Soil’s Microbial Market Shows the Ruthless Side of Forests

In the “underground economy” for soil nutrients, fungi strike hard bargains and punish plants that won’t meet their price.

 Gabriel Popkin

Contributing Writer

[[This article is astonishingly beautiful,   and a great illustration of my new motto: eveything is much much much more complicated that we thought. ]]

---

August 27, 2019

BacteriaBiologyBotanyCellsCommunicationCooperationEcologyEvolutionary BiologyMicrobiomePlants

 https://www.quantamagazine.org/soils-microbial-market-shows-the-ruthless-side-of-forests-20190827/?utm_source=Quanta+Magazine&utm_campaign=26189edbe2-RSS_Daily_Biology&utm_medium=email&utm_term=0_f0cb61321c-26189edbe2-389846569&mc_cid=26189edbe2&mc_eid=61275b7d81

Beneath the green vegetable world we see is a dark microbial world we don’t. The crops we eat, the forests that sustain us and most other life forms, even the regulation of Earth’s climate — all benefit from a shadowy network of fungi and bacteria that mobilize soil nutrients and trade them with plants for sugars and fats. And yet the workings of this subterranean society are almost unknown to scientists. For example, researchers just mapped for the first time the global distribution of three major groups of these microbes. Even in 2019, what lies beneath our feet remains a true scientific frontier.

Despite this epistemological murkiness, public interest in the underground ecosystem has exploded. TED talks and bestselling books extol the benevolent, cooperative “wood wide web” of subsurface organisms that communicate, share nutrients and sustain each other.

Toby Kiers, an evolutionary biologist at VU University Amsterdam, is at the vanguard of a new generation of scientists questioning that gauzy view. Through innovative and groundbreaking studies, Kiers and her collaborators have gathered evidence that plants and their fungal conspirators are not just cooperating with each other but also engaging in a raucous and often cutthroat marketplace ruled by supply and demand, where everyone is out to get the best deal for themselves and their kind.

Doesn’t it strike you as odd? … This is undoubtedly the most important network for our ecosystems, but we just don’t know anything about it.

Toby Kiers, VU University Amsterdam

Key to this picture is the revelation that the unseen underground world is just as complex, sophisticated and purposeful as the visible aboveground world we inhabit. Microbes are not simple, passive accessories to plants, but dynamic, powerful actors in their own right. Fungi can hoard nutrients, they can reward plants that are generous with their carbon reserves and punish ones that are stingy, and they can deftly move and trade resources to get the best “deal” for themselves in exchange.

Those are probably just the beginning of their talents. In a paperpublished in June, Kiers and her colleagues pioneered a method to illuminate the fungal marketplace in action — to make the invisible visible. The tantalizing research hints at a capability that has been suspected but never proved: that fungi might not be just nutrient traders but also sophisticated information processors.

Kiers is the first to admit that scientists have far to go in puzzling out the hidden rules of the tiny networked organisms that somehow support all the rest of us. “Doesn’t it strike you as odd?” she said. “We know so much about other types of networks. This is undoubtedly the most important network for our ecosystems, but we just don’t know anything about it. … It’s radically under-studied.”

Ancient Partners

When plants crept onto land some 500 million years ago, microbes were waiting. Fungi and bacteria struck up relationships with their new neighbors. Plants, after all, could do something most microbes could not: harness solar energy to split apart atmospheric carbon dioxide and construct energy-rich sugars and fats from the pieces. The microbes, in turn, had mastered the art of freeing up the nutrients that plants needed from the soil — phosphorus especially, but also nitrogen; there is evidence that microbes help plants gain access to water as well. Some 80 percent or more of today’s land plants form partnerships with fungi; still other plants partner with bacteria. If the soil were somehow purged of its microbes, the plant and animal worlds would take a big hit. The views of the great naturalist E.O. Wilson notwithstanding, it’s microbes, not insects, that run the world.

And yet the soil microbiome is little known and even less appreciated. There are reasons for this: Soil is opaque and microbes are, well, microscopic. They’re also hard to study; many won’t grow in the lab, and the wispy fungal networks that pervade soil break easily when extracted. Most of all, though, microbes confound our understanding of life, which has been shaped by our experience of the aboveground world. Some fungi don’t have proper cells, for example: Their DNA-containing nuclei float through threadlike subterranean networks that can be kilometers long. It’s often hard to say what it means for fungi to have sex, or even to define what constitutes an individual.

Kiers saw instead a world ordered by individual interests, where potential cheaters lurked everywhere and species needed complex strategies to keep their trading partners in line.

As scientists began to appreciate the importance of microbe-plant partnerships, at least in outline, many came to see the natural world as a cooperative, even communitarian kind of place. In the 1970s, the microbiologist Lynn Margulis and the chemist James Lovelock developed the Gaia hypothesis, which posits that Earth’s biosphere is, in some sense, a unified self-regulating organism. The existence of mutually beneficial inter-species and even inter-kingdom relationships fit right into this picture.

In the mid-1990s, a young biologist named Suzanne Simard, now at the University of British Columbia, decided to test this concept in a forest. “Some people thought I was crazy,” Simard (who did not respond to multiple interview requests) said in a 2016 TED talk. For her doctoral project at Oregon State University, she took carbon dioxide with radioactive carbon isotopes and injected it into bags installed around pint-size birch trees growing near Douglas fir seedlings. After a little while, she ran a Geiger counter along the Douglas fir trees, and the device beeped like crazy. Moreover, she found that the radioactive carbon could also flow from the Douglas firs to the birches if she planted the bags near the firs. She had discovered that the trees shared carbon via underground networks. Her findings, published in Nature in 1997, lit a fire under scientists and the public alike.

In describing what she found, Simard has emphasized the cooperation she views as inherent in nature. “A forest is a cooperative system, and if it were all about competition, then it would be a much simpler place,” she said to Yale Environment 360 in 2016. “Why would a forest be so diverse? Why would it be so dynamic?” In her TED talk, Simard referred to forests as “supercooperators” and made the bold assertion that trees don’t just cooperate but communicate. She described birches and Douglas firs as engaging in a “lively two-way conversation” mediated by their underground collaborators. “I had found solid evidence of this massive belowground communications network,” she said, adding later in the talk, “Through back and forth conversations, they increase the resilience of the whole community.”

Suzanne Simard, professor of forest ecology at the University of British Columbia, drew researchers’ attention to the evidence that diverse forest plants don’t just cooperate but communicate with one another.

Mike Hitchner

Not long after Simard bagged saplings in the misty Pacific Northwest, Kiers decamped from Bowdoin College in frigid Maine to the warm and humid Barro Colorado Island in Panama. She had become fascinated by the underground world, and, at her undergraduate mentor’s suggestion, she spent a year at the Smithsonian’s renowned tropical field station in the main channel of the Panama Canal, studying how soil fungi help tropical trees grow. She then finished up her bachelor’s degree and headed to the University of California, Davis. There she began studying one of the world’s most famous mutualisms, the one between plants in the legume family — which includes important crops like soybeans and alfalfa and trees like the locust — and rhizobial bacteria. These specialized microbes nestle into spherical white nodules on plants’ roots and become nutrient factories, converting inorganic nitrogen from the air into biologically useful forms. They then trade the nitrogen to plants for carbon-rich sugars.

It seems like a balanced, helpful exchange between friendly partners. But to Kiers, the benevolent, cooperation-focused view promoted by thinkers like Margulis and Simard was suspect. Kiers saw instead a world ordered by individual interests, where potential cheaters lurked everywhere and species needed complex strategies to keep their trading partners in line.

“I had this realization … that I’m less interested in cooperation and I’m actually much more interested in the tension,” Kiers said. “I think there’s an underappreciation of how tension drives innovation. Cooperation to me suggests a stasis.”

As she soon learned, the legume-bacteria interaction is not so simple. A single legume plant can host 10 or more strains of bacteria. To Kiers, this evoked a concept from the ecologist Garrett Hardin, whose famous 1968 essay in Science, “The Tragedy of the Commons,” argued that individuals pursuing their own interests can destroy a common environment or resource. The legume plant itself could be seen as a commons, and any given bacterial strain could hoard nitrogen while continuing to feast on the plant’s sugars. “Why should they fix nitrogen — what’s in it for them?” asked Ford Denison, Kiers’ adviser, now an adjunct professor who runs an ecology and evolution lab at the University of Minnesota.

Along with Stuart West, an evolutionary theorist at the University of Oxford whom Kiers had met in Panama, she and Denison modeled the legume-rhizobia interaction mathematically, showing that if bacteria could “cheat” their plant hosts, the relationship would fall apart as more and more strains defected. Kiers, Denison and their colleague Robert Rousseau then designed an experiment that would essentially force some of the bacteria to cheat.

Toby Kiers, an evolutionary biologist at VU University Amsterdam, finds that the interactions among plants and their fungal symbiotes resemble a cutthroat marketplace in which the species negotiate their exchanges of nutrients ruthlessly.

Kiers surrounded some nodules on soybean plants with an almost nitrogen-free air supply, making the bacteria in those nodules useless to the plant. She found that the plant reacted by shutting off the supply of oxygen to those bacteria, drastically reducing their reproduction. It seemed the relationship between the bacteria and the soybeans, far from being a happy friendship, was an uneasy détente, with the plant imposing crippling sanctions on any bacterial partners that failed to earn their keep. The paper, which was published in Nature before Kiers even received her doctorate, made a huge splash. “It’s the most cited paper in my career,” Denison said.

Kiers then switched from bacteria to fungi. While bacteria might nestle into the roots of select groups of plants, fungi are without question the masters of the underground domain. Certain fungi spread through vast areas and commingle with just about every plant they encounter, even sending thready tendrils known as hyphae directly into plants’ roots. (The name for these fungi — “mycorrhizae” — literally fuses the Latin myco-, meaning “of fungi,” with the Greek rhiza, or “root.”) Indeed, the mycorrhizal world forms a sort of inversion of the vegetable one, with branching fungal networks extending downward, mirroring the branching stems and limbs of the plants reaching skyward.

But what really distinguishes the fungal world is its diversity and complexity. A spoonful of soil contains more microbial individuals than there are humans on Earth. “It’s the most species-dense habitat we have,” said Edith Hammer, a soil ecologist at Lund University in Sweden. A single plant might be swapping molecules with dozens of fungi — each of which might in turn be canoodling with an equal number of plants. It’s a promiscuous party down there.

Faced with such overwhelming complexity, scientists must simplify, just not too much (as Einstein is alleged to have said). Kiers and her colleagues did this with a petri dish divided into three equal-size compartments, like a Mercedes symbol. In one, they grew carrot roots deprived of leaves along with fungal species known to associate with carrots. The fungal hyphae, but not the plant roots, were able to grow into the other two compartments to look for nutrients. The researchers gave a special “heavy” form of carbon (the isotope carbon-14) to the carrots; they also made phosphorus available to the fungi that reached into one of the other two compartments. In this way, the scientists could track the movement of sugars and nutrients through the simplified ecosystem. After some time, they measured the fungi’s growth and found that the fungi with phosphorus to trade received much more carbon from the plants.

By growing carrot roots and fungi in segmented petri dishes that held unequally distributed nutrient resources, Kiers and her colleagues demonstrated that the organisms favor symbiotic partners that have more to offer in nutrient exchanges.

What Kiers’ team did next was the real coup. They inverted the setup: carrot roots with fungi in two compartments and one compartment that only the fungus could reach. They gave one carrot compartment more sugar than the other. They waited, and then measured. The plant with more sugar to trade had received far more fungal phosphorus (which in this experiment was recognizable as the “heavy” isotope phosphorus-32).

In 2011, Kiers’ team reported in Science that not only can plants reward high-performing fungal partners and punish poor performers, fungi apparently do the same.

Around the same time, Hammer reported evidence from experiments that fungi have a second trick: They can store nutrients when a plant isn’t paying well, withholding them until they get a better offer.

Together, the results turned scientists’ understanding of the plant-fungal relationship on its head. No longer could mycorrhizal fungi be seen as servants or passive accessories to their plant masters. Rather, life forms below the surface control their own fate, just as much as those above. It’s a dynamic marriage of equals.

“I don’t know if we should say that we enlightened the field, but I think a lot of people thought [fungi] were much simpler” and mostly only responded to plants’ signals, Hammer said.

The findings also established Kiers as an important thinker in her own right. “It’s pretty amazing that this one person provided the first solid evidence for sanctions or discrimination in what are arguably the two most important symbioses — the legume-rhizobia system and the mycorrhizae,” Denison said.

She followed up the lab experiment with a less artificial one: She grew plants connected to mycorrhizae in shade and in full sunlight. She found that the shaded plants, which photosynthesized less and thus had fewer sugars to share, received less phosphorus from their underground fungal counterparts.

She also began developing an economic framework for thinking about relationships between plants and fungi. Based on observations of the free-market system, Kiers suspects that what has stabilized plant-fungal mutualisms for at least 470 million years is not that individual organisms are committed to the good of the community, but rather that, in most cases, both plants and fungi benefit more from trading with each other than from keeping resources to themselves.

Economics provides “this huge body of literature we can borrow from that’s actually mathematical and predictable,” Kiers said. “It can be used as a tool to test some of these ideas.” For example, Kiers and her colleagues found that mutualisms sometimes break down when plants find another way to get the nutrients they need — by turning carnivorous and catching insects, for example. They published their findings last year in the Proceedings of the National Academy of Sciences.

Not everyone is convinced that the cutthroat economic world of markets and traders describes plants and microbes well. “I think it’s possible that sometimes the fitness interests of partners just happen to be really well aligned,” said Megan Frederickson, a professor of ecology and evolutionary biology at the University of Toronto who has argued in several papers that cheating is far less common in nature than Kiers believes. “I think some other people take the view that it’s probably impossible for two partners in any interaction to have perfectly aligned interests.”

That view has also come to dominate the popular literature. In 2016, the German forester Peter Wohlleben, drawing heavily on Simard and a few other scientists, published The Hidden Life of Trees. The book became an international bestseller. Wohlleben, a strong advocate for a communitarian view of nature, wholeheartedly promoted fungi’s supposed beneficence and cooperative nature, writing, “The fungi that populate [trees’ roots] seem to be intent on compromise.”

Together, the results turned scientists’ understanding of the plant-fungal relationship on its head. No longer could mycorrhizal fungi be seen as servants or passive accessories to their plant masters.

Kiers, characteristically, has turned to different sources for inspiration. A few years ago, she read the economist Thomas Piketty’s Capital in the Twenty-First Century, which emphasizes the role of resource inequality in shaping human societies. Kiers suspected that introducing inequality into her fungus-plant ecosystem would reveal novel insights. But she found herself facing the same challenge that had stymied so many scientists: She had no way to directly see what fungi were doing. She had gotten a lot of mileage out of measuring nutrients going in and seeing where they ended up, but what happened within the fungi themselves remained a black box.

“That’s how we start lab meeting every week: ‘If we can’t talk to them, how do we get at that question?’” Kiers said.

The breakthrough came unexpectedly. Kiers had gotten a Dutch government grant to work with an artist on a stop-motion video depicting phosphorus and other nutrients moving through the fungal network, using LEDs to represent nutrients. She showed the animation at a 2014 scientific conference. “Wouldn’t it be cool if we could do that?” she asked her audience.

Matthew Whiteside, a chemist then working at the University of British Columbia, approached Kiers afterward. “We really can do that,” he said.

Whiteside — who, incidentally, had known Margulis as a family friend and spoken with her about pursuing science — had developed a way to tag biomolecules with quantum dots, nanometer-scale bits of semiconductor that absorb certain short wavelengths of light (usually ultraviolet) and re-emit light, or “fluoresce,” at longer wavelengths. Kiers hired him. The two spent several years developing the technique and ensuring that they could distinguish the dots’ emissions from the plant cells’ natural fluorescence. They also had to ensure that the dots wouldn’t poison fungi or plants.

In this micrograph of fungal hyphae, nutrient biomolecules containing phosphorus have been tagged with quantum dots, which fluoresce as green. This labeling technique allowed Kiers’ team to capture photographically what had only been imagined in their video animation.

Victor Caldas

Kiers and Whiteside then set up another experiment in the three-compartment petri dish. They grew carrot roots with fungi in one compartment and allowed the fungi to expand into the other two. Then they introduced apatite, a phosphorus-containing mineral, into the fungi-only compartments. In one, they labeled the apatite with a red-fluorescing quantum dot; in the other they used a blue dot. They used specialized microscopes to quantify the emitted light.

At first they gave each fungal compartment equal amounts of apatite. As expected, the fungi took up the phosphorus-laden compound through their networks but didn’t grow much, choosing instead to store much of the nutrient rather than trade it.

Then came the part inspired by Piketty’s work on economics. Kiers and Whiteside added more phosphorus to one compartment than the other, setting up resource inequalities, with one fungal group controlling up to 90% of the element. The fungi responded by trading with the plants far more than when phosphorus was evenly distributed in the environment.

Most impressively, the fungi moved nutrients from the “rich” to the “poor” region and grew faster in the poor region. Kiers’ team believes that’s because the fungi could extract a higher “price” from the plants in the form of carbon-rich sugars where phosphorus was scarce — though Kiers notes that they couldn’t track the carbon directly.

We’re starting to deconstruct [the fungal network] piece by piece.

Toby Kiers, VU University Amsterdam

“This is totally opposite of what we had anticipated,” she said. She thought trading would be highest where nutrients were already abundant.

The demonstration impressed others. “We often think of fungi or other microbes as not particularly intelligent,” said Kabir Peay, a biologist at Stanford University. “This study goes to show that across these networks, one of the reasons they can be so successful is that they can make what seem to be fairly sophisticated decisions about where to allocate resources to optimize the return they get.”

But scientists must take care when applying market concepts to the biological word, noted Ronald Noë, a biologist and professor emeritus at the University of Strasbourg in France who helped pioneer biological market theory in studies of primates, and who assisted Kiers’ team with the economic analysis. “What they describe is a mechanism by which you could be a trader in the market — you can see how they could do it. But they didn’t actually prove that they did it,” Noë said. “If there would be a market, the fungus would bring its nutrients from one plant to the other. But in the experiment, there’s only one plant. The fungus is not choosing.”

The experiment revealed a second surprise. Phosphorus did not just flow from the rich region to the poor one. Kiers and Whiteside caught some of the glowing nutrients oscillating back and forth through the network every five minutes in a regular rhythm.

The scientists don’t know what these oscillations mean. But they do know that oscillations are common ways to encode information. For example, radios work by encoding information in oscillations of radio waves, which are low-frequency electromagnetic waves. Could the fungal oscillations be a form of information transfer across the network?

There is, in fact, strong evidence that information flows across fungal networks. In 2013, David Johnson, a biologist then at the University of Aberdeen in Scotland, discovered that bean plants attacked by aphids sent chemicals underground through fungal networks and into nearby plants that were then alerted to the presence of the pests. Simard has found similar chemical releases in the forests she studies.

But in those studies, scientists inferred what was happening underground from measurements of chemicals in trees. Left unanswered was whether fungal networks are merely conduits of plant-to-plant signals, or if they can process the information they receive. If someone were able to prove the latter, said Erik Verbruggen, a biologist at the University of Antwerp in Belgium who did his doctorate with Kiers, “that would be quite extraordinary.”

Kiers has illuminated the fungal network for the first time. Her next goal is to narrow the gulf between her experimental setups and the complexity of nature. For example, Kiers’ petri dishes flatten the plant-fungus world into two dimensions, but in real soils, fungal networks are 3D. Different species’ networks overlap and interweave, making the wood wide web more like a wood wide tangle with dozens of independent wiring schemes.

“I’d be the first to admit that this is incredibly artificial,” Kiers said of her lab setup. “But that’s actually the beauty of it. We’re starting to deconstruct [the fungal network] piece by piece.”

Kiers is now teaming up with the fluid dynamics researcher Howard Stone of Princeton University, the biophysicist Tom Shimizu of VU University, and the network ecologist Hirokazu Toju of Kyoto University. (Their work is funded by a grant from the Human Frontier Science Program to encourage international collaborations on high-risk projects.) They plan to use microfluidic tools to create intricate 3D environments that more closely resemble real-world fungal networks. Stone hopes to augment Kiers’ quantum-dot tags with other techniques for tracking resources such as the carbon that plants trade; some commentators critiqued Kiers’ inequality study for failing to account for that carbon.

“There’s a whole system that we have to set up, including imaging and understanding and possibly modeling, that no one’s done,” Stone said.

For Kiers, the collaboration promises further revelations about how much we’ve misjudged our microbial counterparts.

“Even I, still to this day, underestimate fungi,” she said.

Many Genes Influence Same-Sex Sexuality, Not a Single ‘Gay Gene’

By Pam Belluck

 Aug. 29, 2019

https://www.nytimes.com/2019/08/29/science/gay-gene-sex.html?campaign_id=60&instance_id=0&segment_id=16585&user_id=feb81aacf509dc0b47069671919630f6&regi_id=1028677ing-news

[[Two points of commentary. 1. Shock! Surprise! Astonishment! But it has been long known that man y traits depend upon interaction of many genes. Wikipedia:

Often different genes can interact in a way that influences the same trait. In the Blue-eyed Mary (Omphalodes verna), for example, there exists a gene with alleles that determine the color of flowers: blue or magenta. Another gene, however, controls whether the flowers have color at all or are white. When a plant has two copies of this white allele, its flowers are white—regardless of whether the first gene has blue or magenta alleles. This interaction between genes is called epistasis, with the second gene epistatic to the first.^[43]

Many traits are not discrete features (e.g. purple or white flowers) but are instead continuous features (e.g. human height and skin color). These complex traits are products of many genes.^[44] The influence of these genes is mediated, to varying degrees, by the environment an organism has experienced. The degree to which an organism's genes contribute to a complex trait is called heritability.^[45] Measurement of the heritability of a trait is relative—in a more variable environment, the environment has a bigger influence on the total variation of the trait. For example, human height is a trait with complex causes. It has a heritability of 89% in the United States. In Nigeria, however, where people experience a more variable access to good nutrition and health care, height has a heritability of only 62%.

2. The most important conclusion is that now a case for genetic determinism of sexual orientation is impossible to make..]]

How do genes influence our sexuality? The question has long been fraught with controversy.

An ambitious new study — the largest ever to analyze the genetics of same-sex sexual behavior — found that genetics does play a role, responsible for perhaps a third of the influence on whether someone has same-sex sex. The influence comes not from one gene but many, each with a tiny effect — and the rest of the explanation includes social or environmental factors — making it impossible to use genes to predict someone’s sexuality.

“I hope that the science can be used to educate people a little bit more about how natural and normal same-sex behavior is,” said Benjamin Neale, a geneticist at the Broad Institute of M.I.T. and Harvard and one of the lead researchers on the international team. “It’s written into our genes and it’s part of our environment. This is part of our species and it’s part of who we are.”

The study of nearly half a million people, funded by the National Institutes of Health and other agencies, found differences in the genetic details of same-sex behavior in men and women. The research also suggests the genetics of same-sex sexual behavior shares some correlation with genes involved in some mental health issues and personality traits — although the authors said that overlap could simply reflect the stress of enduring societal prejudice.

Even before its publication Thursday in the journal Science, the study has generated debate and concern, including within the renowned Broad Institute itself. Several scientists who are part of the L.G.B.T.Q. community there said they were worried the findings could give ammunition to people who seek to use science to bolster biases and discrimination against gay people.

One concern is that evidence that genes influence same-sex behavior could cause anti-gay activists to call for gene editing or embryo selection, even if that would be technically impossible. Another fear is that evidence that genes play only a partial role could embolden people who insist being gay is a choice and who advocate tactics like conversion therapy.

“I deeply disagree about publishing this,” said Steven Reilly, a geneticist and postdoctoral researcher who is on the steering committee of the institute’s L.G.B.T.Q. affinity group, Out@Broad. “It seems like something that could easily be misconstrued,” he said, adding, “In a world without any discrimination, understanding human behavior is a noble goal, but we don’t live in that world.”

Discussions between Dr. Neale’s team and colleagues who questioned the research continued for months. Dr. Neale said the team, which included psychologists and sociologists, used suggestions from those colleagues and outside L.G.B.T.Q. groups to clarify wording and highlight caveats.

“I definitely heard from people who were kind of ‘why do this at all,’ and so there was some resistance there,” said Dr. Neale, who is gay. “Personally, I’m still concerned that it’s going to be deliberately misused to advance agendas of hate, but I do believe that the sort of proactive way we’ve approached this and a lot of the community engagement aspects that we’ve tried were important.”

The moment the study was published online Thursday afternoon, the Broad Institute took the unusual step of posting essays by Dr. Reilly and others who raised questions about the ethics, science and social implications of the project.

“As a queer person and a geneticist, I struggle to understand the motivations behind a genome-wide association study for non-heterosexual behavior,” wrote Joe Vitti, a postdoctoral researcher at the Broad Institute, in one essay. “I have yet to see a compelling argument that the potential benefits of this study outweigh its potential harms.”

In a way, the range of opinions by scientists who also identify as L.G.B.T.Q. underscores a central finding of the study: Same-sex sexuality is complicated.

The study analyzed the genetic data of 408,000 men and women from a large British database, the U.K. Biobank, who answered extensive health and behavior questions between 2006 and 2010, when they were between the ages of 40 and 69. The researchers also used data from nearly 70,000 customers of the genetic testing service 23andMe, who were 51 years old on average, mostly American, and had answered survey questions about sexual orientation. All were of white European descent, one of several factors that the authors note limit their study’s generalizability. Trans people were not included.

The researchers mainly focused on answers to one question: whether someone ever had sex with a same-sex partner, even once.

A much higher proportion of the 23andMe sample — about 19 percent compared to about 3 percent of the Biobank sample — reported a same-sex sexual experience, a difference possibly related to cultural factors or because the specific 23andMe sexual orientation survey might attract more L.G.B.T.Q. participants.

Despite its limitations, the research was much larger and more varied than previous studies, which generally focused on gay men, often those who were twins or were otherwise related.

“Just the fact that they look at women is hooray,” said Melinda Mills, a professor of sociology at the University of Oxford, who wrote a commentary that Science published alongside the study.

There might be thousands of genes influencing same-sex sexual behavior, each playing a small role, scientists believe. The new study found that all genetic effects likely account for about 32 percent of whether someone will have same-sex sex.

Using a big-data technique called genome-wide association, the researchers estimated that common genetic variants — single-letter differences in DNA sequences — account for between 8 percent and 25 percent of same-sex sexual behavior. The rest of the 32 percent might involve genetic effects they could not measure, they said.

Researchers specifically identified five genetic variants present in people’s full genomes that appear to be involved. Those five comprise less than 1 percent of the genetic influences, they said.

And when the scientists tried to use genetic markers to predict how people in unrelated data sets reported their sexual behavior, it turned out to be too little genetic information to allow such prediction.

“Because we expect the sum of the effects that we observe will vary as a function of society and over time, it will be basically impossible to predict one’s sexual activity or orientation just from genetics,” said Andrea Ganna, the study’s first author, whose affiliations include the Institute of Molecular Medicine in Finland.

While many genetic variants tend to have the same effect in both men and women, Dr. Mills said, two of the five variants the team found were discovered only in males and one was discovered only in females. One of the male variants might be related to sense of smell, which is involved in sexual attraction, the researchers report. The other male variant is associated with male pattern balding and sits near genes involved in male sex determination.

Steven Reilly, a geneticist on the steering committee of the Broad Institute’s L.G.B.T.Q. affinity group, said of the study, “It seems like something that could easily be misconstrued or it seems problematic.”Kayana Szymczak for The New York Times

In a finding that could be especially sensitive, the researchers found that whether someone ever engaged in same-sex sexual behavior showed genetic correlations with mental health issues, like major depressive disorder or schizophrenia, and with traits like risk-taking, cannabis use, openness to experience and loneliness.

They emphasized that the study does not suggest that same-sex sexual behavior causes or is caused by these conditions or characteristics, and that depression or bipolar disorder could be fueled by prejudicial social experiences.

“We are particularly worried that people will misrepresent our findings about mental health,” Dr. Neale said.

“That right there is the big issue with looking for the genetics of sexual orientation — social context could be a big part of the expression of the trait,” said Jeremy Yoder, an assistant professor of biology at California State University, Northridge, who is gay and follows genetic research in the field.

Dr. Neale said increasing social acceptance might be reflected in the fact that younger study participants were much more likely than older ones to report same-sex sexual experiences. He and others noted that older participants came of age when homosexual behavior was criminalized in Britain and that for much of their life homosexuality was classified as a psychiatric disorder.

Dr. Reilly and others said such stark differences between older and younger participants show the trickiness of trying to draw representative biological information from a study population so strongly influenced by society’s changing attitudes. People steeped in a culture that demonized same-sex encounters might only have the gumption to admit it in a study if they were risk-takers to begin with.

Later, the researchers compared the genetic underpinnings of whether people ever had same-sex sex with their answers to what proportion of same-sex partners they had. They found there was little genetic correlation between answers to the “ever-never” question and whether someone ended up having a bisexual mix of partners, said Dr. Neale, who sees those results as a genetic reflection of the variety of sexual orientations within the expanding alphabet of the L.G.B.T.Q. community.

The researchers also looked at answers to other questions in the 23andMe survey, including people’s sexual identity and what gender they fantasized about. There, they found considerable genetic overlap between those results and whether people ever engaged in same-sex sex, suggesting that these aspects of sexual orientation share common genetics, they said.

Dean Hamer, a former National Institutes of Health scientist who led the first high-profile study identifying a genetic link to being gay in 1993, said he was happy to see such a large research effort.

“Having said that, I’d like to emphasize that it’s not a gay gene study — it’s a study of what makes people have a single same-sex experience or more,” said Dr. Hamer, now an author and filmmaker. The gene he identified was on the X chromosome, one of the sex chromosomes, a location the new study did not flag as being significant for same-sex sexual behavior.

“Of course they didn’t find a gay gene — they weren’t looking for one,” Dr. Hamer said.

Experts widely agree that the research was conducted by first-rate scientists.

“I kind of held my breath when I first saw the study — I thought, oh no,” said Dr. Mills of Oxford. “But it’s the top geneticists and some of the top social scientists in the field working on this, so if somebody was going to do it, I’m glad they did it.”

Indeed, Dr. Neale, who also consults for several pharmaceutical companies, said one reason his team did the study was to ensure less careful researchers would not tackle it first, “given how sensitive and hot-button this topic really is and how personal it is.”

Robbee Wedow, a member of the research team who also belongs to Out@Broad, served as a kind of bridge, organizing meetings between the researchers and their Broad Institute critics.

“I grew up in a highly religious evangelical family,” said Dr. Wedow, a research fellow with the Broad Institute and Harvard’s sociology department. “Being confused about not being attracted to women and being attracted to men, being convinced it was a sin and that I would go to hell.”

For a long time, “I definitely tried to pray it away, tried to like girls, tried to have girlfriends,” he said. “This wasn’t something I, of all people, would have chosen. There must be some sort of biological background.”

He concluded: “Saying ‘sorry, you can’t study this’ reinforces it as something that should be stigmatized.”

Outside L.G.B.T.Q. groups that were consulted did not seem as strongly concerned as some of the Out@Broad members, he said. Zeke Stokes, chief programs officer at GLAAD, who was shown the findings several months ago, said, “Anyone who’s L.G.B.T.Q. knows that their identity is complicated and to have science sort of bear that out is a positive thing.”

Over all, Dr. Neale said he believes the study shows that “diversity is a natural part of our experience and it’s a natural part of what we see in the genetics. I find that to actually just be beautiful.”

[Like the Science Times page on Facebook. | Sign up for the Sci

10 differences between artificial intelligence and human intelligence

Posted by Sabine Hossenfelder 

http://backreaction.blogspot.com/2019/08/10-differences-between-artificial.html?utm_source=feedburner&utm_medium=email&utm_campaign=Feed%3A+blogspot%2Fermku+%28Backreaction%29

Today I want to tell you what is artificial about artificial intelligence. There is of course, the obvious, which is that the brain is warm, wet, and wiggly, while a computer is not. But more importantly, there are structural differences between human and artificial intelligence, which I will get to in a moment.

Before we can talk about this though, I have to briefly tell you what “artificial intelligence” refers to.

What goes as “artificial intelligence” today are neural networks. A neural network is a computer algorithm that imitates certain functions of the human brain. It contains virtual “neurons” that are arranged in “layers” which are connected with each other. The neurons pass on information and thereby perform calculations, much like neurons in the human brain pass on information and thereby perform calculations.

In the neural net, the neurons are just numbers in the code, typically they have values between 0 and 1. The connections between the neurons also have numbers associated with them, and those are called “weights”. These weights tell you how much the information from one layer matters for the next layer.

The values of the neurons and the weights of the connections are essentially the free parameters of the network. And by training the network you want to find those values of the parameters that minimize a certain function, called the “loss function”.

So it’s really an optimization problem that neural nets solve. In this optimization, the magic of neural nets happens through what is known as backpropagation. This means if the net gives you a result that is not particularly good, you go back and change the weights of the neurons and their connections. This is how the net can “learn” from failure. Again, this plasticity mimics that of the human brain.

For a great introduction to neural nets, I can recommend this 20 minutes video by 3Blue1Brown.

Having said this, here are the key differences between artificial and real intelligence.

1. Form and Function

A neural net is software running on a computer. The “neurons” of an artificial intelligence are not physical. They are encoded in bits and strings on hard disks or silicon chips and their physical structure looks nothing like that of actual neurons. In the human brain, in contrast, form and function go together.

2. Size

The human brain has about 100 billion neurons. Current neural nets typically have a few hundred or so.

3. Connectivity

In a neural net each layer is usually fully connected to the previous and next layer. But the brain doesn’t really have layers. It instead relies on a lot of pre-defined structure. Not all regions of the human brain are equally connected and the regions are specialized for certain purposes.

4. Power Consumption

The human brain is dramatically more energy-efficient than any existing artificial intelligence. The brain uses around 20 Watts, which is comparable to what a standard laptop uses today. But with that power the brain handles a million times more neurons.

5. Architecture

In a neural network, the layers are neatly ordered and are addressed one after the other. The human brain, on the other hand, does a lot of parallel processing and not in any particular order.

6. Activation Potential

In the real brain neurons either fire or don’t. In a neural network the firing is mimicked by continuous values instead, so the artificial neurons can smoothly slide from off to on, which real neurons can’t.

7. Speed

The human brain is much, much slower than any artificially intelligent system. A standard computer performs some 10 billion operations per second. Real neurons, on the other hand, fire at a frequency of at most a thousand times per second.

8. Learning Technique

Neural networks learn by producing output, and if this output is of low performance according to the loss function, then the net responds by changing the weights of the neurons and their connections. No one knows in detail how humans learn, but that’s not how it works.

9. Structure

A neural net starts from scratch every time. The human brain, on the other hand, has a lot of structure already wired into its connectivity, and it draws on models which have proved useful during evolution.

10. Precision

The human brain is much more noisy and less precise than a neural net running on a computer. This means the brain basically cannot run the same learning mechanism as a neural net and it’s probably using an entirely different mechanism.

A consequence of these differences is that artificial intelligence today needs a lot of training with a lot of carefully prepared data, which is very unlike to how human intelligence works. Neural nets do not build models of the world, instead they learn to classify patterns, and this pattern recognition can fail with only small changes. A famous example is that you can add small amounts of noise to an image, so small amounts that your eyes will not see a difference, but an artificially intelligent system might be fooled into thinking a turtle is a rifle.

Neural networks are also presently not good at generalizing what they have learned from one situation to the next, and their success very strongly depends on defining just the correct “loss function”. If you don’t think about that loss function carefully enough, you will end up optimizing something you didn’t want. Like this simulated self-driving car trained to move at constant high speed, which learned to rapidly spin in a circle.

But neural networks excel at some things, such as classifying images or extrapolating data that doesn’t have any well-understood trend. And maybe the point of artificial intelligence is not to make it all that similar to natural intelligence. After all, the most useful machines we have, like cars or planes, are useful exactly because they do not mimic nature. Instead, we may want to build machines specialized in tasks we are not good at.