Saturday, August 17, 2019

Polygenic Architecture and Risk Prediction for 14 Cancers and Schizophrenia

Two recent papers on polygenic risk prediction. As I've emphasized before, these predictors already have real clinical utility but they will get significantly better with more training data.
Assessment of Polygenic Architecture and Risk Prediction based on Common Variants Across Fourteen Cancers

Yan Zhang et al.

We analyzed summary-level data from genome-wide association studies (GWAS) of European ancestry across fourteen cancer sites to estimate the number of common susceptibility variants (polygenicity) contributing to risk, as well as the distribution of their associated effect sizes. All cancers evaluated showed polygenicity, involving at a minimum thousands of independent susceptibility variants. For some malignancies, particularly chronic lymphoid leukemia (CLL) and testicular cancer, there are a larger proportion of variants with larger effect sizes than those for other cancers. In contrast, most variants for lung and breast cancers have very small associated effect sizes. For different cancer sites, we estimate a wide range of GWAS sample sizes, required to explain 80% of GWAS heritability, varying from 60,000 cases for CLL to over 1,000,000 cases for lung cancer. The maximum relative risk achievable for subjects at the 99th risk percentile of underlying polygenic risk scores, compared to average risk, ranges from 12 for testicular to 2.5 for ovarian cancer. We show that polygenic risk scores have substantial potential for risk stratification for relatively common cancers such as breast, prostate and colon, but limited potential for other cancer sites because of modest heritability and lower disease incidence.

Some people are surprised that a mental disorder might be strongly controlled by genetics -- why? However, it has been known for some time that schizophrenia is highly heritable. I anticipate that good predictors for Autism and Alzheimer's disease will be available soon.
Penetrance and Pleiotropy of Polygenic Risk Scores for Schizophrenia in 106,160 Patients Across Four Health Care Systems

Amanda B. Zheutlin et al.

Individuals at high risk for schizophrenia may benefit from early intervention, but few validated risk predictors are available. Genetic profiling is one approach to risk stratification that has been extensively validated in research cohorts. The authors sought to test the utility of this approach in clinical settings and to evaluate the broader health consequences of high genetic risk for schizophrenia.

The authors used electronic health records for 106,160 patients from four health care systems to evaluate the penetrance and pleiotropy of genetic risk for schizophrenia. Polygenic risk scores (PRSs) for schizophrenia were calculated from summary statistics and tested for association with 1,359 disease categories, including schizophrenia and psychosis, in phenome-wide association studies. Effects were combined through meta-analysis across sites.

PRSs were robustly associated with schizophrenia (odds ratio per standard deviation increase in PRS, 1.55; 95% CI=1.4, 1.7), and patients in the highest risk decile of the PRS distribution had up to 4.6-fold higher odds of schizophrenia compared with those in the bottom decile (95% CI=2.9, 7.3). PRSs were also positively associated with other phenotypes, including anxiety, mood, substance use, neurological, and personality disorders, as well as suicidal behavior, memory loss, and urinary syndromes; they were inversely related to obesity.

The study demonstrates that an available measure of genetic risk for schizophrenia is robustly associated with schizophrenia in health care settings and has pleiotropic effects on related psychiatric disorders as well as other medical syndromes. The results provide an initial indication of the opportunities and limitations that may arise with the future application of PRS testing in health care systems.

Thursday, August 15, 2019

Bruno Maçães on The Power Game in a Connected World

Bruno Maçães in Singapore at IRAHSS Geopolitics Reimagined, 22 July 2019.

Maçães is author of Belt and Road - A Chinese World Order and former Europe Minister of Portugal. He discusses the trade war, his recent visit to a Huawei factory, and the idea of hybrid warfare or weaponized interdependence.

I met with Bruno in Beijing last month. He is among the most insightful geopolitical thinkers today.
I was shown the assembly line for the P30 smartphone [~$1k flagship using Huawei chipset] and told that this assembly line just two or three years ago was operated by 140 operators people it is now down to 17 by the end of this year we'll be down to 15 it's a very long assembly line perhaps 200 250 meters takes about 30 minutes more important than the time it takes to assemble a P30 is the time between each unit and that's now down to 29 seconds so every 29 seconds a fully produced P30 comes out at the end 17 people operate now this this assembly line but the remarkable thing is that I actually looked very carefully at what the 17 were doing and it's very obvious they're not doing anything of significance they left there more in order to keep a certain control over the process...

this is not a new Cold War and I see no indications that were moving in that direction China and the United States continue to be turned towards to each other continue to be very interested in learning from each other and I think this is an important point their way of life their ideology the way they look at the world is not predicated on a negation of the other side the Soviet Union was from the very start a revolutionary movement whose whole identity was the negation of capitalist Western Way of life and organizing society now China and the United States in a way are much less connected they are not part of the same history and their dispute is not a dispute about who is fundamentally right about questions that involved both...

they're not necessarily involved in a death and life struggle between them the world we live in is I'll sum it up this way a world where and this is I think the puzzling element of it we are neither at war nor at peace we are somewhere in the middle conflict takes below takes place below the threshold of kinetic war and other forms of direct confrontation but it is no less intense because of that...

the tactics might include the purchase of infrastructure in other states the corruption or blackmail of foreign officials important elements of this new world that is not often talked about [CALLING EPSTEIN AND GHISLAINE MAXWELL] manipulation of energy flows or energy prices all of these elements are magnified in an integrated global economy the networks that bring us together are used as tools or instruments of conflict...
More Bruno, on the Belt and Road initiative.

Wednesday, August 14, 2019

Epstein and the Big Lie

The biggest Epstein conspiracy mystery is not how he died. The more important mystery is how he managed to operate out in the open for 15-20 years. Rumors concerning Epstein and leading figures like Bill Clinton have been around for at least that long. I have been following his activities, at least casually, for well over a decade.

In the 1990s I was a Bill Clinton supporter. I voted for him twice and supported his efforts to move the democratic party in a centrist, pro-business direction. But my brother is a Republican. He fed me a steady stream of anti-Clinton information that I (at the time) dismissed as crazy right-wing conspiracy theories. However, with the advent of the internet in the late 90s it became easier to obtain information that was not filtered by corrupt mainstream media outlets. I gradually realized that at least some of my brother's claims were correct. For example, Clinton's first presidential bid was almost derailed by charges of adultery by women like Gennifer Flowers. Supporters like myself dismissed these charges as a right-wing smear. However, years later, Clinton admitted under oath that he had indeed had sex with Flowers.

My first exposure to Hillary Clinton was her appearance on 60 Minutes after the SuperBowl in 1992. This was widely regarded to be the emotive performance ("stand by your man") that saved Bill Clinton's presidential candidacy. Hillary affects a fake southern accent and (I believe) lies boldly and convincingly about Flowers to an estimated 50 million Americans. Quite a display of talent.

A side-effect of my history as a Clinton supporter (and gradual enlightenment thanks to my brother!) is that I became quite interested in the tendency of the media to hide obvious truths from the general public. We Americans accept that foreign governments (e.g., the Soviets and "ChiComs") successfully brainwash their people to believe all sorts of crazy and false things. But we can't accept that the same might be true here. (The big difference is that people in the PRC and former Soviet states  -- especially intellectuals -- know propaganda when they see it, whereas most Americans do not...)

It was natural for me to become aware of Epstein once he was linked to Bill Clinton at the very birth of the Clinton Foundation. It was easy to uncover very disturbing aspects of the Epstein story -- including details of his private island, traffic in young women, connections to the rich, the powerful, and even to leading scientists, academics, (many of whom I know) and Harvard University. Almost anyone with access to the internet (let alone an actual journalist) could have discovered these things at any point in the last decade.

But just 6 months ago I could mention Epstein to highly educated "politically aware" acquaintances with absolutely no recognition on their part.

Some obvious, and still unanswered, questions:

Former Federal prosecutor and Labor Secretary R. Alexander Acosta said he was told to lay off Epstein, as he "belongs to intelligence" -- why no media followup on this? (Still don't believe in a Deep State?)

Clinton said he only flew on Epstein's plane 4 times (but 26 is also commonly reported) and never visited the island (despite many eyewitness claims to the contrary). No investigative reporting on this by mainstream media?

Epstein's partner Ghislaine Maxwell is the daughter of Robert Maxwell, a billionaire with possible Mossad connections. What were Epstein's links to Israeli intelligence and national interests? (Robert Maxwell's death is at least as mysterious as Epstein's ...)

Why did it take the FBI so long to get to Epstein's island? What have they found in Epstein's house and on his island? How much blackmail material is there and who is implicated?

Were it not for the possibility that the Epstein scandal might be damaging to Trump, would there be anything close to this level of mainstream media interest?

Why was there almost zero interest in Epstein in the previous 15-20 years?

Someone was protecting Epstein (someone with influence on the DOJ, FBI, perhaps US intelligence) long before Donald Trump had political power of any kind. Why?

What other obvious scandals are hidden in plain sight? Iraq WMD? Spygate? Compromised politicians and national leaders? Blackmail by national intelligence services? Ideology-driven Social Media and Search filtering of information? Ivy League discrimination against Asian Americans? ...

Thursday, August 08, 2019

Manifold Episode #16: John Schulman of OpenAI

John Schulman is a research scientist at OpenAI. He co-leads the Reinforcement Learning group and works on agent learning in virtual game worlds (e.g., Dota) as well as in robotics. John, Corey, and Steve talk about AI, AGI (Artificial General Intelligence), the Singularity (self-reinforcing advances in AI which lead to runaway behavior that is incomprehensible to humans), and the creation and goals of OpenAI. They discuss recent advances in language models (GPT-2) and whether these results raise doubts about the usefulness of linguistic research over the past 60 years. Does GPT-2 imply that neural networks trained using large amounts of human-generated text can encode "common sense" knowledge about the world? They also discuss what humans are better at than current AI systems, and near term examples of what is already feasible: for example, using AI drones to kill people.

John Schulman


Better Language Models and Their Implications (GPT-2)

Transcript of show

man·i·fold /ˈmanəˌfōld/ many and various.

In mathematics, a manifold is a topological space that locally resembles Euclidean space near each point.

Steve Hsu and Corey Washington have been friends for almost 30 years, and between them hold PhDs in Neuroscience, Philosophy, and Theoretical Physics. Join them for wide ranging and unfiltered conversations with leading writers, scientists, technologists, academics, entrepreneurs, investors, and more.

Steve Hsu is VP for Research and Professor of Theoretical Physics at Michigan State University. He is also a researcher in computational genomics and founder of several Silicon Valley startups, ranging from information security to biotech. Educated at Caltech and Berkeley, he was a Harvard Junior Fellow and held faculty positions at Yale and the University of Oregon before joining MSU.

Corey Washington is Director of Analytics in the Office of Research and Innovation at Michigan State University. He was educated at Amherst College and MIT before receiving a PhD in Philosophy from Stanford and a PhD in a Neuroscience from Columbia. He held faculty positions at the University Washington and the University of Maryland. Prior to MSU, Corey worked as a biotech consultant and is founder of a medical diagnostics startup.

Sporting my OpenAI t-shirt. Wish I had worn this at Number 10 Downing Street earlier this week ;-)

Friday, August 02, 2019

Different Class Altogether

BBC Radio 4 profile of Dominic Cummings. (Sorry, didn't see any embed code.)

Some interesting comments from Dom's Oxford tutor, Robin James Lane Fox: (@4m50s)
He was extremely sharp, very sure of his own abilities, but had every reason to be... not narrow minded in any way...

BBC: Who is cleverer, Boris Johnson or Dominic Cummings?

Oh Dominic, by a long way.

BBC: A long way?

Different class altogether.

Robin James Lane Fox (Wikipedia), FRSL (born 5 October 1946)[1] is an English classicist, ancient historian and gardening writer known for his works on Alexander the Great.[2] Lane Fox is an Emeritus Fellow of New College, Oxford and Reader in Ancient History, University of Oxford. Fellow and Tutor in Ancient History at New College from 1977 to 2014...
See The Differences are EnormousCreators and Rulers, and The Gulf is Deep.

How Brexit was won, and the unreasonable effectiveness of physicists.

Blog Archive