# News this Week

Science  16 May 2014:
Vol. 344, Issue 6185, pp. 676
1. # Around the World

1 - College Park, Maryland
El Niño Coming Back
2 - Canberra
Budget Slashes Science Funding
3 - Palo Alto, California
Stanford Strips Coal Investments From Endowment
4 - Brussels
E.U. Unveils University Ranking System

## College Park, Maryland

### El Niño Coming Back

El Niño, a periodic warming in the eastern tropical Pacific Ocean, has a 58% chance of emerging in the next 3 months, according to an 8 May forecast by the Climate Prediction Center (CPC) of the National Oceanic and Atmospheric Administration. That likelihood rises to 78% by the early fall, CPC found.

If strong, the El Niño event could wreak havoc on weather around the world, kicking off weather patterns that can cause droughts, storms, fires, and floods. The last El Niño occurred in 2009 to 2010, and the last big one, in 1997 to 1998, caused billions of dollars of damage worldwide.

A strong El Niño event could also trigger a resumption of global warming, seemingly stalled for the last 15 years. The Pacific Decadal Oscillation (PDO), a 20- to 30-year climate cycle, has been in a cool phase for nearly 20 years. But a strong El Niño event could push the PDO back into a warm phase, allowing heat from the ocean to reach the atmosphere—and causing a jump in atmospheric global warming. http://scim.ag/elnino2014

## Canberra

### Budget Slashes Science Funding

Australia's new conservative government released its first federal budget on 13 May, with major cuts to science funding outside of biomedical research. Spending at major science agencies will be at least AU$420 million less than envisioned by government projections: The Commonwealth Scientific and Industrial Research Organisation, for instance, would lose AU$111.4 million, the Defence Science and Technology Organisation AU$120 million, the Australian Nuclear Science and Technology Organisation AU$27.6 million, and the Australian Institute of Marine Science AU$7.8 million. The big winner in the budget is biomedical research; the government will establish a Medical Research Future Fund, and AU$42 million will go to expand the Australian Institute of Tropical Health and Medicine at James Cook University. http://scim.ag/AUbud

4. # Beyond the Temples

Turning their backs on spectacular monuments, archaeologists are studying ordinary households to uncover the daily rhythms of long-lost cities.

TAPACHULA, MEXICO—Kneeling in the cacao tree-shaded ruins of a 2000-year-old house, Rebecca Mendelsohn carefully scrapes soil off the fractured edge of a red ceramic plate and into a plastic bag. The archaeology graduate student from the University at Albany, State University of New York (SUNY), will bring hundreds of such samples to a lab in the mountain city of San Cristóbal de las Casas, where she will analyze them for traces of the food that the mysterious residents of Izapa, one of Mesoamerica's earliest cities, prepared and ate.

Izapa, 10 kilometers outside of the modern city of Tapachula in the Soconusco region of Chiapas state, arose around 850 B.C.E., possibly as people moved north from the Guatemalan coast to take advantage of a better climate for growing maize. Over at least the next 800 years, Izapa became the major economic and cultural hub along a trade route linking Olmec cities of the Gulf Coast and Maya strongholds in Central America.

Why Izapa flowered and who its inhabitants were are riddles that Mendelsohn hopes to solve from the bottom up. By excavating in several places around Izapa's periphery, she aims to compare the jobs, possessions, diet, and economic well-being of the city's residents, and how those patterns changed over time. And by plotting that information on a map that she and her adviser, Robert Rosenswig, created by surveying the site with an airborne laser, Mendelsohn hopes to uncover something that past archaeologists never expected to find in the region's ancient settlements: neighborhoods.

Studying pyramids and deciphering cryptic writing systems have helped archaeologists piece together the political, cultural, and religious characteristics of many Mesoamerican civilizations. But ceremonial architecture and official records may not reveal how societies actually work. "Tell me what the normal people were doing," Mendelsohn says. "That won't be on your monuments." Mapping lost neighborhoods can help archaeologists see an ancient city through the eyes of its residents, rather than through its leaders. What Mendelsohn and others are discovering through their bottom-up approach to places like Izapa are cities that look like nothing found in the modern world.

## The new urbanism

Cities, and therefore neighborhoods, were once considered a rarity in ancient Mesoamerica. Scholars have long defined cities as places with large, densely packed populations, intertwined economic activities—trades such as tailors, jewelers, soldiers, or manual laborers—and, often, a splash of cultural diversity, says Michael Smith, an archaeologist at Arizona State University, Tempe. "That definition makes sense because it fits our preconceptions of what cities are like today."

Teotihuacan, which lies 50 kilometers northeast of Mexico City and was occupied from roughly 100 B.C.E. to 650 C.E., was one of the few Mesoamerican cities that conformed to those expectations. Centered on the imposing Pyramids of the Sun and the Moon, it was laid out on a grid, and its more than 100,000 residents, many living in apartment buildings, were crammed into just 20 square kilometers.

But many other settlements in the ancient world don't fit the high-density, modern sense of a city. The Maya capitals Tikal and Copán, for example, were long thought to consist of unoccupied temples and administrative buildings surrounded by a haphazard smattering of villages spread across vast swaths of landscape. But as archaeologists studied sprawling ancient cities in Africa and in Southeast Asia, including Cambodia's Angkor Wat, they developed a concept of low-density urbanism, which defines a city not by size or density, but by what Smith calls its "urban function": the economic, political, or religious effect it has on a hinterland, or what is today known as a metropolitan area. A city, in other words, casts a spell over an entire region.

This idea raised a tantalizing possibility for Mayanists and other Mesoamerican archaeologists working at sites other than Teotihuacan: What if the landscape around a ceremonial center were not wasteland, but hinterland? What if the clusters of households weren't independent, self-governing villages, but rather interconnected nodes in a complex network of neighborhoods—one that was just as urban as high-density Teotihuacan?

## Hidden patterns

At Izapa, Mendelsohn is looking for signs of those neighborhoods. On a map, she points to a cluster of small structures near a larger mound far to the southeast of downtown. These are the households she's excavating in the cacao field. Another of her field teams is excavating a mound a few hundred meters to the west. Mendelsohn hopes to discern, 2000 years later, whether these were separate neighborhoods.

Preliminary findings hint at a major lifestyle gap. The ceramics and grinding stones found in the cacao-shaded dwellings suggest an abode of commoners. The pits at the mound to the west have yielded valuable jade beads, imported obsidian, and high-status pottery. Quite simply, Mendelsohn says, it "seems like rich people were here."

That may not mean, however, that all their neighbors were well-to-do. In Teotihuacan, for example, neighborhoods were economically mixed, says Ian Robertson, an archaeologist who studied Teotihuacan at Stanford University in California. Elite residences, temples, and administrative buildings were scattered throughout the ancient city, and the poor lived alongside the wealthy.

Some archaeologists believe Teotihuacan's economic mosaic offers a clue to its political structure. Based on 8 years of excavations in the center of an economically diverse neighborhood dubbed Teopancazco, Linda Manzanilla, an archaeologist at the National Autonomous University of Mexico in Mexico City, has proposed that Teotihuacan operated as a collection of "house societies," in which neighborhood leaders commanded the labor and loyalty of nearby lower status residents. This relationship resembled the feudal system of medieval Europe, she explains.

If it holds up under future excavations, the house society model may explain how Teotihuacan first formed, Robertson says. "Compared to most other cities, [Teotihuacan] was there very fast"—faster than the city's birthrate could account for. That means that people must have been migrating to the new city in droves. "I imagine that sometimes what you had were whole towns pulling up stakes, moving into Teotihuacan, and reproducing their own social structure in the new urban center," he proposes. As time went on, those uprooted towns likely morphed into neighborhoods.

Marilyn Masson, an archaeologist at the University at Albany, SUNY, sees a similar pattern in Mayapán, an ancient city in the dry interior of the Yucatán Peninsula. From the early 1200s C.E. to its collapse in the 1400s, Mayapán was the political and economic capital of the Maya area. Spanish accounts of Mayapán's history suggest that as the city consolidated, its rulers—who came from all over the peninsula—"forcibly relocated" people to populate the new center, Masson says. Mayapán's first neighborhoods, therefore, probably would have resembled the uprooted villages.

Masson believes she's found the primary way order was imposed on Mayapán's unruly melting pot: walls. Walls were everywhere in the city, and after years of fieldwork and a laser mapping survey in 2013, Masson and Timothy Hare, an anthropologist at Morehead State University in Kentucky, have mapped every single one. "With the walls, [the city] starts making sense," Hare says.

Not only did Mayapán boast a defensive wall surrounding a dense 4-square-kilometer downtown, but smaller walls also separated houses from one another, demarcated roads, and, overall, organized how people moved through the city. They also dictated how buildings were oriented, a key feature of neighborhoods. "When people walk out of the front door of their houses, what do they see?" Masson asks. Likely, the dwellings would be facing their neighborhood's "most important symbol." Downtown, she says, that would be the temples and administrative buildings in Mayapán's ceremonial center. But farther out—and especially in low-density areas beyond the city's defensive wall—the buildings tend to be oriented toward complexes of elite residences and smaller ritual buildings. According to Masson, those secondary centers probably played a role in knitting the surrounding households into neighborhood communities, and perhaps served as landmarks to help residents navigate the city.

## Ethnic diversity

Back in the cacao field, Mendelsohn points out an odd feature of the mound she's excavating: The dwellings here are arranged around a shared patio. This architectural pattern is more common in the Maya area to the south and east. She wonders if she might be looking at an ethnic enclave. Izapa's location along a major trade route meant that Maya and Olmec people probably passed through frequently, and it's possible some of them stayed. "It would be completely reasonable to find a Maya neighborhood or an Olmec neighborhood" in Izapa, Mendelsohn says.

Although most archaeologists, including Mendelsohn, doubt that artifacts can reliably distinguish their owners' ethnicity thousands of years later, at least two ethnic enclaves have been identified at Teotihuacan. A cluster of apartment compounds on the city's western edge contained ceramics resembling those made by the Zapotecs, who lived 500 kilometers to the south in modern-day Oaxaca. These out-of-place cooking tools and ritual objects were made of local clay but fired using an Oaxacan technique that darkened the pottery. Later excavations of the area uncovered an ornate Zapotec-style tomb and a temple clearly modeled after its cousins in Oaxaca—structures that, according to Michael Spence, an archaeologist at the University of Western Ontario in London, Canada, have "got no business being in Teotihuacan." Meanwhile, a separate cluster of similarly anomalous pottery styles from the Gulf Coast and the Maya region was found in apartment compounds on the other side of the city. The two enclaves were the first neighborhoods identified in Teotihuacan.

At Izapa, neighborhood outlines are more elusive. Its city plan is still coming into view; Rosenswig's laser survey doubled the early city's estimated size, revealing dozens of hidden mounds. And now Mendelsohn's excavations hint that Izapa may have been occupied far longer than anyone thought. Its neighborhoods had centuries to shift borders and change character, adding texture to the city—and intrigue to its ruins.

5. # The Hunt for Missing Genes

1. Jocelyn Kaiser

Identifying healthy human "knockouts"—people completely lacking a specific gene—may suggest new biomedical treatments.

Daniel MacArthur's quest for the genes we can live without began with two sick boys. In 2000, as an undergraduate student, he began working in the laboratory of Kathryn North, a geneticist who studies rare, mysterious muscle diseases. Her group at the University of Sydney in Australia had recently published the possible cause of two brothers' early-onset form of muscular dystrophy. They suspected a pair of faulty copies of ACTN3, a gene that codes for a protein in the fast-twitch muscles that generate short bursts of power.

But later, the group tested the parents to confirm what seemed obvious: that each had two versions of ACTN3—a working copy and a broken one—and both had passed on the latter to their ill children. To the team's surprise, the parents, like their sons, both lacked any functioning copy of the gene or any trace of its protein in their muscles. Yet both mother and father appeared healthy.

North's group eventually did find another mutated gene responsible for the boys' disease, but at the time they also realized they had documented something significant. "It was actually incredibly exciting," says MacArthur, who joined the lab soon after the misstep was recognized. "It was one of the first examples of a gene that should be important" but that people can live without.

Nor was that family a rarity: North's group found that about 16% of the global population had two broken copies of the gene, without obvious disease as a result. As he continued in North's lab into graduate school, MacArthur and co-workers also discovered that the functional version of the gene was more prevalent in Australia's elite sprinters and jumpers.

But MacArthur became most intrigued by the prospect that hidden in the human population were people who lacked certain genes yet remained healthy. Researchers routinely disable, or "knock out," a specific gene in mice to learn what the gene does, but the results don't always translate into people. Ethically, knocking out genes in humans is off-limits. But the sick children's parents offered another route—finding natural human knockouts and looking for differences between their physiology and that of people with the intact gene. "I became fascinated by the idea that these individuals serve as experiments of nature," says MacArthur, now at Massachusetts General Hospital (MGH) in Boston and the Broad Institute in Cambridge, Massachusetts.

Those natural experiments could have biomedical payoffs. Researchers often seek new drug targets by identifying genes that cause disease when mutated and looking for molecules that can compensate. Healthy knockouts suggest a different approach: hunting for genes that, when missing, actually confer a health benefit, then trying to mimic that effect by blocking the normal gene's protein. AIDS researchers have already found that certain people lacking a working gene for a specific cell surface protein suffer no ill effects and are resistant to HIV; that protein is now a drug target and figures into other anti-HIV strategies. More recently, pharmaceutical firms began developing a potential new blockbuster class of cholesterol-lowering drugs, inspired by a woman missing a cholesterol-regulating gene.

In a survey of scores of human genomes 2 years ago, MacArthur and co-workers caught a glimpse of a bigger universe of missing genes. The survey showed that the average healthy person has about 20 genes knocked out. Now, he and several other groups want to similarly comb through many more thousands of people's genomes for missing genes to seed what MacArthur calls the Human Knockout Project.

Such an effort may be the only way to fully understand the function of many of our genes, he and others contend. But identifying missing genes is only the start of the challenge. Then, investigators have to link a dispensable gene to the human knockout's phenotype—their health measures and other traits—and that information is costly and time-consuming to collect. "Getting good phenotypes makes genome sequencing look cheap (and easy)," says cardiovascular disease researcher Helen Hobbs of the University of Texas (UT) Southwestern Medical Center in Dallas.

## The poster gene

Hobbs speaks from experience. About a decade ago, she and her UT Southwestern colleague Jonathan Cohen wondered if mutations that hampered or disabled a gene called PCSK9 explained why some people in the Dallas Heart Study had unusually low levels of harmful cholesterol. Recent studies had suggested that the gene's product, an enzyme, might regulate the body's blood levels of low-density lipoprotein (LDL) cholesterol, the type that raises the risk of heart disease.

The pair's hypothesis proved to be correct. They even found one 34-year-old woman who was a PCSK9 knockout—she completely lacked the enzyme due to mutations in both the genes for it. She had the lowest LDL cholesterol levels of all, yet was perfectly healthy. Based on the team's work, drug companies realized that blocking the PCSK9 enzyme might lower cholesterol levels alone or in combination with the immensely popular statins.

That hunch is paying off in clinical trials of PCSK9-inhibiting drugs; early results suggest they can lower blood cholesterol levels by up to 57%. PCSK9 has become the poster child for the idea that genes whose absence confers some benefit can be an extremely attractive drug target. The very existence of a healthy knockout shows that the gene isn't essential, so blocking its protein to mimic the beneficial effect should not cause harmful side effects. The human knockout is "a shortcut. You know you could inhibit" the gene's protein, says human geneticist David Altshuler of the Broad Institute.

Until recently, however, no one knew how many genes like PCSK9 existed. That's why MacArthur did his initial survey. After moving from Australia to a postdoc at the Sanger Institute in the United Kingdom, MacArthur and his team scanned the genomes of 185 individuals who were part of a study of human genetic variation. They homed in on DNA errors that incapacitate a gene, known as loss-of-function mutations. Because not all DNA alterations do so—and many apparent DNA errors are actually sequencing mistakes—identifying true gene knockouts was a huge analytical task.

This laborious effort paid off, revealing that the average person carries about 100 incapacitated genes—and in 20 of those cases, both the maternal and paternal copies of a gene are missing, creating a complete knockout, the team reported 2 years ago in Science (17 February 2012, p. 823). "More than a few of us were surprised" by such a large number, says human geneticist David van Heel of Queen Mary, University of London.

Many of the more common missing genes were involved in smell; they may have been important for helping our ancestors find food, but are unlikely to affect a modern human's fitness. Others belonged to families of related genes that serve similar functions, suggesting the missing genes were not needed because the cell has backups.

Still, that left a small, but not insignificant, number of missing genes that just might protect against disease. Other scientists took notice. National Institutes of Health (NIH) Director Francis Collins remarked at a meeting that MacArthur's study suggested "a systematic, comprehensive way of identifying where the other several dozen PCSK9s might be out there."

## Hunting for gold

Even before MacArthur's paper, other disease researchers had followed Hobbs and Cohen's example and begun looking at the genomes of people who seem protected against disease by missing genes. For example, earlier this year, Altshuler and others reported finding a gene that, when one copy is nonfunctional, lowers a person's risk of type 2 diabetes—by a stunning 65%.

Now, MacArthur and other knockout seekers plan to widen the search. They will sequence the DNA of a large number of healthy people, see who lacks potentially interesting genes, then study whether those individuals are somehow protected—if they're less prone to heart attacks or high blood pressure, for example—or if they are unusual in some other unexpected way. This strategy "allows you to discover things you didn't know were there," says a fan of the idea, human geneticist Leslie Biesecker of NIH.

The most efficient way to do this is not to look in the general population; novel knocked-out genes will be too rare. A faster way may be to study historically isolated populations, such as the Finns. Some 4000 years ago, presumably when a small number of settlers moved to Finland, that population passed through a bottleneck. Because of this initial small gene pool, the frequency of some loss-of-function mutations is "enriched," says Aarno Palotie of the Broad Institute and the University of Helsinki. As part of a project pooling samples and data on 200,000 Finns from various biobanks, he and Mark Daly of MGH and the Broad Institute led a pilot study in which they scoured the protein-coding portions of the genome, or exomes, of 3000 Finns; they found twice as many knocked-out genes as in a northern European comparison group.

Marrying that work with DNA data and health records for 35,000 Finns has already led to potential gold: 227 individuals entirely lacked a gene called LPA that codes for a blood lipoprotein implicated in cardiovascular disease. These people had a significantly lower risk of heart attacks and stroke, according to a retrospective analysis that the team presented last fall at the annual meeting of the American Society of Human Genetics.

"This is a proof of concept that, indeed, you can find protective loss-of-function variants using isolated populations," Palotie says. "The excitement is that we can find many more of these." Finland's vast electronic system of medical and death records should give the team an advantage in the hunt.

Cultural traditions may also help the search for knockouts. In some cultures, it is common practice for relatives as close as first cousins to marry. That increases the odds a child will inherit two copies of the exact same nonfunctioning genetic variant, says Yale University human geneticist Richard Lifton. This has a downside if the mutations are in necessary genes—countries such as Turkey and Saudi Arabia have relatively high rates of inherited genetic diseases—but it ups the odds of finding beneficial cases.

To take advantage of this, van Heel and Richard Trembath, his colleague at Queen Mary, University of London, along with MacArthur and others, plan to sequence the exomes of up to 25,000 adults of Pakistani and Bangladeshi descent living in East London. They are now engaging with community leaders to build support for the Wellcome Trust–funded project and plan to begin the search next year. In a pilot study looking at the genomes of 1103 healthy British Pakistani individuals, they've already found about 200 potentially interesting knocked-out genes that differ from the ones found in other studies. "Some 20 are reportedly lethal in mice [if knocked out], but clearly these fit adults are doing just fine," van Heel says.

In another effort, starting this summer, a team in Saudi Arabia will collect blood samples and basic clinical data from volunteers recruited in public areas such as malls and parks—"places where healthy people are likely to go," says study leader Fowzan Alkuraya of King Faisal Specialist Hospital and Research Center in Riyadh. The goal is to enroll 10,000 people whose parents are first cousins and sequence their exomes over the next year; it is a subproject of the country's recently announced 100,000 genomes project. Like other knockout hunters, the team plans to later recall individuals with interesting loss-of-function genes for more clinical testing. "I'm pretty sure we're going to find something as exciting as the PCSK9 story," Alkuraya says.

## A lofty goal

Not everybody thinks that the search for knockouts should focus exclusively on healthy people. Lifton suggests that more drug targets will come from people in whom a missing gene causes a disease or some kind of obvious abnormality. He cites a gene involved in making a neurotransmittor that causes narcolepsy when both copies are disabled. Studying this gene has led to a new kind of sleeping pill, now awaiting regulatory approval. Another example is the rare individuals who feel no pain because they lack a gene for a particular cell receptor. The gene could eventually result in a new class of painkillers.

Hobbs and Cohen add that sequencing large groups of healthy people without having good clinical information to guide the search may lead nowhere. They credit their success at identifying the PCSK9 knockout and several others to having years of detailed clinical data on people they have studied. When they find a mutation, they can immediately know a lot about the medical consequences, the pair says.

That is especially important because the consequences of missing a gene won't always be straightforward. The missing gene may have no effect in some people; or it may matter only later in life, or when the person eats certain plants or is exposed to a specific disease, Hobbs says.

MacArthur has heard the warnings, and he plans to bring together both healthy knockouts and those linked to disease, along with as much clinical data as he can amass, in one loss-of-function mutation database—the foundation, he hopes, for an eventual Human Knockout Project. Besides gathering published data on diseases caused by nonfunctioning genes, his group is building a list of novel knocked-out genes by combing through the exomes of more than 80,000 ill and healthy people sequenced for disease studies at the Broad Institute and other research centers. Eventually, he hopes other researchers, including those working in the United Kingdom, Finland, and Saudi Arabia, will contribute their knockouts, along with data on clinical consequences.

The proposal is "massively ambitious," MacArthur acknowledges, and will require buy-in from the broader genetics community. For now, he adds, "This is the beginning of an idea."

For all the voices of caution, few doubt that tallying up the world's human knockouts is a worthy goal. "We think the key is to have an outstanding, centralized database," Hobbs and Cohen note in an e-mail, "that contains systematic phenotypic information in each human knockout." Adds Lifton: "It's an obvious thing to do."