancer cells, normal cells, and a variety of microorganisms such as bacteria and viruses," says the study's senior author, Matthew Meyerson, MD, PhD, of Dana-Farber and the Broad Institute. "Over the past decade, there has been an increasing focus on the relationship between cancer cells and their ‘microenvironment,' specifically on the cell-to-cell interactions that may promote cancer formation and growth." 

While the relationship between colorectal cancer and Fusobacterium is unclear, there are intriguing hints that the bacterium may play a role in the cancer, says Meyerson.  "At this point, we don't know what the connection between Fusobacterium and colon cancer might be," Meyerson observes. "It may be that the bacterium is essential for cancer growth, or that cancer simply provides a hospitable environment for the bacterium."

Recent studies have also found a link between bacteria and pancreatic cancer.

The buzz on genes and behavior

Researchers have long worried that correlating genes and behavior was "too complex to fathom, because so many factors can act together to regulate behavior."

Now new research shows how a few "master genes" can switch on and off complex behaviors, at least in bees.  

Researchers found that "naturally occurring behaviors are under the influence of discrete regulatory networks in the brain ... there is a close relationship between changes in gene expression ... and behavior."

They also found "a high degree of modularity in the regulation of genes and behavior, with distinct behavioral states".  Words like "discrete" and "modular" are important, as they show that complex networks created by genes can be controlled and achieve a steady state.

"The team focused their analysis on ... genes implicated in ... foraging, such as scouting for flowers ... maturation, ... and aggression, or hive defense."

Look for more informatio on the BeeSpace project which developed "a catalog of genes that turn on or off in the bee brain in response to social cues, environmental changes or as a result of hereditary factors."

A virus that cures cancer

Scientists have recently discovered that certain viruses can selectively kill cancer cells.  Here's how it works:

The virus, which is injected into the bloodstream, targets only cancer cells that express a certain high-growth factor.  Once inside a cancer cell, the virus triggers an immune response that kills the cell, leaving healthy cells unaffected.

World's coolest robot

The Nao is officially the world's coolest robot.  See the video.

But it's not just the physical robot that's cool.  The robot's activity can be fully simulated on a computer.   See, for example, a soccer match conducted either with real, physical robots, or purely virtual - in silico.

 

Are identical twins identical?

Identical robots (i.e. robots with identical programming) aren't necessarily identical.  Consider, this program:

set red_light_exposure = false;  // initial state

loop

    if (red_light_detected()==true)

       then // cause permanent change to internal state

         set red_light_exposure=true;  

    if (red_light_exposure==true)

       then run_scenario_A

       else run_scenario_B;

repeat loop // endlessly

Two robots with this same programming (i.e. same DNA) would operate differently if one was ever exposed to red light (and the other wasn't).  Different sections of code would be executed, and the change would be semi-permanent.

Epigenetics in humans operates by the same mechanism.  It causes a semi-permanent, extra-genetic change, triggered by one's experiences.  If one twin has a certain experience (and the other doesn't), then it's possible the twins will develop different behavior or appearance.

It also appears - contrary to recent scientific belief - that identical twins may also have slightly different genetic material, known as copy number variations.

Does this invalidate twin studies, which seek to clarify nature/nurture effects, as some believe?

It depends.  The only experience that can affect us must be wired into our innate detection circuitry.  Otherwise, that experience is ignored.  The robot example (above) shows that exposure to "red light" (but not blue light) can cause a permanent change to our internal state, because it's explicitly included in the programming.

So we'd need a better understanding of which environments are innately detected, and whether an innate switch (or latch) retains the experience.

We're 99% microbe

A nice article in the New York Times points out that we humans are 99% microbe.

99 percent of the functional genes in the body are microbial. Researchers [are uncovering] how genes in this microbiome — exchange messages with genes inside human cells.

Most DNA, for example, was long considered junk — a netherworld of detritus that had no important role... [However] signals coming from both regions of the genome participate in the delicate balance between normal cellular behavior and [cancer] malignancy.

Some 90 percent of the protein-encoding cells in our body are microbes. We evolved with them in a symbiotic relationship, which raises the question of just who is occupying whom.

As I've previously described, this mechanism is also a possible engine of "directed evolution".

Don't like your DNA? Replace it!

Hemophilia, a straightforward genetic disease, can now be cured in mice. There's no reason the technique can't be applied to humans as well. 

Here's how it works.  Using a specialized enzyme, snip out the offending gene, and employ a virus to deliver a new, better functioning gene in its place.

The mice had been engineered to carry a human gene for hemophilia, and the enzymes and virus had been specifically engineered as well: the enzymes would cut only certain sequences of DNA, patterns that were known to appear on either side of the defective gene, and the virus, which naturally infects the liver cells where the clotting factor is made, would swap in an unmutated gene, instead of the viral genes it would carry in nature.

How genetics will solve the debt crisis

According to the New York Times:

WHEN we think of the qualities we seek in visionary leaders, we think of intelligence, creativity, wisdom and charisma, but also the drive to succeed, a hunger for innovation, a willingness to challenge established ideas and practices.

By implication, not everyone has these qualities. Key leaders and highly productive managers and employees - i.e. the people who pay most of the taxes - are fairly rare. So where do these qualities come from?

Republicans believe that you're just "born that way". (See the Bell Curve, for example.) They oppose excessive government spending on education and social programs, because they believe such efforts are wasted, and that existing programs have largely failed (e.g. the collapse of "no child left behind").

Democrats realize this, yet their political constituents demand these programs (in return for votes). And why not? If Republican claims are true, it only bolsters the rationale for redistributionist social programs. What's the alternative? Throw people without the "right" innate traits out on the streets?

Yet America can no longer afford all the social programs. There are too few highly productive workers to be taxed to supply the revenue. Unemployment is high in the U.S. because goods can be produced in China and elsewhere more cheaply.

So what's the answer? Genetics.

Here's how it works. Identify the genes held by those with the "New York Times" traits for success. Then transfer those genes using artificial chromosomes into the rest of us. (You don't need to know which gene does what. Simply transfer all the genes - around 6,000 of them - that are active in the brain of elite individuals.... and throw in some Michael Phelps sports genes while you're at it.).

As the next generation of workers receive these artificial chromosomes, America will experience a renaissance in productivity. New ideas, faster, cheaper, more efficient. Businesses will boom, paying more taxes. Enough to solve the debt crisis, to be sure.

The Brain on Trial - errata

It's good to see the legal community finally grappling with the implications of biology. David Eagleman writes in the Atlantic about criminal behavior and culpability, and concludes that

Whether in empathy, intelligence, impulse control, or aggression … People are not created equal … Not everyone is equally “free” to make socially appropriate choices … If free will does exist, it has little room in which to operate … [there is a] myth of human equality

Eagleman argues that “blameworthiness should be removed from the legal argot”. Regarding the choices a criminal - or anyone else - makes:

it no longer makes sense to ask, ‘To what extent was it his biology, and to what extent was it him?’ because we now understand that there is no meaningful distinction between a person’s biology and his decision-making. They are inseparable.

Here's my issue. The desire to commit crimes (or lack of control) comes from my wants, impulses and interests, which come from my genes (i.e. genetically constructed regions of the brain). Yet I freely choose, based on who I am (what I innately want). In other words: "I am my genes". I commit crimes based on who I innately am. That's perfect free will - freedom to follow your innate programming. What you want and what you choose (and who you innately are) are all identical.

This implies that the legal system is simply a set of rules imposed by one set of genetic beings over another. This has its origins 15,000 years ago, when human civilization began its differentiation into specialized roles of leaders and followers, to allow society to grow and scale. (Perhaps ancient “hierarchy genes” stored in our junk DNA were brought back for this purpose to re-establish our specialized roles as worker bees, drones, and queens.)

By this account, there is no right and wrong in the law. There's no implied fairness for individuals. Laws are a control mechanism to allow society to function more smoothly, a cultural reflection of genetic specialization. If you have the “poor impulse control gene” then you're incarcerated. Like quantum physics, there is no separation of law-giver and law abider, just genetic beings interacting with other (specialized) genetic beings.

Sure, we can try “customized rehabilitation” of criminals and “tailor sentencing and rehabilitation for the individual” based on genetics (or brain hardwiring) as Eagleman suggests. We can try to encourage people to tune their frontal lobes to control impulses using biofeedback. But all this complexity will add confusion and reduce the law's clarity and impact.

Sooner or later, we will all ask the question: why accept this genetic hierarchy, when with a few genetic changes we can pluck out our innate “poor impulse control” and “aggression” and replace them with new genes for success (and no jail time)?

A critique of David Brooks' The Social Animal

I’m fascinated by David Brooks new book "The Social Animal", not because the ideas are originally his, but because Brooks synthesizes recent findings of researchers in many fields.

The message is an essentially negative and pessimistic one, especially if you believe in free will:

• "The unconscious parts of the mind are most of the mind – where most of the decisions and many of the most impressive acts of thinking take place"
• The unconscious realm includes "emotions, intuitions, biases, longings, genetic predispositions, [and] character traits"
• “The conscious mind merely confabulates stories that try to make sense of what the unconscious mind is doing of its own accord.”
• This “removes the conscious mind from its privileged place at the center of human behavior”

Yet Brooks tries to put a positive spin on it. “This is the happiest story you’ve ever read,” he writes. Brooks invents imaginary characters – Harold and Erica – with an admirable set of rare innate traits:

• “persistent after setbacks”
• “possessed enough confidence to take risks”
• could “control their worst impulses”
• could “read people”
• “you could put them in front of a crowd”

Sure, it’s a happy story if you’re born with the “predispositions” of the meritocratic elite. But how about the rest of us? “More than a quarter of high-school students drop out” and 70% of adults never graduate from college.

Today’s corporate CEOs struggle to find such talent who are comfortable in their own skin and can be “put in front of a crowd”. These are the “seedbeds of accomplishment”, according to Brooks. Yet most people simply want approval from their superiors. Most of us don’t want to stand out. And certainly, we get panic attacks when presenting in front of crowds.

Brooks tries to have it both ways. He writes that “the conscious mind can influence the unconscious” and the emotions can be “properly nurtured” and “educated” and “don’t control our lives”. Yet how can the conscious mind choose to go against the unconscious mind, if "what we want [and] how we perceive the world” are determined by the unconscious? The conscious mind doesn’t have preferences of its own. So how can it override the subconscious? Even Brooks admits that “mental processes ... inaccessible to consciousness organize our thinking, shape our judgments [and desires, and] form our characters”.

Brooks refers to an “outer mind [that] hungers for status, money, and applause, [while] the inner mind hungers for harmony and connection”. In reality, there is no inner/outer distinction. Need for status is unconscious. Power-hungry people are not conscious of why they feel that way. Traits are not something you have; they are who you are.

“Character is formed,” says Brooks, and this is certainly true. Nobody is born power-hungry. But how does it develop? Certainly the brain requires experience in order to wire itself. But it’s not just any experience; it’s expected experience. Any other experience is essentially ignored.

We evolved in a social context. Our genes expect other people to be around. Our genes hardwire us to suckle at our mother’s breast. Our genes hardwire us to turn our heads and look up, and remember what we see. Later, this imprinted memory of “mother” can be exploited by the genes for other purposes. For example, the genes can establish a hardwired reaction to our mother, once we learn to recognize who she is. So (expected) experience is required for proper hardwiring to occur.

The development of character proceeds in a similar way. Some people are excited by the presence of others, especially those who can help them socially or politically. The mental capacity to perceive the “right people” takes years to develop. But it all starts with interest. If you’re not interested in forming networks – if you’re a homebody or a shy loner – you won’t do it. If you can’t stand kissing up to power, you probably won’t get promoted.

The unconscious mind “coats things with emotional significance”. But no two people are alike in their interests, and thus their capabilities.

If you’re born with the (unconscious) traits to be a meritocrat (as Brooks is), it must be a happy life of easy connections, influence and easy money. Your wants, desires, and passions are highly rewarded by society. Yet the rest of us struggle with an unconscious that leads to introversion, insecurity and followership. We’re easily exploited by the Oprah-esque meritocrats who tell us we can change with conscious effort to be just like them ("just buy my book!"), and who possess the confidence to pat us on the back, without their own need for approbation.

Evolution is not random

For once, I agree with the Pope. Evolution is not random.

Evolution is a process by which genetic changes are passed from parents to children. Children live or die, depending on the success they have with their new genes in the current environment. If they survive to bear their own children, the new genes are passed along to the next generation.

Nothing random about that. It's the environment that determines which new genes remain and which disappear with the death of the childless organism. The genes that remain somehow resonate with the environment. Thus, genes reflect evolutionary experience.

And the genetic change from parent to child is not random either. "Do genetic mutations really occur at random spots along the genome, as researchers have long supposed? Maybe not" according to a new study. Why? I assume it's because the genes have long ago evolved strategies to direct their own evolution.

So the recent announcement of new fetal DNA testing techniques should come as no surprise.

If you're planning children ... you may want to know if you're a carrier for a serious single-gene disorder such as Tay-Sachs; if close relatives have had breast cancer, you may want to learn whether you have the mutation that significantly raises your risk of the rare familial form of the cancer.

[A new] technique being developed analyzes fetal DNA that is collected from women's blood as early as five weeks into a pregnancy. So-called "noninvasive prenatal diagnosis," or NIPD, may hit the market as a test for Down syndrome later this year. [With] NIPD you'd simply give a little extra blood at the lab at your first prenatal checkup. [Painful amniocentesis would become a thing of the past.] There would be no risk at all to you or the fetus. And you'd get the results [and choose whether or not to keep the fetus] before you were visibly pregnant, before you'd told your mother or your friends.

Scary, but it's the latest example of genes (i.e. the organisms that develop from genes) choosing to direct their own evolution.

The problem with David Brooks' "new humanism"

Political and social commentator David Brooks has written a new book called "The Social Animal" in which he tries to elevate the role of human emotion and the unconscious mind into the political debate. He appears to agree with Dennis Overbye who writes,

A bevy of experiments in recent years suggest that the conscious mind is like a monkey riding a tiger of subconscious decisions and actions in progress, frantically making up stories about being in control. [Our] conscious brain [is] only playing catch-up to what the unconscious brain [is] already doing.

Our preferences, says Brooks, have deep innate underpinnings, of which we are not consciously aware.  Most human decision-making emanates from the unconscious mind, where emotion provides a largely hidden ability to assign value to our decision-making options.  A man without emotion is not rational, but rather indecisive and lost. 

Brooks asserts, however, that you can learn strategies to control your impulses through repetition, and thus gain self-control. Brooks claims that "character" can be trained through rigorous "discipline" and 10,000 "small repetitions" to develop "habits and strategies" that can overcome your perceptions.  You can "learn to see" things differently through absorption of new "norms". 

What Brooks won't tell you

Is it true?  Can we "learn strategies to control impulses" through practice?  Certainly you can teach yourself anything if you're interested and motivated.  But if you're not interested, you'll stop practicing.

The philosopher, Arthur Schopenhauer, said that we can do as we will, but we cannot will as we will. In other words, you can't change your (innate) preferences (consider this case for example).  What people don't understand is, if you don't want to do something, it's the same as not being able to do it. Some people don't want to practice, because of who they genetically are.  Freedom to follow your innate preferences is not really "free will".

Genes place a leash on our capacities, because they determine what things interest us enough to practice them.  Practice leads to habit, and gradually to expertise.  Since we all differ genetically, we all have different interests, desires, cravings, and passions. Different interests lead to different practice regimens, which lead (deterministically) to inequality in talent.  Psychologist William James "believed in free will on 'ethical grounds,' [but] he did not believe that there was evidence for it on scientific grounds."

In Mischel's famous marshmallow test, for example, "the ability of 4-year-olds to postpone gratification by leaving a marshmallow uneaten for a time as a condition of receiving a second marshmallow was a very good predictor of success in life."  Brooks writes: "The kids who could wait a full 15 minutes had, 13 years later, SAT scores that were 210 points higher than the kids who could wait only 30 seconds”. He doesn't mention that differences among children in the ability to delay gratification are genetically linked.

The truly novel implication that Brooks doesn't fully explore is human diversity; that we're not all the same, genetically speaking. While Brooks hints at these issues (the "meritocratic class re-inforces itself through genes" and "kids are born with a certain temperament"), he describes the mind as an "emergent system," a product of murky nature/nurture "interactionism".

Brooks can't have it both ways.  He cites "parental sensitivity" as the "causal arrow" of a child's success (a little lie to boost book sales among the Oprah crowd), even as he admits that 30% of abused children turn out just fine, since they possess an innately resilient temperament.  In other words, genes trump environment, but only if you're a member of the genetic meritocracy, as Brooks is.  Born leaders have traits like high energy, passion, ability to motivate others, fascination with power hierarchies, and extreme self-confidence. These are rare traits, which is why leaders are so highly compensated for their rare gene variants.  In other words, they're paid just to be themselves, i.e. who they genetically are.

Obviously both political parties would shun Brooks if he followed this argument to its logical conclusion, which is why he cloaks his message in muddy sentiment. Yet even as Republicans promote individual responsibility, they secretly believe in a natural God-given hierarchy in society.  Democrats, on the other hand, endlessly exploit the human hardwired belief in "free will" (despite evidence to the contrary), and thus gain political power by cynically promising that anyone can do anything with the right government programs. 

Yet the Democratic myth remains, for now, the more politically compelling one, until such time we can rectify the unfairness of natural inequalities in society.

Race and drugs

The FDA recently approved a new drug for lupus, even as that drug (like all drugs) doesn't work for everyone:

African American patients ... did not appear to respond to treatment with Benlysta. ... To address this concern, the sponsor has agreed to conduct an additional study of people with those backgrounds.

The FDA has approved drugs for specific racial groups before. For example, "BiDil" was approved in 2005 for Blacks only, prompting critics to complain that,

by approving a drug for one group, the FDA was using race as a crude shortcut for genetic typing ... Endorsing a drug for one race gave official government imprimatur to the discredited notion of race as a biological category.

Computer beats human at Jeopardy

IBM's Watson computer recently beat a human opponent at Jeopardy by employing an exciting new approach as described by Peter Norvig, Google's director of research:

This approach of relying on examples — on massive amounts of data — rather than on cleverly composed rules, is a pervasive theme in modern A.I. work. It has been applied to closely related problems like speech recognition and to very different problems like robot navigation. IBM’s Watson system also relies on massive amounts of data, spread over hundreds of computers, as well as a sophisticated mechanism for combining evidence from multiple sources.

The current decade is a very exciting time for A.I. development because the economics of computer hardware has just recently made it possible to address many problems that would have been prohibitively expensive in the past. In addition, the development of wireless and cellular data networks means that these exciting new applications are no longer locked up in research labs, they are more likely to be available to everyone as services on the web.

Genetic Bill of Rights

Massachusetts is proposing a Genetic Bill of Rights to address gaps in the federal Genetic Information Non-Discrimination Act (GINA), passed three years ago to prohibit genetic discrimination in employment and insurance coverage.

Hopefully the bill will also address the usurpation of individual genetic rights now practiced by the FDA.

According to CGS the Massachusetts bill would establish:

• a right to privacy with respect to genetic information and material
• that genetic information and material are personal property
• that health care facilities cannot deny treatment & wellness services solely based on the discovery of a genetic marker
• that genetic information and material may not be used in conveyance of life, long-term care and disability insurance
• that auto insurers cannot use genetic information in establishing policies and rates
• other protections, making identity theft stemming from misappropriation of genetic data a criminal offense, for example, and forbidding bankers from using genetic profiles.

Molecular Motors

Two of the strangest proteins in the human body - helicase and kinesin - act as molecular-size motors.

When DNA unpacks itself from tightly wound chromosomes during cell division and gets twisted and tangled, like yarn carelessly pulled from a hank, helicase makes a cut in the DNA, allows it to spin and untwist, then repairs the cut.  Nifty.

Kinesins, on the other hand, are crawlers. During cell division, they drag replicated chromosomes into one of the two new cells.  Using cellular rope called "microtubules," kinesins alternatively grasp, pull and let go - hand over hand - to draw the chromosomes into position.  There are 124 different types of kinesins in a cell.  Very cool.

 

Why machines will one day surpass the human mind

A recent cover story in the Atlantic magazine claims that artificial intelligence and machines "will never beat the human mind". While it's easy to make trivial objections to this argument (e.g., "never" is a long time, and "the brain is itself a machine constructed of squishy, organic parts"), the claim has some validity.

We humans have predicted artificial intelligence and friendly talking robots for decades now, and they never arrived.

But I'm not pessimistic. Certainly we'll need another Einstein or Niels Bohr to make the next theoretical leap, and it's difficult to predict the timing of such creative insight. It could come next week, or in 100 years. For now, we're stuck in our current paradigm.

To get unstuck, we must open our minds to new approaches to the problem. Don't just write a computer program that "acts like a brain". Conduct a thought experiment, as Einstein might do:

• Consider this: If you stare at an object without looking away, the object eventually disappears. Perception requires change.
• Also consider this: If you live next to a railway and have grown accustomed to hearing a train pass at 3am every night, you'll wake up if there is no train at 3am. Perception is prediction.
• Everything in the brain is about perceiving sequences (of events, objects, etc) through time and space. But perhaps the brain represents time spatially? Or by "multiplexing"?
• Forget language. Focus on emotion. Try to create a robot that's shy in certain situations.
• The number of different emotions we can feel is finite (e.g., jealousy, happiness, etc) and genetically prescribed. Yet the object of emotion is not, e.g. I can be jealous of someone else's brand new computer. How is this implemented in the brain?
• Some people are born extroverts, some are born introverts. How is that implemented in the brain?
• How would you construct a robot that enjoys practicing, and can direct itself to those things it must practice to gain skills and talent and expertise?
• Imagine a robot that can balance and walk on two broom handles in a wind storm. No one has solved that problem yet. There's no reason to build a physical robot. Everything can be simulated on a computer these days, including the laws of physics.
• Imagine an ant or cockroach. No one has ever simulated them accurately in a computer, and their brains are tiny. Start small.
• The brain doesn't represent the world so much as resonate with it. Like particles in quantum physics, perhaps concepts in our brain don't exist until interaction causes them to exist.
• Study the brain for clues. The human brain contains 100 billion neurons. That's 100 billion machines acting semi-autonomously, receiving signals from thousands of other machines every second. And yet there is an emergent effect of thought and intelligence
• Computer beats human at the game of Jeopardy by employing massive numbers of examples, not elegant rules

And so on... Yes, the brain is a machine. We don't understand it yet, that's why it appears magical and unbeatable to us. For now.

We can handle the truth

The FDA has been attempting to censor an individual's right to know our own genes, by shutting down access to personal genetic testing.

Thomas Goetz, who is just as outraged as I am, writes: "The controversy here rests on a single assumption: That the typical ... American isn't prepared to learn about their DNA ... We can't handle the truth."

Fortunately, Scripps has recently offered supporting evidence for those who oppose such government overreaching.

In any case, there is a face-saving alternative for the FDA, which is to partition the two activities (offering genetic testing services, and making scientific claims on genetic variants).  Perhaps no single entity or private company should do both.  But don't take away our right to know our own genes, under the guise of consumer protection.

Genetics and social networks

A new study finds that people tend to "friend" others who have similar or complementary genes:

Researchers from the University of California, San Diego found that gene clusters were comparable among friends, even after accounting for the predisposition to befriend others in the same geographic area.

"It does blur the line between nature and nurture, that's for sure," said study author James Fowler, a professor of medical genetics and political science. "We show in this paper that our environments, in part, are based on the genes of our friends. What the study brings to light is that genes of other people are part of what affects our environment."

The team was surprised to find that individuals also tend to befriend those with opposite but complementary traits. People who carried a gene known as CYP2A6, which is associated with an open personality, tended to form friendships with others who lacked that genotype.  [In other words, innate leaders tend to befriend innate followers.]

It's disrespectful to question Professor Amy Chua

As a student at Harvard Law, Amy Chua, "wasn’t naturally curious or skeptical like other law students... [she] 'just wanted to write down everything the professor said and memorize it.'”  I can imagine her in class, never raising her hand to question or contradict or contribute, but always getting A's; stealing the intellectual production of others and pawning it off like fake Rolexes.

Now she's a Yale professor, writing “Why Chinese Mothers Are Superior" as a testament to her extreme parenting style that encourages rote memorization without critical analysis or the "creative destruction" of opposing ideas.  She assumes it's her turn not to be questioned.  How embarrasing for Yale! 

First, her basic premise is wrong:  "Amy Chua has smart kids. They would [still] be really smart if she had put them up for adoption at birth ... Similarly, if Amy Chua had adopted two children at birth who turned out to have [low IQ] she would have struggled to get them through high school, no matter how fiercely she battled for them. Accepting both truths—parenting does matter, but genes constrain possibilities—seems peculiarly hard for some parents and almost every policy maker to accept."

Second, it's wrong to raise unquestioning kids in any case.  America will surely decline if we adopt the Chinese style of raising our children.  In China, anyone who questions authority is thrown in jail.  Corruption reigns.  Innovation - which requires independent, controversial thinking - does not exist, since anyone can steal your idea or trade secret with impunity. 

In America, the humanist movement has reached its zenith.  The individual is king.  Every individual has equal moral worth.  Yet as we enter an era of extreme genetics, our moral equality will be challenged.  There are already hints that the government may soon engage in widespread genetic screening to identify talent.  They won't be stopped by a credulous people who only believe what they are told.

If we let down our guard with Chinese-style docility, we are lost.  Let's show Amy Chua some disrespect.

The brain cools the blood?

Aristotle famously declared that the brain's purpose is to cool the blood.  (It was left to Herophilus to point out that the brain actually generates a lot of heat, and is the true source of passions and the intellect.)

But it turns out the Peripatetic One was on to something after all.  The next generation of computer chips (in order to satisfy Moore's Law) will resemble the brain's 3D design.  As such, internal heat dissipation will be a big problem:

[Researchers seek] to understand how the latest chip cooling techniques can support a 3D chip architecture [with] an interconnect density from 100 to 10,000 connections per millimeter square. [They] believe ... the use of hair-thin, liquid cooling microchannels measuring only 50 microns in diameter ... are the missing links to achieving high-performance computing with future 3D chip stacks.

DNA in the movies

Hollywood maintains a Freudian, Woody Allen-esque view of human behavior, which is to say, everything you are comes from your experience (not your DNA).  After the Gattaca debacle (the movie cost $36 million to make, and grossed only $12 million), Hollywood has been reluctant to make movies with a nature (as opposed to nurture) perspective.

Until now.  Or... maybe, if I can find an agent for my new screenplay.  The working title is "GINA" after the Genetic Information Nondiscrimination Act (GINA) signed into law in 2008.

Here's my pitch:  What if you applied for a promotion, and were told the traits for the job were "just not in your DNA"?  The protagonist in the screenplay sues his employer for genetic discrimination under the GINA law.

I've decided to submit the screenplay to the new Amazon Studios.

Myth: Human behavioral traits are 50% genetic and 50% environment

The latest “common wisdom” goes something like this: While our genes have an impact on our behavior, the environment is still responsible for half the behavioral variability between us. This feel-good, politically-correct compromise allows social scientists to appear reasonable and cede some ground, while claiming that genes aren’t everything.

Unfortunately, it’s not true, and here's why:

First, genes encapsulate experience from the evolutionary past. They transport knowledge from the past to the present. Thus genes interacting with environment are better considered as past environment interacting with present environment.

Second, as we develop, our genes can assume the presence of a certain environment, like a kite whose design can assume the presence of wind to realize its full shape and function. Genes are economical; if something can be assumed to exist in the environment, genes can simply exploit it, and not bother to represent it.

Consider a toaster. Its design anticipates a certain environment (i.e. someone to press its lever and start the toasting process). Does this mean the result (i.e. toasted bread) is 50% explained by nature (the design of the toaster, it's DNA) and 50% by nurture (its experience)? Of course not. In general, when an object’s design (nature) anticipates a very specific environment (nurture), the variability in behavior outcome is 100% explained by the scenarios manifest in its design.

Humans are more complex than toasters, of course. Human behavior is driven by thousands of such mechanisms (innate modules). But the quantity of gears in our head does not invalidate the principle itself. We only remember things - or respond to things - for which we were designed.

Here's a more complex example: Say our genes wanted to construct a module in our brain that can recognize that someone is our mother. These brain genes could exploit the fact that any face-like pattern that appears with regularity when we cry is probably our mother. And if a face-like pattern doesn't appear when we cry, that means a no-mother scenario exists (fairly common in evolutionary history), and a different brain module can be constructed.

How to create an intelligent robot

I'm interested in how our genes construct the brain to guide our behavior.  What better way to study this than by simulating "behavioral genetics" in robots?

To construct an intelligent robot, you need to provide it with the following:

• basic senses and prime directives
• ability to recognize patterns in its senses (environment)
• ability to predict future patterns, and use prediction to drive its own behavior

The robot's senses represent what's happening both outside and inside itself.  If it sees food, that's external.  If it moves its muscles or feels full after eating, that's internal.  The robot's prime directive tell it "what's important", and we'll assume "stomach fullness" is an important outcome of its behavior.

First, the robot must recognize patterns in its senses.  Stacked neural networks (and eventually memristors) can be used for this purpose.  The robot's senses train the network, and, once trained, the network recognizes patterns, even those not identical with the training set.

Second, the robot must predict future patterns.  This is not as hard as it seems.  When trained in the previous step, the stacked neural network should be presented not only with the present state of the senses, but also with several past states, simultaneously.  In other words, the neural network makes no distinction between past and present, as it associates inputs from its external senses and its internal senses (e.g. muscle tension and how full is its stomach).

We can simulate all this virtually in a computer program (no need to build a physical robot) -

• Let's assume that "food" falls randomly from the sky, and obeys certain simplified laws of physics, such as the steady rate of fall (one row per clock tick).  A single piece of food appears at a time, and it is dropped at a random angle (diagonal left, right, or straight) from the top row.  Food bounces off the sides but never bounces upward.  The robot's mouth can only move one space at a time (left or right, on the bottom row) per clock tick.  Only if the mouth arrives at the same position as the food (when it reaches the bottom row) can the robot eat it.
• From the diagram, you can see that the robot has (8 x 8) + 8 = 72 inputs.  The external inputs (the visual sense, an 8 x 8 grid) take up most of the inputs, but there are also internal inputs (shown on the bottom row) such as: the position of the mouth (used to catch falling food), incremental mouth movement (whether the left or right mouth muscle is contracted), and how full is the robot's stomach.

How can we make the robot learn?  By training its neural network brain using the sensory inputs.  But not only current sensory inputs: we must combine the present inputs and past inputs into a single neural network.  If we assume the robot can remember the inputs from each of the last 8 clock ticks, the total number of inputs presented to the network will be 72 x 8  = 576 inputs.  Past and present are thus combined into a single network.  At the next clock tick, the oldest 72 "remembered inputs" are dropped, and the current 72 are added to the new training set.

The prime directive is hard-wired into the robot.  It simply states that the robot will be "happier" if its stomach is full.  We should assume a certain rate of digestion, whereby the stomach becomes less full over time unless it consumes more food.  Obviously, if the robot can be trained to move its mouth muscles back and forth to catch food from the sky, it will achieve greater "happiness".  If it doesn't learn this quickly enough, it will starve.

At first, the robot moves its mouth muscles randomly, back and forth.  But eventually, given the high weight (importance) assigned to the prime directive input (i.e. stomach fullness), the robot eventually learns to predict how to move its mouth left or right (one space per clock tick) to anticipate where the food will land a few ticks in the future.

Most interestingly, the robot uses prediction (or anticipation) to guide its own muscles.  It predicts where its mouth will be in the future, and this very act of prediction "causes" its muscles to contract and its mouth to move to catch the food (or play ping pong).

This simulated robot, although simplistic, demonstrates a primitive form of behavioral genetics.  Humans are more complex, and our hard-wired "prime directives" are subtle (and no two people have exactly the same prime directives), but the principle should be the same.

HIV Forensics

DNA tools can now trace the path of HIV infection back to the source, using natural variation in the genetic sequence of the virus as a unique fingerprint. 

According to researchers, the technique could also be used "in the identification of food-borne diseases where identifying the source of contamination could be very important".

What does it mean for a trait to be genetic?

I often say 'leadership is innate' or 'leadership traits are genetic'.  But what does it really mean for a trait to be genetic?

It clearly doesn't mean you're a leader from the moment you're born.  Nobody takes orders from a baby.  Leadership takes time to develop.  So 'leadership is genetic' is shorthand for saying 'the development of leadership traits over time is genetic'.

Yet not everyone is a leader.  Some people crave power, attention (status), and resources.  But others simply want to be followers.  So 'leadership is genetic' is also a shorthand way to say 'some people have the natural motivation to lead' and others have an interest in following, due to genetic differences.  Social skills are also genetic and vary across the population.

Richard Arvey and Stephen Colarelli posted a nice paper about the biology of leadership that I'll refer to next.

Leaders (monkeys and college fraternity presidents) have higher levels of the chemical serotonin in their brains than do followers.  So 'leadership is genetic' also means that our genes direct the manufacture of certain chemicals in the brain that correlate with leadership.  The more of a leader you are, the more serotonin is manufactured in your brain. Having high levels of serotonin is correlated with 'social ease' and 'calmer behavior in problem solving situations' and conflict management.  Brain chemicals can have a profound effect on behavior, because the brain is designed (by the genes) to be affected by these chemicals.

OK, so serotonin is correlated with leadership.  But does it actually cause leadership?  In some cases, yes.  Again, quoting from the paper:  Scientists conducted 'experiments where serotonin levels were altered with Prozac ... In leaderless all-male groups, monkeys that received Prozac emerged as alpha males'.

Since serotonin is a simple molecule, there's obviously no inherent 'leadership' in the chemical itself.  So artificially raising serotonin levels must trigger a capability that already lies dormant in the brain. Any gene variant that regulates production of more serotonin is essentially a leadership 'master' gene that exploits this pre-existing (genetically constructed) brain circuitry.

Let's step back for a moment.  Clearly the brain has the ability to recognize when you've attained a leadership role, otherwise the physical production of serotonin wouldn't have an 'on switch'.  In other words 'leadership is genetic' is another way of saying that genes develop a region of the brain that recognizes (via your senses) that you're dominant, and upon that recognition, manufactures and distributes serotonin in your bloodstream.  A different set of genes constructs another brain region that detects serotonin in the bloodstream, and triggers the interest, motivation and calm demeanor.

This brings us back to the question, can anyone become a leader?  Do your unique gene variants cause you to become a leader?  Or do you become a leader first (by having the right mix of qualities) and then innately realize your status, which causes physiological changes (like elevated serotonin levels).

My guess is that correlation and causation are both at work.  Some people are innately interested in acquiring status, navigating power hierarchies, and grabbing resources.  Such interests emanate from the more primitive amygdala region of the brain when you are younger, and the specific skills you acquire from practicing those interests (in these modern times) are stored in the cortex, which is a more general purpose learning engine.  Again, 'leadership is genetic' because your innate interests - which derive from genetically constructed regions of the brain - motivate you to practice and acquire the culturally proscribed skills and expertise to wield leadership. Leaders are thus selected (not trained) for their innate traits and interests, even as their skills have a modern flavor.

But 'born leaders' don't always succeed, and once they recognize their loss of stature, their serotonin level decreases by the natural process.  On the other hand, some people are appointed (or anointed) as leaders through little effort of their own (like hereditary kings and emperors), and once their innate brain circuitry detects this status increase, their serotonin level goes up.

A challenge for all you computational neuroscientists out there!  How are interests and motivations manifest in the brain?  How can our genes guide the development of brain circuitry that detects we are in a leadership position, and, upon detection, reduces (or encapsulates) that experience into a simple signal - the serotonin molecule?

Guilt by genetic association

Interesting story in the NY Times:

Recently, forensic scientists in California used a genetic analysis procedure called “familial searching” or “kinship searching” to help the police identify a suspect in the “Grim Sleeper” serial murder case — and they did so by using a DNA sample collected for another purpose from the suspect’s own son.

Lo and behold, [the suspect's] son had recently been convicted on a felony weapons charge, and his DNA offered a partial match to crime-scene DNA. And for the first time in California, that kind of one-degree-of-separation search ultimately led to an arrest.

The risk is that the police, while looking for a suspect’s family members, might intrude on people who have not committed a crime. Some lawyers call it guilt by genetic association.

Harvard creates genetic test to identify entrepreneurs

Harvard researchers recently announced a psychometric test that identifies individuals with the traits that make for successful entrepreneurs in developing economies.

Yet, as we have seen, such traits are genetically linked and can't be trained.  You develop certain traits because of the genes you inherit.

Knowing the specific genes behind such traits is not necessary to establish a genetic basis.  For example, it's illegal (in America) to ask someone's family history of heart disease in a job interview, because family history is a proxy for genes passed on from one generation to the next.

Psychometric tests are also genetic proxies.  In other words, they're genetic tests.

Should you test your genes?

CNN asks the question: "Should you test your genes?"  But I think it should be treated as two separate issues:

1. Do you have the right to know your genes (i.e. the specific gene variants you possess)?
2. Who is allowed to make claims on gene function?

Certainly, it's everyone's right to know their own genes.  That's what the Human Genome Project was all about, and several individuals like James Watson and Glenn Close have had their own genomes fully sequenced for $100,000 or more. Personal genetics services (like 23andMe) offer an affordable service for the rest of us, by sampling a much smaller amount of genetic differences between individuals.

Yet this individual right to know is being challenged for the following reason: Vendors who provide gene tests are also making claims on the results, e.g. you may have a 1% increased risk of cancer because of the genes you inherited.  For this reason, the FDA is blocking all personal genetic testing.

This decision clearly infringes on an individual's right to know our own genes, which are objective facts about us (our unique and individual fingerprint, responsible for human diversity).  Rich people like Glenn Close are allowed to know their gene sequence, but not the rest of us?

Thomas Goetz, who is just as outraged as I am, writes: "The controversy here rests on a single assumption: That the typical ... American isn't prepared to learn about their DNA ... We can't handle the truth."

Yet there is a solution: Don't treat them as gene "tests".  Separate the identification of gene variants (objective facts about you) from the claims being made on them.  Perhaps partition these activities, so no single entity or private company can do both.  Once you know your gene sequence, you can do your own research on what your individual gene variants really mean, in a grand marketplace of ideas.

But don't take away our right to know our own genes, under the guise of consumer protection.