Times of Israel on Polygenic Embryo Screening

This is a very nice article on polygenic embryo screening and its prospects in Israel. Worth reading in full. Leading statistical geneticist Shai Carmi is interviewed. See also his interview on Manifold (embedded player at bottom of this post).

The Times of Israel 

14 October 2022, 1:27 pm 

Designer babies? Hi-tech preimplantation genetic testing may soon come to Israel 

For generations, the Yu family of Shanghai has suffered from type 2 diabetes. But this summer, as reported in the China Daily, the family welcomed a baby with a better chance of avoiding this disease.

These rosier prospects are the result of a recent breakthrough in assisted reproduction that was advanced with the help of Israeli scientists, called preimplantation genetic testing for polygenic diseases (PGT-P). In addition to China, PGT-P is also gaining ground among couples in the United States who wish to improve health outcomes for their future children. But in Israel, it is illegal.

PGT-P is carried out on an embryo during in vitro fertilization (IVF), prior to its transfer from the Petri dish to the womb. Viable embryos with the probable lowest disease risk can then be selected for implantation.

Since this innovative testing takes into account a complex combination of factors that are not broached in more traditional testing, in some ways it’s almost like an educated guess. Accordingly, polygenic screening is not a diagnosis: It is a prediction of relative future risk compared to other people.

Israeli academics have published peer-reviewed research advancing the science behind polygenic screening, including Shai Carmi, Ehud Karavani, Or Zuk, Gil Atzmon, and Einat Granot-Hershkovitz.

But Start-Up Nation is not yet implementing this cutting-edge tech in the field of fertility. Although fertility treatments are subsidized by the Israeli government, it is still unclear whether Israeli couples ever will have access to the procedure, which screens for polygenic diseases such as diabetes, heart disease and cancer — or whether they would even want it.

PGT-P is different from prior technology in important ways, creating new opportunities and challenges for parents while raising profound ethical dilemmas for society. Similar to older forms of testing, PGT-P relies on analyzing genetic material from embryos created through IVF before implantation and checking them for certain diseases and conditions. The information then helps the parents and doctors decide which embryos to implant.

However, the biggest difference between PGT-P screening and earlier forms of genetic testing is that the prior tests checked for genetically simple conditions such as Down syndrome, cystic fibrosis, or Tay Sachs disease. These diseases, which are serious or fatal, have extremely high “penetrance,” which means that if the gene mutation is seen in the embryo’s DNA, it is nearly certain that the child will have that condition. The appearance of the disease-linked gene is the basis of a clear diagnosis.

This “simple” genetic screening has already borne fruit in the Jewish community: Decades ago, babies in the Ashkenazi Jewish community were nearly 100 times more likely to be born with Tay Sachs than babies in the general US population. Today, because of genetic screenings, the disease is “virtually wiped out.”

In contrast, PGT-P screening can’t tell you with assurance if an embryo will develop a genetic disease such as cancer or Crohn’s disease. That’s because this new screening checks for polygenic diseases – complex conditions caused by the combined impact of possibly thousands of different genes, as well as lifestyle and other environmental factors.

Instead of a clear diagnosis, prospective parents receive a “polygenic risk score,” basically the probability of a child developing a certain disease or condition.

Noa and her husband went through 10 IVF cycles to build their family. “We now have two wonderful boys,” she says.

If I had an opportunity to reduce disease risk in my kids, I would do it

She knows what she would have said if doctors had offered her polygenic screening: “I want that technology.” As a speech therapist who works with kids facing a lot of health challenges, she was very worried about what her own kids would face.

“If I had an opportunity to reduce disease risk in my kids, I would do it. It would definitely help my peace of mind as a mother. Everyone here in Israel should have the option of using it,” Noa says.

No clear-cut answers

Scientists at the US-based Genomic Prediction, Inc. published an article in 2019 describing the “first clinical application” of polygenic screening of embryos. Genomic Prediction is a polygenic screening company based in New Jersey that partners with various IVF clinics around the world.

However, to date, the Israeli Health Ministry has yet to even issue a statement on the use of polygenic screening on embryos.

For some, the fact that PGT-P screening isn’t available, or even legal, in Israel is somewhat counterintuitive, given Israel’s prominence in the fields of both assisted reproduction and genetic testing.

Israelis undergo more rounds of IVF per capita than any other nation in the world. This is largely due to religious and cultural norms that are highly supportive of child-bearing, combined with the nationally financed healthcare system that provides full coverage for as many IVF cycles as needed, up to two children per family.

In addition, “in Israel there is a lot more openness to preimplantation genetic testing in general because of the high prevalence of various disease mutations in our community,” says Carmi, an associate professor at the Hebrew University School of Public Health and Faculty of Medicine.

Today, Carmi is a leading researcher on the accuracy of polygenic screening. As part of his post-doctoral project at Columbia University in New York, he helped generate important genetic sequencing data for Ashkenazi Jews.

Israel’s embrace of most genetic testing is reflected in the Israeli Health Ministry’s website, which lists dozens of recommended genetic screenings, broken down by ethnic sub-community. But these screenings are for monogenic disorders, easily diagnosed by looking for a single gene mutation.

“In Israel, the Health Ministry controls which diseases can be screened for, and candidate variants need to have high penetrance and lead to diseases with severe symptoms,” says Carmi.

Playing the odds

Miri is a consultant originally from central Israel. Although she did not have any known fertility problems, she chose to undergo IVF specifically because it would allow her to screen for a certain hereditary disease. She and her husband are both carriers of a rare mutation, so a natural conception meant a 25 percent chance of the fetus suffering from this generally fatal condition.

“For me, it was a choice between the extra physical hardship of IVF, or the extra emotional hardship of a pregnancy where, for months, we would not know if the baby would have this disease,” Miri said.

In contrast, PGT-P can’t provide conclusive information, because in the context of polygenic diseases like diabetes and heart disease “nothing is deterministic,” says Carmi.

According to Carmi, a child may develop the condition or may not, and non-genetic factors can certainly affect the outcome. Based on his peer-reviewed research on statistical modeling of polygenic screening, though, Carmi notes that “you can get quite a substantial risk reduction.”

The “relative risk reduction” projected to be accomplished by PGT-P varies depending on the disease. However, according to a 2021 research paper by Carmi and his collaborators, for schizophrenia and Crohn’s disease, around a 45% relative risk reduction is achievable for parents testing five embryos and choosing the best scoring, compared to implanting a randomly chosen one of the five.

The testing, of course, comes with a fee: Costs vary, but Genomic Prediction in New Jersey charges a $1,000 up-front fee, plus $400 per embryo analyzed. Of course, this is an add-on cost for people already doing IVF, which in the US can cost up to tens of thousands of dollars per cycle.

Pricing can get even more complicated, however, because different services end up bundled together, or are offered as add-ons once related costs are already accounted for. But one of the earlier forms of embryonic screening (PGT-A, which checks for aneuploidies, giving rise to Down syndrome for example) can cost several thousand dollars.

By contrast, carrier screening, which is a blood test performed on the parents to check for “simple” monogenic-disease carrier status, costs only several hundred dollars, and is often also covered by insurance.

In Israel, for couples whose family history or carrier-screening blood tests reveal a heightened risk for having children with a specific monogenic disease, the Health Ministry promotes the benefits of traditional genetic testing of embryos prior to implantation in the womb.

According to its website, “Pre-implantation Genetic Diagnosis (PGD) is today considered to be one of the practical options for couples who are at high risk for giving birth to a baby with a chromosomal abnormality or a genetic disease. This is because the process allows pregnancies to be achieved with healthy fetuses, and avoids the need for pregnancy termination, a procedure that constitutes a problem for many couples for religious, ethical and/or moral reasons.”

The nuts and bolts

PGT-P was developed using artificial intelligence technology applied to huge databases containing the genetic and health information of hundreds of thousands of people. Statistical data analysis of DNA and health outcomes allows scientists to see which complex genetic patterns more frequently show up in individuals who also develop a certain disease, such as schizophrenia. By genetically analyzing an embryo and then comparing its genetic information to this population data, the embryo’s polygenic risk score can be calculated for a given disease. This can already be done for a great many common diseases, with varying levels of predictive power, and as genetic databases grow, the reliability of these risk scores will continue to improve.

The couple also receives the raw data about their embryos’ genes and risk scores, so if they prefer to implant the embryo with the lowest risk of type 2 diabetes rather than the lowest combined disease risk, they can do that

“For prospective parents undergoing IVF and electing to use polygenic screening, somewhere between 10 and 20 polygenic risk scores are combined in a weighted average, with more serious diseases given greater weight in the final figure. This averaging provides a single number for each embryo — a health index — that can be used to rank the available embryos, so that the one with the best health index can be implanted,” says Carmi.

“Of course, the couple also receives the raw data about their embryos’ genes and risk scores, so if they prefer to implant the embryo with, let’s say, the lowest risk of type 2 diabetes, rather than the lowest combined disease risk, they can do that,” says Carmi.

An emotional decision

Michal Amrani, 32, lives in the central Israeli town of Ramat Hasharon and is working toward a master’s degree in chemistry from the Weizmann Institute. Through a four-year IVF process, she and her husband Sarel welcomed a son, and later, a set of twins. They say they are unlikely to use polygenic screening, even if it becomes available in Israel.

“As it is, we opted not to do some of the genetic testing that was already available to us,” Amrani says. “I work in science, so I am more open to these things, but my husband doesn’t really like all these genetic tests. For him, there’s risk in lots of things, and his optimistic nature helps him be comfortable that things will work out.”

Others, like Noa, are more interested in trying out preimplantation polygenic screening of their embryos, but even if Israel would change its rules to allow it, it’s a tricky issue. First, there are concerns about the psychological difficulties that this technology may pose for prospective parents.

Rona Langer Ziv is a social worker and cognitive behavioral psychotherapist who counsels IVF patients — both couples and singles — at a large Israeli hospital, as well as through her private clinical practice.

“Due to the potential implications of this new technology,” she says, “I would be concerned about a higher risk for depression and anxiety among the IVF patients.”

“Even if they feel they understand what they are signing up for at the beginning of the journey, they may not appreciate the emotional, ethical, and psychosocial implications of polygenic screening several IVF cycles down the road,” says Langer Ziv. “They may find themselves worrying that the embryos’ scores are not good enough, or that they won’t have any viable embryos left to choose from.”

Even if they feel they understand what they are signing up for at the beginning of the journey, they may not appreciate the emotional, ethical, and psychosocial implications of polygenic screening several IVF cycles down the road

Because polygenic screening predicts relative risk rather than providing an affirmative disease diagnosis, “women, especially those over 40 who may have very few embryos to work with, end up facing a serious dilemma — they may be discarding an embryo that could have resulted in a healthy child,” says Langer Ziv.

Amrani is in a similar situation. She and her husband are ready for more kids, but right now they have just one embryo available, so that’s the embryo they will try to implant. Even though she won’t be using polygenic screening, Amrani says that “it does sound very innovative. It’s nice that there’s something like this.”

Social worker and cognitive behavioral psychotherapist Rona Langer Ziv. (Courtesy) Indeed, Langer Ziv acknowledges that some people would find polygenic screening appealing, particularly those with higher education levels.

“There’s definitely coolness in the technology. It’s scientifically advanced, and it could offer interesting health insights about your future children. Everyone would theoretically like to use a technology that potentially predicts a more healthy child, although there is disagreement among fertility specialists about the benefits involved,” Langer Ziv says.

“And for some IVF patients, it might also provide a feeling of control during a process that involves so much stress, uncertainty, luck, and randomness,” she says.

Risk of eugenics

Regardless of how polygenic screening would be received by potential consumers, there are grave concerns about the impacts of this new technology on society. Various ethical issues have been raised for decades about older forms of genetic screenings, including fears of stigmatizing those living with genetic diseases, and questions about equitable access to these technological advances.

Perhaps the most significant ethical concern, and one that looms larger with polygenic screening than with older tests for monogenic diseases, is the potential for eugenics. This is the infamous and dangerous philosophy, practiced in Nazi Germany and elsewhere, that society should try to promote the creation of the most genetically “superior” babies.

Miri and her husband now have a baby boy and are looking forward to having more children — they still have three embryos to choose from. Asked whether she would be interested in polygenic screening if it became available in Israel, Miri says she’s unsure.

“I would love to see less suffering in the world from diseases. But where do we draw the line?” she says.

Indeed, the potential for eugenics is most stark when screenings cross over from the realm of disease prevention to the world of intelligence and aesthetic traits such as height or eye color. As such, some laboratories preemptively claim they will only screen for health concerns: An American polygenic screening company currently states that it does not test for “high IQ,” nor for “purely cosmetic traits such as hair color and eye color.”

But complicating the “noble” stance, genetic researchers have shown that “IQ is negatively correlated with most psychiatric disorders [and] positively correlated with autism and anorexia,” meaning that a high IQ comes with a lower risk of most psychiatric diseases and a higher risk of certain other neurological and mental health conditions.

As such, while some companies may currently refuse to offer IQ screening, it is not hard to imagine a health-based argument for loosening such protocols in the future, particularly as society becomes more used to the practice of PGT-P.

Similarly, a large study was published this year by researchers at Brown University and Peking University that found that “light eye colors were associated with high risks” of certain forms of skin cancer. Again, one can picture checks for eye color making their way into future genetic screenings through a backdoor of disease relevance.

In Carmi’s view, the responsible way for Israel to approach the prospect of polygenic screening is a gradual one.

“Ideally, we would start by recruiting Israeli participants for local academic research, with oversight by the Health Ministry,” Carmi says. “Once we develop more insight into how predictive polygenic screening is in our population, the relevant stakeholders — including patients, professional organizations, and regulators — can balance competing interests and local values, and come up with tailored guidance on its use in Israel.”

For some Israeli citizens, of particular concern is the idea of the wealthy trying to create perfect babies.

“If polygenic screening came to Israel, I would want to see a lot of regulation about who gets to use it, how it is used, and what reasons it is used for,” Miri says.

See also

WIRED: Genetic Screening Now Lets Parents Pick the Healthiest Embryos 

Genomic Prediction in Bloomberg

The Future of Human Evolution -- excerpts from podcast interview with Brian Chau

The transcript excerpts below are from my recent conversation with Brian Chau (aka Cactus Chu) on his podcast.

Segments begin at 47m and 1h07m.

They have been lightly edited.https://t.co/RJ68FasfJchttps://t.co/TnSuNaxhnQ

— steve hsu (@hsu_steve) September 28, 2022

Full excerpts.

1. The prospect of predicting cognitive ability from DNA, and the consequences. Why the main motivation has nothing to do with group differences. This segment begins at roughly 47 minutes. 

2. Anti-scientific resistance to research on the genetics of cognitive ability. My experience with the Jasons. Blank Slate-ism as a sacralized, cherished belief of social progressives. This segment begins at roughly 1 hour 7 minutes. 

1. Starts at roughly 47 minutes. 

Okay, let's just say hypothetically my billionaire friend is buddies with the CEO of 23andMe and let's say on the down low we collected some SAT scores of 1M or 2M people. I think there are about 10M people that have done 23andMe, let's suppose I manage to collect 1-2M scores for those people. I get them to opt in and agree to the study and da da da da and then Steve runs his algos and you get this nice predictor. 

But you’ve got to do it on the down low. Because if it leaks out that you're doing it, People are going to come for you. The New York Times is going to come for you, everybody's going to come for you. They're going to try to trash the reputation of 23andMe. They're going to trash the reputation of the billionaire. They're going to trash the reputation of the scientists who are involved in this. But suppose you get it done. And getting it done as you know very well is a simple run on AWS and you end up with this predictor which wow it's really complicated it depends on 20k SNPs in the genome ... 

For anybody with an ounce of intellectual integrity, they would look back at their copy of The Mismeasure of Man which has sat magisterially on their bookshelf since they were forced to buy it as a freshman at Harvard. They would say, “WOW! I guess I can just throw that in the trash right? I can just throw that in the trash.” 

But the set of people who have intellectual integrity and can process new information and then reformulate the opinion that they absorbed through social convention – i.e., that Gould is a good person and a good scientist and wise -- is tiny. The set of people who can actually do that is like 1% of the population. So you know maybe none of this matters, but in the long run it does matter. … 

Everything else about that hypothetical: the social scientists running the longitudinal study, getting the predictor in his grubby little hands and publishing the validation, but people trying to force you to studiously ignore the results, all that has actually already happened. We already have something which correlates ~0.4 with IQ. Everything else I said has already been done but it's just being studiously ignored by the right thinking people. 

 … 

Some people could misunderstand our discussion as being racist. I'm not saying that any of this has anything to do with group differences between ancestry groups. I'm just saying, e.g., within the white population of America, it is possible to predict from embryo DNA which of 2 brothers raised in the same family will be the smart one and which one will struggle in school. Which one will be the tall one and which one will be not so tall. 

[more]

2. Starts at roughly 1 hour 7 minutes. 

I've been in enough places where this kind of research is presented in seminar rooms and conferences and seen very negative attacks on the individuals presenting the results. 

I'll give you a very good example. There used to be a thing called the Jasons. During the cold war there was a group of super smart scientists called the Jasons. They were paid by the government to get together in the summers and think about technological issues that might be useful for defense and things like war fighting. … 

I had a meeting with the (current) Jasons. I was invited to a place near Stanford to address them about genetic engineering, genomics, and all this stuff. I thought okay these are serious scientists and I'll give them a very nice overview of the progress in this field. This anecdote takes place just a few years ago. 

One of the Jasons present is a biochemist but not an expert on genomics or machine learning. This biochemist asked me a few sharp questions which were easy to answer. But then at some point he just can't take it anymore and he grabs all his stuff and runs out of the room. ...

[more]

"The Pressure to Conform is Enormous": Steve Hsu on Affirmative Action, Assimilation and IQ Outliers (CSPI Podcast with Richard Hanania)

 

Another great conversation with Richard Hanania. 

Some rough timestamps: 

Begin: American society, growing up as child of immigrants 

18m: Russia-Ukraine conflict (eve of invasion), geopolitical implications (China, India, Germany, EU) 

38m: Affirmative Action, Harvard case at SCOTUS 

54m: Woke leftists at the university, destruction of meritocracy, STEM vs Social Justice advocacy, Sokal Hoax 

1h25m: Academic economics, 2008 credit crisis, Do economists test theories? 

1h33m: Maverick thinking, Agreeableness, Aspergers, Pressure to conform 

1h39m: Far-tail intelligence, Jeff Bezos and physics, progress in science and technology

Full transcript at Richard's substack.

BOLA2 Copy Number Variation: Phenotype Effects From A Human Accelerated Region

Human Accelerated Regions (HARs) are regions of DNA that were conserved throughout prior (e.g., vertebrate) evolution but are significantly different in the human genome.

Allen Institute: ... of the known 3,171 human accelerated regions, 99 percent of these human-specific mutations fall into "non-coding" regions of DNA, or regions of DNA that don't contain instructions for making a protein. Many of them are in stretches of our genome known as enhancers, regions which regulate nearby genes, and about half of those are nestled in enhancers that are active in the developing human brain.

Our analysis of DNA regions used in predictors for common diseases and complex human traits found that large portions of phenotype variance reside in non-coding regions. This has important consequences for pleiotropy and for our understanding of genetic architecture. 

Regarding HARs, in a 2013 post Neanderthals Dumb? I wrote:

This figure is from the Supplement (p.62) of a recent Nature paper describing a high quality genome sequence obtained from the toe of a female Neanderthal who lived in the Altai mountains in Siberia. Interestingly, copy number variation at 16p11.2 is one of the structural variants identified in a recent deCODE study as related to IQ depression; see earlier post Structural genomic variants (CNVs) affect cognition.

From the Supplement (p.62):

Of particular interest is the modern human-specific duplication on 16p11.2 which encompasses the BOLA2 gene. This locus is the breakpoint of the 16p11.2 micro-deletion, which results in developmental delay, intellectual disability, and autism5,6. We genotyped the BOLA2 gene in 675 diverse human individuals sequenced to low coverage as part of the 1000 Genome Project Phase I7 to assess the population distribution of copy numbers in homo-sapiens (Figure S8.3). While both the Altai Neandertal and Denisova individual exhibit the ancestral diploid copy number as seen in all the non-human great apes, only a single human individual exhibits this diploid copy number state.

Modern humans typically have many (e.g., 3-10) copies of BOLA2. In Neanderthals and apes, 2 copies. 

Variation in copy number presumably affects gene expression, even if the actual protein (coding base pairs) structure is not changed. There may be other mechanisms at work, of course.

Mutations in this 16p11.2 region are associated with schizophrenia, autism, brain size, reduced IQ, anemia, and other things. 

Since 2013 a number of papers have investigated the phenotype effects of BOLA2 copy number variation (CNV) and/or the 16p11.2 duplication/deletion. The latter is more complex as it affects multiple genes in addition to BOLA2. In the future, using whole exome or whole genome data in UKB, it should be possible to focus more specifically on effects of BOLA2 CNV.

For reference I note some of the results below.

Phenome-wide Burden of Copy-Number Variation in the UK Biobank (2019) 

16p11.2 C deletion: "We observe significant increases, on the order of one standard deviation, in weight, BMI, hip and waist circumference, reticulocyte count, and Cystatin C measures for these individuals. The larger 593 kb CNV associates with similar measures of body size and fat, as well as hypertension, diabetes/HbA1c, and abdominal hernia. These results are also indicative of effects due to developmental delay; namely, decreased measures of memory, higher Townsend deprivation (an index of material deprivation which considers employment, home/auto ownership, and household overcrowding in a person's neighborhood) ..."   

Note the effect sizes, e.g., on Townsend deprivation index, are extremely large, roughly 1 SD. The effect size for Prospective Memory score (related to ability to read, remember, and execute directions) is 2 SD!

 

 

Medical consequences of pathogenic CNVs in adults: analysis of the UK Biobank (2019)

Population percentage in parenthesis: 

See also:

The Human-Specific BOLA2 Duplication Modifies Iron Homeostasis and Anemia Predisposition in Chromosome 16p11.2 Autism Individuals (2019)

Quantifying the Effects of 16p11.2 Copy Number Variants on Brain Structure: A Multisite Genetic-First Study (2018)

Social and Educational Mobility: Denmark vs USA (James Heckman)

Despite generous social programs such as free pre-K education, free college, and massive transfer payments, Denmark is similar to the US in key measures of inequality, such as educational outcomes and cognitive test scores. 

While transfer payments can equalize, to some degree, disposable income, they do not seem to be able to compensate for large family effects on individual differences in development. 

These observations raise the following questions: 

1. What is the best case scenario for the US if all progressive government programs are implemented with respect to child development, free high quality K12 education, free college, etc.?

2. What is the causal mechanism for stubborn inequality of outcomes, transmitted from parent to child (i.e., within families)? 

Re #2: Heckman and collaborators focus on environmental factors, but do not (as far as I can tell) discuss genetic transmission. We already know that polygenic scores are correlated to the education and income levels of parents, and (from adoption studies) that children tend to resemble their biological parents much more strongly than their adoptive parents. These results suggest that genetic transmission of inequality may dominate environmental transmission.

  

See 

Genetic correlation of social outcomes between relatives (Fisher 1918) tested using lineage of 400k English individuals 

Game Over: Genomic Prediction of Social Mobility

The Contribution of Cognitive and Noncognitive Skills to Intergenerational Social Mobility (McGue et al. 2020)

Note: Denmark is very homogenous in ancestry, and the data presented in these studies (e.g., polygenic scores and social mobility) are also drawn from European-ancestry cohorts. The focus here is not on ethnicity or group differences between ancestry groups. The focus is on social and educational mobility within European-ancestry populations, with or without generous government programs supporting free college education, daycare, pre-K, etc.

Lessons for Americans from Denmark about inequality and social mobility 

James Heckman and Rasmus Landersø 

Abstract Many progressive American policy analysts point to Denmark as a model welfare state with low levels of income inequality and high levels of income mobility across generations. It has in place many social policies now advocated for adoption in the U.S. Despite generous Danish social policies, family influence on important child outcomes in Denmark is about as strong as it is in the United States. More advantaged families are better able to access, utilize, and influence universally available programs. Purposive sorting by levels of family advantage create neighborhood effects. Powerful forces not easily mitigated by Danish-style welfare state programs operate in both countries.

Also discussed in this episode of EconTalk podcast. Russ does not ask the obvious question about disentangling family environment from genetic transmission of inequality.

 

The figure below appears in Game Over: Genomic Prediction of Social Mobility. It shows SNP-based polygenic score and life outcome (socioeconomic index, on vertical axis) in four longitudinal cohorts, one from New Zealand (Dunedin) and three from the US. Each cohort (varying somewhat in size) has thousands of individuals, ~20k in total (all of European ancestry). The points displayed are averages over bins containing 10-50 individuals. For each cohort, the individuals have been grouped by childhood (family) social economic status. Social mobility can be predicted from polygenic score. Note that higher SES families tend to have higher polygenic scores on average -- which is what one might expect from a society that is at least somewhat meritocratic. The cohorts have not been used in training -- this is true out-of-sample validation. Furthermore, the four cohorts represent different geographic regions (even, different continents) and individuals born in different decades.

The figure below appears in More on SES and IQ.

Where is the evidence for environmental effects described above in Heckman's abstract: "More advantaged families are better able to access, utilize, and influence universally available programs. Purposive sorting by levels of family advantage create neighborhood effects"? Do parents not seek these advantages for their adopted children as well as for their biological children? Or is there an entirely different causal mechanism based on shared DNA?

 

The figure below appears in Low SES does not decrease heritability of cognitive ability (N=300k).

 

Kathryn Paige Harden Profile in The New Yorker (Behavior Genetics)

This is a good profile of behavior geneticist Paige Harden (UT Austin professor of psychology, former student of Eric Turkheimer), with a balanced discussion of polygenic prediction of cognitive traits and the culture war context in which it (unfortunately) exists.

Can Progressives Be Convinced That Genetics Matters? 

The behavior geneticist Kathryn Paige Harden is waging a two-front campaign: on her left are those who assume that genes are irrelevant, on her right those who insist that they’re everything. 

Gideon Lewis-Kraus

Gideon Lewis-Kraus is a talented writer who also wrote a very nice article on the NYTimes / Slate Star Codex hysteria last summer.

Some references related to the New Yorker profile:

1. The paper Harden was attacked for sharing while a visiting scholar at the Russell Sage Foundation: Game Over: Genomic Prediction of Social Mobility 

2. Harden's paper on polygenic scores and mathematics progression in high school: Genomic prediction of student flow through high school math curriculum 

3. Vox article; Turkheimer and Harden drawn into debate including Charles Murray and Sam Harris: Scientific Consensus on Cognitive Ability?

A recent talk by Harden, based on her forthcoming book The Genetic Lottery: Why DNA Matters for Social Equality

Regarding polygenic prediction of complex traits 

I first met Eric Turkheimer in person (we had corresponded online prior to that) at the Behavior Genetics Association annual meeting in 2012, which was back to back with the International Conference on Quantitative Genetics, both held in Edinburgh that year (photos and slides [1] [2] [3]). I was completely new to the field but they allowed me to give a keynote presentation (if memory serves, together with Peter Visscher). Harden may have been at the meeting but I don't recall whether we met. 

At the time, people were still doing underpowered candidate gene studies (there were many talks on this at BGA although fewer at ICQG) and struggling to understand GCTA (Visscher group's work showing one can estimate heritability from modestly large GWAS datasets, results consistent with earlier twins and adoption work). Consequently a theoretical physicist talking about genomic prediction using AI/ML and a million genomes seemed like an alien time traveler from the future. Indeed, I was.

My talk is largely summarized here:

On the genetic architecture of intelligence and other quantitative traits 

https://arxiv.org/abs/1408.3421 

How do genes affect cognitive ability or other human quantitative traits such as height or disease risk? Progress on this challenging question is likely to be significant in the near future. I begin with a brief review of psychometric measurements of intelligence, introducing the idea of a "general factor" or g score. The main results concern the stability, validity (predictive power), and heritability of adult g. The largest component of genetic variance for both height and intelligence is additive (linear), leading to important simplifications in predictive modeling and statistical estimation. Due mainly to the rapidly decreasing cost of genotyping, it is possible that within the coming decade researchers will identify loci which account for a significant fraction of total g variation. In the case of height analogous efforts are well under way. I describe some unpublished results concerning the genetic architecture of height and cognitive ability, which suggest that roughly 10k moderately rare causal variants of mostly negative effect are responsible for normal population variation. Using results from Compressed Sensing (L1-penalized regression), I estimate the statistical power required to characterize both linear and nonlinear models for quantitative traits. The main unknown parameter s (sparsity) is the number of loci which account for the bulk of the genetic variation. The required sample size is of order 100s, or roughly a million in the case of cognitive ability.

The predictions in my 2012 BGA talk and in the 2014 review article above have mostly been validated. Research advances often pass through the following phases of reaction from the scientific community:

1. It's wrong ("genes don't affect intelligence! anyway too complex to figure out... we hope")

2. It's trivial ("ofc with lots of data you can do anything... knew it all along")

3. I did it first ("please cite my important paper on this")

Or, as sometimes attributed to Gandhi: "First they ignore you, then they laugh at you, then they fight you, then you win.”

See also To Predict the Future it is useful to understand the Past.

Technical note: 

In 2014 I estimated that ~1 million genotype | phenotype pairs would be enough to capture most of the common SNP heritability for height and cognitive ability. This was accomplished for height in 2017. However, the sample size of well-phenotyped individuals is much smaller for cognitive ability, even in 2021, than for height in 2017. For example, in UK Biobank the cognitive test is very brief (~5 minutes IIRC, a dozen or so questions), but it has not even been administered to the full cohort as yet. In the Educational Attainment studies the phenotype EA is only moderately correlated (~0.3 ?) or so with actual cognitive ability.

Hence, although the most recent EA4 results use 3 million individuals [1], and produce a predictor which correlates ~0.4 with actual EA, the statistical power available is still less than what I predicted would be required to train a really good cognitive ability predictor.

In our 2017 height paper, which also briefly discussed bone density and cognitive ability prediction, we built a cognitve ability predictor roughly as powerful as EA3 using only ~100k individuals with the noisy UKB test data. So I remain confident that  ~million individuals with good cognitive scores (e.g., SAT, AFQT, full IQ test) would deliver results far beyond what we currently have available. We also found that our predictor, built using actual (albeit noisy) cognitive scores exhibits less power reduction in within-family (sibling) analyses compared to EA. So there is evidence that (no surprise) EA is more influenced by environmental factors, including so-called genetic nurture effects, than is cognitive ability.

A predictor which captures most of the common SNP heritability for cognitive ability might correlate ~0.5 or 0.6 with actual ability. Applications of this predictor in, e.g., studies of social mobility or educational success or even longevity using existing datasets would be extremely dramatic.

Enter the Finns: FinnGen and FINRISK polygenic prediction of cardiometabolic diseases, common cancers, alcohol use, and cognition

In 2018 Dr. Aarno Palotie visited MSU (video of talk) to give an overview of the FinnGen research project. FinnGen aims to collect the genomic data of 500k citizens in Finland in order to study the origins of diseases and their treatment. Finland is well suited for this kind of study because it is relatively homogenous and has a good national healthcare system.

Professor Aarno Palotie, M.D., Ph.D. is the research director of the Human Genomics program at FIMM. He is also a faculty member at the Center for Human Genome Research at the Massachusetts General Hospital in Boston and associate member of the Broad Institute of MIT and Harvard. He has a long track record in human disease genetics. He has held professorships and group leader positions at the University of Helsinki, UCLA, and the Wellcome Trust Sanger Institute. He has also been the director of the Finnish Genome Center and Laboratory of Molecular Genetics in the Helsinki University Hospital.

FinnGen is now producing very interesting results in polygenic risk prediction and clinical / public health applications of genomics. Below are a few recent papers.

1. This paper studies the use of PRS in prediction of five common diseases, with an eye towards clinical utility.

Polygenic and clinical risk scores and their impact on age at onset and prediction of cardiometabolic diseases and common cancers 

Nature Medicine volume 26, 549–557(2020) 

Polygenic risk scores (PRSs) have shown promise in predicting susceptibility to common diseases1,2,3. We estimated their added value in clinical risk prediction of five common diseases, using large-scale biobank data (FinnGen; n = 135,300) and the FINRISK study with clinical risk factors to test genome-wide PRSs for coronary heart disease, type 2 diabetes, atrial fibrillation, breast cancer and prostate cancer. We evaluated the lifetime risk at different PRS levels, and the impact on disease onset and on prediction together with clinical risk scores. Compared to having an average PRS, having a high PRS contributed 21% to 38% higher lifetime risk, and 4 to 9 years earlier disease onset. PRSs improved model discrimination over age and sex in type 2 diabetes, atrial fibrillation, breast cancer and prostate cancer, and over clinical risk in type 2 diabetes, breast cancer and prostate cancer. In all diseases, PRSs improved reclassification over clinical thresholds, with the largest net reclassification improvements for early-onset coronary heart disease, atrial fibrillation and prostate cancer. This study provides evidence for the additional value of PRSs in clinical disease prediction. The practical applications of polygenic risk information for stratified screening or for guiding lifestyle and medical interventions in the clinical setting remain to be defined in further studies.

2. This paper is a well-powered study of genetic influence on alcohol use and effects on mortality.

Genomic prediction of alcohol-related morbidity and mortality 

Nature Translational Psychiatry volume 10, Article number: 23 (2020) 

While polygenic risk scores (PRS) have been shown to predict many diseases and risk factors, the potential of genomic prediction in harm caused by alcohol use has not yet been extensively studied. Here, we built a novel polygenic risk score of 1.1 million variants for alcohol consumption and studied its predictive capacity in 96,499 participants from the FinnGen study and 39,695 participants from prospective cohorts with detailed baseline data and up to 25 years of follow-up time. A 1 SD increase in the PRS was associated with 11.2 g (=0.93 drinks) higher weekly alcohol consumption (CI = 9.85–12.58 g, p = 2.3 × 10–58). The PRS was associated with alcohol-related morbidity (4785 incident events) and the risk estimate between the highest and lowest quintiles of the PRS was 1.83 (95% CI = 1.66–2.01, p = 1.6 × 10–36). When adjusted for self-reported alcohol consumption, education, marital status, and gamma-glutamyl transferase blood levels in 28,639 participants with comprehensive baseline data from prospective cohorts, the risk estimate between the highest and lowest quintiles of the PRS was 1.58 (CI = 1.26–1.99, p = 8.2 × 10–5). The PRS was also associated with all-cause mortality with a risk estimate of 1.33 between the highest and lowest quintiles (CI = 1.20–1.47, p = 4.5 × 10–8) in the adjusted model. In conclusion, the PRS for alcohol consumption independently associates for both alcohol-related morbidity and all-cause mortality. Together, these findings underline the importance of heritable factors in alcohol-related health burden while highlighting how measured genetic risk for an important behavioral risk factor can be used to predict related health outcomes.

3. This paper examines rare CNVs (Copy Number Variants) and PRS (Polygenic Risk Score) prediction using a combined Finnish sample of ~30k for whom education, income, and health outcomes are known. The study finds that low polygenic scores for Educational Attainment (EA) and intelligence predict worse outcomes in education, income, and health.

Polygenic burden has broader impact on health, cognition, and socioeconomic outcomes than most rare and high-risk copy number variants 

https://www.nature.com/articles/s41380-021-01026-z 

Abstract Copy number variants (CNVs) are associated with syndromic and severe neurological and psychiatric disorders (SNPDs), such as intellectual disability, epilepsy, schizophrenia, and bipolar disorder. Although considered high-impact, CNVs are also observed in the general population. This presents a diagnostic challenge in evaluating their clinical significance. To estimate the phenotypic differences between CNV carriers and non-carriers regarding general health and well-being, we compared the impact of SNPD-associated CNVs on health, cognition, and socioeconomic phenotypes to the impact of three genome-wide polygenic risk score (PRS) in two Finnish cohorts (FINRISK, n = 23,053 and NFBC1966, n = 4895). The focus was on CNV carriers and PRS extremes who do not have an SNPD diagnosis. We identified high-risk CNVs (DECIPHER CNVs, risk gene deletions, or large [>1 Mb] CNVs) in 744 study participants (2.66%), 36 (4.8%) of whom had a diagnosed SNPD. In the remaining 708 unaffected carriers, we observed lower educational attainment (EA; OR = 0.77 [95% CI 0.66–0.89]) and lower household income (OR = 0.77 [0.66–0.89]). Income-associated CNVs also lowered household income (OR = 0.50 [0.38–0.66]), and CNVs with medical consequences lowered subjective health (OR = 0.48 [0.32–0.72]). The impact of PRSs was broader. At the lowest extreme of PRS for EA, we observed lower EA (OR = 0.31 [0.26–0.37]), lower-income (OR = 0.66 [0.57–0.77]), lower subjective health (OR = 0.72 [0.61–0.83]), and increased mortality (Cox’s HR = 1.55 [1.21–1.98]). PRS for intelligence had a similar impact, whereas PRS for schizophrenia did not affect these traits. We conclude that the majority of working-age individuals carrying high-risk CNVs without SNPD diagnosis have a modest impact on morbidity and mortality, as well as the limited impact on income and educational attainment, compared to individuals at the extreme end of common genetic variation. Our findings highlight that the contribution of traditional high-risk variants such as CNVs should be analyzed in a broader genetic context, rather than evaluated in isolation. 

From the paper:

 ... we compared the impact of CNVs to the impact of the PRSs for educational attainment [24], schizophrenia [25], and general intelligence [26] on general health, morbidity, mortality, and socioeconomic burden. We analyzed these effects in two cohorts: one sampled at random from the Finnish working-age population (FINRISK), the other a Finnish birth cohort (Northern Finland Birth Cohort 1966; NFBC1966). Both cohorts link to national health records, enabling analysis of longitudinal health data and socioeconomic status data over several decades. 

... we observed a clear polygenic effect on socioeconomic outcome with educational attainment and IQ PRS scores. Belonging to the matched lowest PRS extremes (lowest 2.66%) of educational attainment or IQ had an overall stronger impact on the socioeconomic outcome than belonging to most high-risk CNV groups, and a generally stronger impact on health and survival, with the exception of household income-associated CNVs. 

... odds for subsequent level of education were even lower at the matched lowest extreme of PRSEA (OR = 0.31 [0.26–0.37]) and PRSIQ (OR = 0.51 [0.44–0.60]).

... Rare deleterious variants, including CNVs, can have a major impact on health outcomes for an individual and are thus under strong negative selection. However, such variants might not always have a strong phenotypic impact (incomplete penetrance), and as observed here, can have a very modest—if any—effect on well-being. The reason for this wide spectrum of outcomes remains speculative. From a genetic perspective, one hypothesis is that additional variants, both rare and common, modify the phenotypic outcome of a CNV carrier (Supplementary Figs. 11 and 12). This type of effect is observable in analyzes of hereditary breast and ovarian cancer in the UK Biobank [40] and in FinnGen [41], where strong-impacting variants’ penetrance is modified by compensatory polygenic effects. 

... As stated above, the observed effect of polygenic scores was broader than that of structural variants. We observed strong effects in PRSs for intelligence and educational attainment on education, income and socioeconomic status. 

The Inheritors and The Grisly Folk: H.G. Wells and William Golding on Neanderthals

Some time ago I posted about The Grisly Folk by H.G. Wells, an essay on Neanderthals and their encounters with modern humans. See also The Neanderthal Problem, about the potential resurrection of early hominids via genomic technology, and the associated ethical problems. 

The Grisly Folk: ... Many and obstinate were the duels and battles these two sorts of men fought for this world in that bleak age of the windy steppes, thirty or forty thousand years ago. The two races were intolerable to each other. They both wanted the eaves and the banks by the rivers where the big flints were got. They fought over the dead mammoths that had been bogged in the marshes, and over the reindeer stags that had been killed in the rutting season. When a human tribe found signs of the grisly folk near their cave and squatting place, they had perforce to track them down and kill them; their own safety and the safety of their little ones was only to be secured by that killing. The Neandertalers thought the little children of men fair game and pleasant eating. ...

William Golding was inspired by Wells to write The Inheritors (his second book, after Lord of the Flies), which is rendered mostly (until the end, at which point the perspective is reversed) from the Neanderthal point of view. Both Wells and Golding assume that Neanderthals were not as cognitively capable as modern humans, but Golding's primitives are peaceful quasi-vegetarians, quite unlike the Grisly Folk of Wells.

The Inheritors 

Golding considered this his finest novel and it is a beautifully realised tale about the last days of the Neanderthal people and our fear of the ‘other’ and the unfamiliar. The action is revealed through the eyes of the Neanderthals whose peaceful world is threatened by the emergence of Homo sapiens. 

The struggle between the simple Neanderthals and the malevolent modern humans ends in helpless despair ... 

From the book jacket: "When the spring came the people - what was left of them - moved back by the old paths from the sea. But this year strange things were happening, terrifying things that had never happened before. Inexplicable sounds and smells; new, unimaginable creatures half glimpsed through the leaves. What the people didn't, and perhaps never would, know, was that the day of their people was already over."

See this episode of the podcast Backlisted for an excellent discussion of the book. 

I am particularly interested in how Golding captures the perspective of pre-humans with limited cognitive abilities. He conveys the strangeness and incomprehensibility of modern humans as perceived by Neanderthals. In this sense, the book is a type of Science Fiction: it describes a first encounter with Aliens of superior capability.

We are approaching the day when modern humans will encounter a new and quasi-alien intelligence: it may be AI, or it may be genetically enhanced versions of ourselves.

On a scientific note, can someone provide an update to this 2013 work: "... high quality genome sequence obtained from the toe of a female Neanderthal who lived in the Altai mountains in Siberia. Interestingly, copy number variation at 16p11.2 is one of the structural variants identified in a recent deCODE study as related to IQ depression"? Here is an interesting follow up paper: Nature 2016 Aug 11; 536(7615): 205–209.

   

Audiobook:

 

Wall Street Journal on the Twitter Mob Attack and MSU Moral Panic

I've added some comments in [ ..brackets.. ] to the article excerpt below. More background.

WSJ: ‘We are scientists, seeking truth,” Michigan State University physicist Stephen Hsu wrote in a 2018 blog post. “We are not slaves to ideological conformity.” That might have been too optimistic. Last week MSU’s president, Samuel L. Stanley Jr., yielded to a pressure campaign, based in part on that post, and asked Mr. Hsu to resign as senior vice president for research and innovation.

The trouble began June 10, when MSU’s Graduate Employees Union composed a lengthy Twitter thread denouncing Mr. Hsu as, among other things, “a vocal scientific racist and eugenicist.” The union claimed Mr. Hsu believes “in innate biological differences between human populations, especially regarding intelligence.”

[ This is obviously a very serious accusation. It is contradicted by PUBLIC REMARKS I have made in multiple interviews over the years, as well as in blog posts. ]

Mr. Hsu says these accusations “were made in bad faith.” Take that 2018 blog post, which responded to New York Times articles that, in his words, linked “genetic science to racism and white supremacy.” In it, he wrote: “All good people abhor racism. I believe that each person should be treated as an individual, independent of ancestry or ethnic background. . . . However, this ethical position is not predicated on the absence of average differences between groups. I believe that basic human rights and human dignity derive from our shared humanity, not from uniformity in ability or genetic makeup.” Mr. Hsu doesn’t work in this field but rejects the idea that scientists should categorically exclude the possibility of average genetic differences among groups.

[ To be precise, in THAT BLOG POST I am defending researchers in genomics, at universities like Oxford, Harvard, and Columbia (see papers cited there), who study signals of recent natural selection from computational analysis of many human genomes. In 2018 the New York Times had attempted to create a moral panic, labeling them as "racists" ... Again, this is not my area of science -- I am defending the right of others to do their research. ]

In a 2011 post, Mr. Hsu argued that standardized tests are predictive of the quality of graduate-school candidates. The post mentioned nothing about race, but the union imputed to him a belief “that lack of Black & Hispanic representation in higher ed reflects lower ability, despite evidence these tests negatively impact diversity.”

[ The reporter is correct: RACE IS NEVER MENTIONED IN THAT BLOG POST. It merely reviews published results from very large studies, showing the predictive power of standardized tests in graduate and professional school outcomes. These results have nothing to do with race. For example, professors on admissions committees might be interested in whether undergraduate GPA or GRE score are better predictors of success. This kind of analysis is necessary to answer that question. 

However, this topic is heavily politicized and very few people, even professors, are familiar with the scientific results. But experts in psychometrics and personnel selection all know that these tests have real world utility. ]

The union also faulted him for having “directed funding to research downplaying racism in bias in police shootings.” The MSU professor who conducted that work, psychologist Joe Cesario, tells me that “we had no idea what the data was going to be, what the outcome was going to be, before we did this study.” Mr. Cesario has collected evidence from a simulator and from real-world interactions between police and citizens. He concluded that “the nature of the interaction really matters the most, and officers were not more likely to be ready to shoot upon encountering a black versus white citizen.”

[ Several years ago Cesario was granted a rare opportunity to study police shootings and officer behavior in simulators in a large city. My office provided him with a small amount of funding to create realistic simulator video of police stops and other situations. This is an important topic to study if we want to understand and improve policing. ]

A June 3 op-ed in these pages cited Mr. Cesario’s work, and the MSU communications team highlighted the mention in the June 9 edition of their email newsletter, InsideMSU. The next day, the Graduate Employees Union denounced Mr. Hsu. By June 11, editors of the newsletter had apologized “for including the item and for the harm it caused.” Hundreds of MSU students and employees signed petitions calling for Mr. Hsu to be fired from the administration.

[ NO, YOU ARE NOT MISTAKEN: THIS IS ORWELLIAN. YES THIS REALLY HAPPENED TO A PROMINENT RESEARCHER ON OUR CAMPUS WHO IS DOING VERY IMPORTANT WORK. YES, THEY APOLOGIZED FOR CALLING ATTENTION TO HIS WORK. YES, IT IS CHILLING. ]

Mr. Hsu says he felt compelled to step down because he served at the pleasure of the president. But he thinks Mr. Stanley handled the matter badly. “The first action of the university should be to investigate, find the truth, and defend the person if the claims are false.” Mr. Hsu says MSU undertook no such investigation.

[ I am among a shrinking subset of the faculty that believes the pursuit of truth should be the core value of a university.

Almost 2000 PEOPLE SIGNED THE PETITION supporting me, including hundreds of professors from MSU and around the world. Many letters from prominent scientists were sent to the president -- my guess is about 30, SOME OF WHICH ARE PUBLIC. MSU conducted no investigation of the facts. ]

...

... “I don’t personally believe that kind of enforcing a higher conduct to administrators will necessarily chill faculty,” Mr. Bird says.

Mr. Cesario disagrees. It’s “bad or worse that they are doing this to an administrator,” he says. “If anybody should be allowed to explore all topics, speak on all topics, and go where the data leads them, it’s administrators.” He expects the activists who won Mr. Hsu’s dismissal won’t stop “pushing for a narrowing of what kinds of topics people can talk about, or what kinds of conclusions people can come to.” The number of administrators willing to defend scientific inquiry, Mr. Cesario adds, is “now down by one.”

Media coverage:

A Twitter Mob Takes Down an Administrator at Michigan State (Wall Street Journal June 25)

Scholar forced to resign over study that found police shootings not biased against blacks (The College Fix)

On Steve Hsu and the Campaign to Thwart Free Inquiry (Quillette)

Michigan State University VP of Research Ousted (Reason Magazine, Eugene Volokh, UCLA)

Research isn’t advocacy (NY Post Editorial Board)

Podcast interview on Tom Woods show (July 2)

College professor forced to resign for citing study that found police shootings not biased against blacks (Law Enforcement Today, July 5)

"Racist" College Researcher Ousted After Sharing Study Showing No Racial Bias In Police Shootings (ZeroHedge, July 6)

Twitter mob: College researcher forced to resign after study finding no racial bias in police shootings (Reclaim the Net, July 8)

Horowitz: Asian-American researcher fired from Michigan State administration for advancing facts about police shootings (The Blaze, July 8)

I Cited Their Study, So They Disavowed It: If scientists retract research that challenges reigning orthodoxies, politics will drive scholarship (Wall Street Journal July 8)

Conservative author cites research on police shootings and race. Researchers ask for its retraction in response (The College Fix, July 8)

Academics Seek to Retract Study Disproving Racist Police Shootings After Conservative Cites It (Hans Bader, CNSNews, July 9)

The Ideological Corruption of Science (theoretical physicist Lawrence Krauss in the Wall Street Journal, July 12)

Foundation for Individual Rights in Education: "chilling academic freedom" (Peter Bonilla, July 22)


Here is a Reddit comment that succinctly summarizes the story:

I'd recommend directly reading the Twitter accusations as well as the actual blog posts in question and Hsu's rebuttal. There's nothing like reading the primary sources for yourself. The most precise secondary source on this is probably WSJ.

You can see for yourself that none of the blog posts in question are studies done by Hsu himself. They are discussions of actual research papers published by other scientists in top-tier journals (Cell, Nature, Science, etc.). I've followed Hsu's blog for years, and he is always very careful to make clear that he doesn't think current research clearly supports genetic IQ differences. He is being targeted here just because he doesn't categorically rule out the possibility of genetic differences between populations and kowtow to the wokeness overlords. Some prominent psychologists and geneticists who signed his support letter include Steven Pinker, Robert Plomin, Robert Gordon, Linda Gottfredson, Jonathan Haidt, Lee Jussim... these are just a few I caught quickly scrolling through. Hsu is well within the mainstream and the controversy is pretty manufactured IMO.

The article and accusations makes it sound like Hsu's reactions to the George Floyd murder was to fund research that finds no racial disparities in police killing, which is incorrect. He was the Senior Vice President for Research and Innovation at MSU and one of his responsibilities is to allocate funding to different research projects. He allocated funding to this professor (Joe Cesario) before knowing what kind of conclusions his study would find later. This funding was decided a few years ago and the research concluded last year, well before the George Floyd incident. You can read a letter co-written by Joe Cesario here. The lab's research is actually really cool - they build a shooting simulator and subject an entire police department (few hundred officers at Milwaukee PD I believe) to different tests, and measure response times and decisions to shoot, etc. For more info you can read his papers or the NSF award abstract here.

I consider this a great loss, because he was one of the most highly positioned Asian-American academics and he was not shy about representing us and criticizing excessive affirmative action policies (although he favors having some mild preferences). A few years back, he and others made a very strong push for election to Harvard's alumni board of overseers on a slate for increasing admissions transparency. If he had succeeded, he would have done more for equality for Asians in this country than anyone else in recent memory. His efforts led to others taking up the mantle in the recent Harvard affirmative action case. It's perhaps because of this that he started to have a target on his back.

Zach Hambrick on Psychometrics and the Science of Expertise -- Manifold Podcast #28

MSU Psychology Professor Zach Hambrick joins Corey and Steve to discuss general cognitive ability, the science of personnel selection, and research on the development of skills and expertise. Is IQ really the single best predictor of job performance? Corey questions whether g is the best predictor across all fields and whether its utility declines at a certain skill level. What does the experience of the US military tell us about talent selection? Is the 10,000 hour rule for skill development valid? What happened to the guy who tried to make himself into a professional golfer through 10,000 hours of golf practice?

Transcript

Science of Expertise

Zach Hambrick (Faculty Profile)

Armed Services Vocational Aptitude Battery

Project 100,000 (1960s DoD Program)

Test Validity Study Report (CLA)

The Validity and Utility of Selection Methods in Personnel Psychology

Feynman and Tukey (Working Memory); Dom and Brexit

I received this message over the weekend.

Dear Dr. Hsu,

With great interest I regularly read your excellent Information Processing Blog. With regard to your assessment of Dom Cummings' achievements I am at variance with yours. But I guess you will like the anecdote referring to Feynman.

I tried to comment directly on your blog but the whole procedure was somewhat cumbersome, so I mail my comment directly to you. Please feel free to post it at the comment section under my full name. See the comment attached.

I am a retired psychology prof from University of Mannheim, Germany specializing in intelligence research, research methodology, assessment and evaluation research.

Best regards
Werner W. Wittmann

The letter:

IQ makes the difference

If you want to learn more about what kind of difference differences in IQ make read the research of Dave Lubinski and Camilla Benbow what differences highly gifted youngsters accomplish after several decades. Dave makes their publications available at https://my.vanderbilt.edu/smpy/publications/david-lubinski/

But let me turn to a funnier anecdote for physicists like Steve Hsu

A 35 year gap:

Physicists are among the smartest high IQ people, there is no doubt. If you want a single case example take Richard Feynman. If we could have lured him to psychology an important concept probably would have been published 35 years earlier.

In 1939 Feynman as a graduate student at Princeton experimented just for fun together with his friend John Tukey (who later became the famous statistician) to assess the ability of measuring time by counting.(Gleick,1992) They run stairs up and down to accelerate their heartbeats and trained themselves at the same time to count seconds and steps. Feynman’s performance deteriorated when he talked but not when he read. Tukey instead performed well when he recited poems aloud and worse when he read. So both have detected what is now known as the two slave systems of working memory, namely the phonological loop and the visuo-spatial sketchpad. Now you get a feeling how much more psychology would have been advanced if brains like theirs had been invested in my discipline at that time.

As a true and convinced European I am really sorry that the English left us, the Scots and the Northern-Irish didn’t want it and maybe one day the fame of tearing the United Kingdom into parts goes to Cummings as well?

What I would say to Cummings:

“If a thing is not worth doing, it is not worth doing well.” ― John W. Tukey

But he did it and now…

Boris Johnson probably to Cummings: “The moor has pled guilty the moor can go” ?

References:

Gleick,J.(1992) Genius. The life and science of Richard Feynman. New York: Pantheon Books. 

Lubinski, D., Benbow, C.P., & Kell, H.J. (2014). Life paths and accomplishments of mathematically precocious males and females four decades later. Psychological Science, 25, 2217–2232.

From Wikipedia about Working Memory 

In 1974, Baddeley and Hitch[11] introduced the multicomponent model of working memory. The theory proposed a model containing three components: the central executive, the phonological loop, and the visuospatial sketchpad with the central executive functioning as a control center of sorts, directing info between the phonological and visuospatial components.[12] The central executive is responsible inter alia for directing attention to relevant information, suppressing irrelevant information and inappropriate actions, and coordinating cognitive processes when more than one task is simultaneously performed. A "central executive" is responsible for supervising the integration of information and for coordinating "slave systems" that are responsible for the short-term maintenance of information. One slave system, the phonological loop (PL), stores phonological information (that is, the sound of language) and prevents its decay by continuously refreshing it in a rehearsal loop. It can, for example, maintain a seven-digit telephone number for as long as one repeats the number to oneself again and again.[13] The other slave system, the visuospatial sketchpad, stores visual and spatial information. It can be used, for example, for constructing and manipulating visual images and for representing mental maps. The sketchpad can be further broken down into a visual subsystem (dealing with such phenomena as shape, colour, and texture), and a spatial subsystem (dealing with location).

Re: Brexit, see these remarks from Now it can be told: Dominic Cummings and the Conservative victory 2019

I don't know enough to have a high confidence or high conviction opinion concerning Brexit. Intelligent and thoughtful people disagree strongly over whether it is a good idea or a potential disaster.

Nevertheless, I can admire Dom's effectiveness as a political strategist and chief advisor to the Prime Minister. I do know him well enough to state with high confidence that his intentions are idealistic, not selfish, and that he (someone who has spent decades thinking about UK government, foreign policy, relations with Europe) sincerely thinks Brexit is in the best interests of the British people. Dom has deeper insights and better intuition about these issues than I do!

Being a rationalist, Dom has pointed out on his own blog that it is impossible to know with high confidence what the future implications of most political decisions are... In that sphere one cannot avoid decision making under extreme uncertainty.

The epistemically careful may end up like Zhou Enlai. When asked about consequences of the French Revolution, the late premier is reported to have said: Too early to tell. Be prepared to find that thoughtful people, pressed for an opinion, can disagree...

Good and Bad Journalism on Embryo Screening: The Economist vs Science Magazine

Note Added in response to 2020 Twitter mob attack which attempts to misrepresent my views:

This post discusses a Science News article that misrepresented the activities of the startup Genomic Prediction (GP), which I helped to found. The news article wrongly conflated screening embryos for intelligence (which GP does NOT do; this is not really possible technically at the moment -- there is still too much noise in the predictors) with testing for unusual disease risk (which GP does).

Media outlets tend to be very sloppy in confusing these two types of screening. Screening for disease risk is very common in IVF. About 2 million embryos per year undergo some level of genetic screening. The original Science News article was corrected after a tedious interaction with the editor. There are many other news articles out there that make the same mistake, but we expected something better from Science. By contrast, The Economist article is very accurate and concise.

#####################################################

Modern genetics will improve health and usher in “designer” children (Economist), which I linked to in the last post, does an excellent job of covering the scientific, technical, and ethical issues raised by recent advances in polygenic risk prediction and embryo screening.

The author, Ananyo Bhattacharya, is an experienced science writer with (if I recall correctly) a degree in Physics. His forthcoming book is an ambitious scientific / intellectual history of John von Neumann!

What did Ananyo get right in his article?

1. He gives an overview of polygenic risk scores (PRS) and the underlying science behind GWAS studies and construction of risk predictors
2. He describes how PRS will have important applications in health care as well as in IVF
3. He discusses the important ethical and societal aspects of embryo screening

As someone who works in this area, I can say that I don't know of any popular work that combines the clarity, precision, and concision of this article (3 pages).

Unfortunately, not all journalism reaches this high standard.

For example, a really terrible ("click-bait") article appeared in the News section of Science recently, which conflated embryo screening to reduce disease risk with the optimization of complex traits such as IQ or height. I had numerous email exchanges with the writer (a self-described "non-scientist"), running to thousands of words, and including references to published work on disease risk reduction from genomic prediction. The resulting story was irresponsible, and very confusing to readers. I can judge this directly and empirically from communications I received in reaction to it.

Here is the letter we submitted to Science in response to the article. We do not know whether our letter will be published, but the News editor has already made significant revisions to the original article in response to our complaints.

Dear Editor,

Your news article Screening embryos for IQ and other complex traits is premature, study concludes (Oct 24 2019) contained significant errors, which we correct below. 

Each year roughly 2 million IVF embryos are genetically screened worldwide. In many developed countries, a significant fraction of all babies are born via IVF (e.g., almost 10% in Denmark). Reproductive health and IVF are serious matters and deserve serious journalism, not the inaccurate sensationalism of your article. Errors persist in the article even after numerous email exchanges (consisting of thousands of words of text, including references to published research) with your writer, informing your journalist clearly of these misrepresentations.

1. Your article failed to cite published work that shows significant risk reduction for complex disease conditions using polygenic predictors to select between sibling embryos. These results, which we emphasized many times to the writer, explicitly contradict this entire paragraph of the article:

The work "is the first to empirically test the viability of screening embryos" for traits that are influenced by many genes, says sociologist and demographer Melinda Mills of the University of Oxford in the United Kingdom. Such embryo screening goes beyond today's testing for single-gene disorders and currently "isn't plausible," she concludes.

[ Note: this paragraph has been altered now in the Science article. The original is given above. Science added this to the modified article, but still without referencing our work: *Clarification, 5 November, 10:05 a.m.: This story has been updated to clarify the context of a quote from Melinda Mills to emphasize that she was referring to screening for desirable traits, not disease risks. ]

Carmi et al. is not the first to empirically test embryo screening. Our published work predates it. Furthermore, Carmi’s work uses far less sibling data than our preceding work - an order of magnitude fewer siblings, 2-3 orders of magnitude fewer families (28 vs several thousand). Carmi’s analysis relies primarily on “simulated” data, ours is 100% empirical. We made your writer abundantly aware of the published work validating differentiation of real siblings (not “synthetic genomes”) by polygenic disease status, linking to it in email correspondence:

“You would be negligent to cite a BioRxiv preprint without thoroughly addressing our peer-reviewed, formally published work in the field, significantly predating this preprint.”

Your article misleads the reader to think that the dozens of IVF clinics and laboratories working with Genomic Prediction to screen embryos for complex (polygenic) disease risk do so without detailed, published validation. This is an irresponsible, unprofessional, and dangerous misrepresentation. We reserve the right to seek damages.

2. The article, and especially the title of the article, conflates screening embryos for disease with optimizing embryos for IQ, and gives the false impression that Carmi address the use-case of our patients: relative risk reduction of disease. This is misleading, as we repeatedly emphasized in writing with your journalist: “You will misrepresent our test if you fail to make this distinction...” , etc. The reader is misled by the article - especially the headline - to think that Carmi’s work addresses the current polygenic use-case of screening embryos for relative risk reduction of disease, rather than Carmi’s futuristic thought experiment of IQ optimization. This conflation is irresponsible, and a disservice to everyone, particularly to the IVF families using screening to reduce polygenic disease risk.

From IVF scientific pioneer Prof. Simon Fishel, external to Genomic Prediction. Fishel is former Deputy Scientific Director of the world's first IVF clinic, which included Nobel prize winning colleagues Dr. Steptoe and Dr. Edwards. His response to the Science article: "IVF itself was a revolutionary new technology that also endured an initial response of similar misrepresentation. There is no reason to repeat the mistakes of the past; Science should aim to convey the state of the field with less inaccuracy."

We ask that you publish this letter, and publish a correction to the article. We also ask that you recommit yourself to serious science reporting.

Sincerely,

Prof. Stephen Hsu
Dr. Nathan Treff
Laurent C. A. Melchior Tellier
Dr. Jia Xu
Prof. Simon Fishel

Note, I commented on Carmi et al. when it first appeared, here. This commentary was one of the first things I shared with the journalist and it makes very clear the difference between optimization of traits such as height or IQ (which Genomic Prediction does not do) and disease risk reduction (which is the main focus of our report).

The following is from an email I sent to the writer and editors:

... disease conditions are themselves complex traits and are typically referred to as such, so the risk of confusion is high. From the Wikipedia article on Complex Traits: "Examples of complex traits include height, circadian rhythms, enzyme kinetics, and many diseases including diabetes and Parkinson's disease." ...

Hence the title of the Science article Screening embryos for IQ and other complex traits is premature, study concludes is extremely misleading.

The most important scientific point: we have demonstrated that polygenic predictors can differentiate between two adult siblings, one with the disease and the other without. This is the gold standard validation relevant for embryo selection -- the predictor can identify from DNA alone the sibling with higher risk of the disease. The evidence is very strong that we can reduce disease risk through embryo screening, and IVF parents have a right to make use of this capability.

In an era of rapid scientific progress and technological change, the public deserves careful, accurate reporting -- not sensationalism.

For future reference, here are simple sentences which a journalist can include in any future article on this topic:

Published validation studies, using genomic data from thousands of families, have shown that polygenic scores can predict which of two adult siblings has the disease, and which one is healthy. It is reasonable to conclude that these predictors can reduce disease risk through IVF embryo screening.