Wednesday, September 28, 2022

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

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. 

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. ...

Tuesday, September 27, 2022

Quantum Hair in Electrodynamics and Gravity (arXiv:2209.12798)

New paper!
Quantum Hair in Electrodynamics and Gravity 
Xavier Calmet, Stephen D. H. Hsu 
We demonstrate the existence of quantum hair in electrodynamics and gravity using effective action techniques. In the case of electrodynamics we use the Euler-Heisenberg effective action while in the case of quantum gravity we use the unique effective action. We give a general formulation of these effects which applies to both theories and discuss analogies and differences between them. Furthermore, we present a QED analog to black hole evaporation. Spontaneous pair production in the external field of a ball of charge is analogous to Hawking radiation from black holes. Assuming spherical symmetry, the Gauss law prevents the external field from depending on the density profile of the ball. Quantum corrections violate these expectations, showing that quantum radiation can encode classically forbidden information about the source.
We found it interesting that quantum hair can already be found using the familiar Euler-Heisenberg effective action, which results from integrating out the electron in QED.

The paper also contains a general argument for why solutions to the semiclassical field equations resulting from the effective action (both in gravity and QED) carry more information about the state of the source than in classical physics.

From the Conclusions:
The quantum effective actions for both electrodynamics and gravity lead to field equations which couple a compact source (charge current or energy-momentum tensor) to external fields (electromagnetic or graviton field) in a manner which, generically, leads to quantum memory and quantum hair effects. External solutions of the field equations deviate, due to quantum corrections, from the familiar classical forms that satisfy the Gauss law. As a specific consequence, more information about the interior source configuration is encoded in the external field than in the classical theory. 
As specific applications, we considered semiclassical sources (large black hole, macroscopic charge distribution), which allowed us to solve the quantum corrected field equations by expanding around a classical solution. However, fully quantum statements regarding quantum hair are also possible, which do not, for example, require a semiclassical source. In [1–3] it was shown that the quantum state of a compact source (e.g., in an energy eigenstate or superposition thereof) determines certain aspects of the quantum state of its external field. In principle, measurements of the external fields can fully determine the interior state of a black hole.

Thursday, September 22, 2022

Rob Henderson: A Journey from Foster Care to the US Military to Elite Academia — Manifold podcast #20


Rob Henderson grew up in foster homes in California, joined the Air Force at 17, attended Yale on the G.I. Bill, and is currently a Gates Fellow at Cambridge University (UK). He is an acute observer of American society and has coined the term Luxury Beliefs to describe ideas and opinions that confer status on the rich at very little cost, while taking a toll on the lower class. 

Steve and Rob discuss: 

00:00 Early life and foster experience 
20:21 Rob’s experience in the Air Force 
31:26 Transitioning from the Air Force to Yale and then Cambridge 
44:04 Dating and socializing as an older student 
50:06 Reflections on the Yale Halloween email controversy 
1:01:10 Personal incentives and careerists in higher education 
1:09:45 Luxury beliefs and how they show up in elite institutions 
1:31:08 Age and moral judgments 
1:42:50 Rob on resisting legacy academia and his future 


Rob's substack 

Luxury Beliefs are the Latest Status Symbol for Rich Americans

Tuesday, September 20, 2022

Sibling Variation in Phenotype and Genotype: Polygenic Trait Distributions and DNA Recombination Mapping with UK Biobank and IVF Family Data (medRxiv)

This is a new paper which uses Genomic Prediction IVF family data, including genotyped embryo samples.
Sibling Variation in Phenotype and Genotype: Polygenic Trait Distributions and DNA Recombination Mapping with UK Biobank and IVF Family Data
L. Lello, M. Hsu, E. Widen, and T. Raben  
We use UK Biobank and a unique IVF family dataset (including genotyped embryos) to investigate sibling variation in both phenotype and genotype. We compare phenotype (disease status, height, blood biomarkers) and genotype (polygenic scores, polygenic health index) distributions among siblings to those in the general population. As expected, the between-siblings standard deviation in polygenic scores is \sqrt{2} times smaller than in the general population, but variation is still significant. As previously demonstrated, this allows for substantial benefit from polygenic screening in IVF. Differences in sibling genotypes result from distinct recombination patterns in sexual reproduction. We develop a novel sibling-pair method for detection of recombination breaks via statistical discontinuities. The new method is used to construct a dataset of 1.44 million recombination events which may be useful in further study of meiosis.

Here are some figures illustrating the variation of polygenic scores among siblings from the same family.

The excerpt below describes the IVF family highlighted in blue above:

Among the families displayed in these figures, at position number 15 from the left, we encounter an interesting case of sibling polygenic distribution relative to the parents. In the family all siblings have significantly higher Health Index score than the parents. This arises in an interesting manner: the mother is a high-risk outlier for condition X and the father is a high-risk outlier for condition Y. (We do not specify X and Y, out of an abundance of caution for privacy, although the patients have consented that such information could be shared.) Their lower overall Health Index scores result from high risk of conditions X (mother) and Y (father). However, the embryos, each resulting from unique recombination of parental genotypes, are normal risk for both X and Y and each embryo has much higher Health Index score than the parents.
This case illustrates well the potential benefits from PGS embryo screening.

The second part of the paper introduces a new technique that directly probes DNA recombination -- the molecular mechanism responsible for sibling genetic differences. See figure above for some results. The new method detects recombination breaks via statistical discontinuities in pairwise comparisons of DNA regions.

From the discussion:
...This new sibling-pair method can be applied to large datasets with many thousands of sibling pairs. In this project we created a map of roughly 1.44 million recombination events using UKB genomes. Similar maps can now be created using other biobank data, including in non-European ancestry groups that have not yet received sufficient attention. The landmark deCODE results were obtained under special circumstances: the researchers had access to data resulting from a nationwide project utilizing genealogical records (unusually prevalent in Iceland) and widespread sequencing. Using the sibling-pair method results of comparable accuracy can be obtained from existing datasets around the world -- e.g., national biobanks in countries such as the USA, Estonia, China, Taiwan, Japan, etc.
The creator of this new sibling-pair method for recombination mapping is my son. He developed and tested the algorithm, and wrote all the code in Python. It's his high school science project :-)

Thursday, September 08, 2022

Lyle Goldstein on U.S. Strategic Challenges: Russia, China, Ukraine, and Taiwan — Manifold #19


Professor Goldstein recently retired after 20 years of service on the faculty of the U.S. Naval War College (NWC). During his career at NWC, he founded the China Maritime Studies Institute (CMSI) and has been awarded the Superior Civilian Service Medal for this achievement. He has written or edited seven books on Chinese strategy and is at work on a book-length project that examines the nature of China-Russia relations in the 21st century. He has a longstanding interest in great power politics, military competition, and security in the pacific region. 

Goldstein is Director of Asia Engagement at the Washington think-tank Defense Priorities, which advocates for realism and restraint in U.S.defense policy, and also a visiting professor at the Watson Institute for International and Public Affairs at Brown University. 

He earned a PhD at Princeton, an MA from Johns Hopkins SAIS, and an AB from Harvard. He is fluent in both Chinese and Russian. 

Steve and Lyle discuss: 

00:00 Early life and background 
18:03 Goldstein’s dissertation on China’s nuclear strategy 
37:35 Pushback on “Meeting China Halfway” 
41:24 Could the U.S. have prevented war in Ukraine? 
46:05 How territorial conflicts are influencing China’s relationship with Russia 
1:00:16 Analyzing war games with U.S., China, and Taiwan 


Watson Institute, Brown University 

Meeting China Halfway (2015) 

Here's Why War With China Could Elevate to Nuclear Strikes The National Interest, January 29 2022 

Goldstein's articles at The National Interest 

Monday, September 05, 2022

Lunar Society (Dwarkesh Patel) Interview


Dwarkesh did a fantastic job with this interview. He read the scientific papers on genomic prediction and his questions are very insightful. Consequently we covered the important material that people are most confused about. 

Don't let the sensationalistic image above deter you -- I highly recommend this podcast!

0:00:00 Intro 
0:00:49 Feynman’s advice on picking up women 
0:12:21 Embryo selection 
0:24:54 Why hasn't natural selection already optimized humans? 
0:34:48 Aging 
0:43:53 First Mover Advantage 
0:54:24 Genomics in dating 
1:01:06 Ancestral populations 
1:08:33 Is this eugenics? 
1:16:34 Tradeoffs to intelligence 
1:25:36 Consumer preferences 
1:30:49 Gwern 
1:35:10 Will parents matter? 
1:46:00 Wordcels and shape rotators 
1:58:04 Bezos and brilliant physicists 
2:10:58 Elite education 

If you prefer audio-only click here.

Blog Archive