Breakthrough of the year

Development cell by cell

With a trio of techniques, scientists are tracking embryo development in stunning detail

Spherical zebrafish embryo in early stage of development.

A zebrafish embryo at an early stage of development. Fluorescent markers highlight cells expressing genes that help determine the type of cell they will become. (Jeffrey Farrell, Schier Lab/Harvard University)

From at least the time of Hippocrates, biologists have been transfixed by the mystery of how a single cell develops into an adult animal with multiple organs and billions of cells. The ancient Greek physician hypothesized that moisture from a mother’s breath helps shape a growing infant, but now we know it is DNA that ultimately orchestrates the processes by which cells multiply and specialize. Now, just as a music score indicates when strings, brass, percussion, and woodwinds chime in to create a symphony, a combination of technologies is revealing when genes in individual cells switch on, cueing the cells to play their specialized parts. The result is the ability to track development of organisms and organs in stunning detail, cell by cell and through time. Science is recognizing that combination of technologies, and its potential for spurring advances in basic research and medicine, as the 2018 Breakthrough of the Year.

Driving those advances are techniques for isolating thousands of intact cells from living organisms, efficiently sequencing expressed genetic material in each cell, and using computers, or labeling the cells, to reconstruct their relationships in space and time. That technical trifecta “will transform the next decade of research,” says Nikolaus Rajewsky, a systems biologist at the Max Delbrück Center for Molecular Medicine in Berlin. This year alone, papers detailed how a flatworm, a fish, a frog, and other organisms begin to make organs and appendages. And groups around the world are applying the techniques to study how human cells mature over a lifetime, how tissues regenerate, and how cells change in diseases.

The ability to isolate thousands of individual cells and sequence each one’s genetic material gives researchers a snapshot of what RNA is being produced in each cell at that moment. And because RNA sequences are specific to the genes that produced them, researchers can see which genes are active. Those active genes define what a cell does.

That combination of techniques, known as single-cell RNA-seq, has evolved over the past few years. But a turning point came last year, when two groups showed it could be done on a scale large enough to track early development. One group used singlecell RNA-seq to measure gene activity in 8000 cells extracted at one time point from fruit fly embryos. About the same time, another team profiled gene activity of 50,000 cells from one larval stage of the nematode Caenorhabditis elegans. The data indicated which proteins, called transcription factors, were guiding the cells to differentiate into specialized types.

This year, those researchers and others performed even more extensive analyses on vertebrate embryos. Using a variety of sophisticated computational methods, they linked single-cell RNA-seq readouts taken at different time points to reveal the turning on and off of sets of genes that defined the types of cells formed in those more complex organisms. One study uncovered how a fertilized zebrafish egg gives rise to 25 cell types; another monitored frog development through early stages of organ formation and determined that some cells begin to specialize earlier than previously thought. “The techniques have answered fundamental questions regarding embryology,” says Harvard University stem cell biologist Leonard Zon.

Researchers interested in how some animals can regrow limbs or whole bodies have also turned to single-cell RNA-seq. Two groups studied gene expression patterns in aquatic flatworms called planaria—among biology’s champion regenerators—after theyhad been cut into pieces. The scientists discovered new cell types and developmental trajectories that emerged as each piece regrew into a whole individual. Another group traced the genes that switched on and off in axolotls, a type of salamander, that had lost a forelimb. The researchers found that some mature limb tissue reverted to an embryonic, undifferentiated state and then underwent cellular and molecular reprogramming to build a new limb.

Because cells must be removed from an organism for single-cell sequencing, that technique alone can’t show how those cells interact with their neighbors or identify the cells’ descendants. But by engineering markers into early embryonic cells, researchers can now track cells and their progeny in living organisms. At least one team exposes early embryos to mobile genetic elements that carry genes for different colored fluorescent tags, which randomly settle into the cells, imparting different colors to each cell lineage. Other teams have harnessed the gene-editing technique called CRISPR to mark the genomes of individual cells with unique barcodelike identifiers, which are then passed on to all their descendants. The gene editor can make new mutations in progeny cells while retaining the original mutations, enabling scientists to track how lineages branch off to form new cell types.

By combining those techniques with single-cell RNA-seq, researchers can both monitor the behavior of individual cells and see how they fit into the organism’s unfolding architecture. Using that approach, one team determined the relationships of more than 100 cell types in zebrafish brains. The researchers used CRISPR to mark early embryonic cells, then isolated and sequenced 60,000 cells at different time points to track gene activity as the fish embryo developed.

Other groups are applying similar techniques to track what happens in developing organs, limbs, or other tissues—and how those processes can go wrong, resulting in malformations or disease. “It’s like a flight recorder, where you are watching what went wrong and not just looking at a snapshot at the end,” says Jonathan Weissman, a stem cell biologist at the University of California, San Francisco. “We can ask questions at a resolution that was just not possible before.”

Although those technologies cannot be used directly in developing human embryos, researchers are applying the approaches to human tissues and organoids to study gene activity cell by cell and characterize cell types. An international consortium called the Human Cell Atlas is 2 years into an effort to identify every human cell type, where each type is located in the body, and how the cells work together to form tissues and organs. Already, one project has identified most, if not all, kidney cell types, including ones that tend to become cancerous. Another effort has revealed the interplay between maternal and fetal cells that allows pregnancy to proceed. And a collaboration of 53 institutions and 60 companies across Europe, called the LifeTime consortium, is proposing to harness single-cell RNA-seq in a multipronged effort to understand what happens cell by cell as tissues progress toward cancer, diabetes, and other diseases.

High-resolution movies of development and disease will only get more compelling. Papers already posted online extend development studies to ever-more-complex organisms. And researchers hope to combine single-cell RNA-seq with new microscopy techniques to see where in each cell its distinctive molecular activity takes place and how neighboring cells affect that activity.

The single-cell revolution is just starting.


E. Pennisi, Chronicling embryos, cell by cell, gene by geneScience, Vol. 360, p. 367, 27 April 2018

R. M. Harland, A new view of embryo development and regenerationScience, Vol. 360, p. 967, 1 June 2018

B. Pijuan-Sala et al., Single-cell transcriptional profiling: a window into embryonic cell-type specificationNature Reviews Molecular Cell Biology, Vol. 19, p. 399, 17 April 2018

B. Raj et al., Simultaneous single-cell profiling of lineages and cell types in the vertebrate brainNature Biotechnology, Vol. 36, p. 442, 28 March 2018

B. Spanjaard et al., Simultaneous lineage tracing and cell-type identification using CRISPR-Cas9-induced genetic scarsNature Biotechnology, Vol. 36, p. 469, 9 April 2018


Messengers from a far-off galaxy

Spherical detectors called DOMs buried in a grid under South Pole ice detect faint bursts of light as neutrinos from distance stellar events collide with atoms.

An illustration of detectors buried in ice beneath the South Pole that record rare flashes triggered by neutrinos. (Jamie Yang and Savannah Guthrie. IceCube/NSF)

Few kinds of messengers from the distant universe are joining the photons collected by telescopes—and revealing what light can’t show. So-called multimessenger astrophysics got started with high-speed particles called cosmic rays and gravitational waves, the ripples in space-time first detected in 2015 that Science named Breakthrough of the Year in 2016. This year, another messenger has joined the party: neutrinos, tiny, almost massless particles that are extraordinarily hard to detect.

Snaring one of these extra-galactic will-o’-the-wisps took a cubic kilometer of ice deep below the South Pole, festooned with light detectors to record the faint flash triggered—very rarely—by a neutrino. Known as IceCube, the massive detector has logged many neutrinos before, some from outside the Milky Way, but none had been pinned to a particular cosmic source. Then, on 22 September 2017, a neutrino collided with a nucleus in the ice, and the light sensors got a good fix on the direction it had come from.

An alert sent out to other telescopes produced, after a few days, a match. As the researchers reported in July, NASA’s Fermi Gamma-ray Space Telescope found an intensely bright source known as a blazar right where the neutrino appeared to come from. A blazar is the heart of a galaxy centered on a supermassive black hole, whose gravity heats up gas swirling around it, causing the material to glow brightly and fire jets of particles out of the maelstrom.

Researchers are pretty sure the blazar, which was flaring up at the time of the detection, is the source of the neutrino—making it the first time a neutrino telescope has identified an extra-galactic source. But the discovery is more than just a proof of principle. A blazar producing gamma rays and neutrinos is likely producing other highenergy particles, too, such as protons. These ultra–high-energy cosmic rays bombard Earth from time to time, but their source has been a mystery. Now, blazars are a suspect.

The IceCube team awaits more fleeting extra-galactic messengers. But having welcomed this first visitor, it is making its case for an enlarged detector enclosing 10 times the current volume of ice.


Ice Cube Collaboration, Multimessenger observations of a flaring blazar coincident with high-energy neutrino IceCube-170922A, Science, Vol. 361, p. 147, 13 July 2018

Ice Cube Collaboration, Neutrino emission from the direction of the blazar TXS 0506+056 prior to the IceCube-170922A alert, Science, Vol. 361, p. 147, 13 July 2018

Molecular structures made simple

Black microcrystals rest on a surface dotted with holes 1 micron in diameter.

Structures can now be gleaned from micrometer-size crystals (black), seen here on an electron microscope slide. (Gonen Lab)

Two research teams simultaneously published papers in October revealing a new way to determine the molecular structures of small organic compounds in just minutes, rather than the days, weeks, or months required by traditional methods.

For decades, the gold standard for molecular mapping has been a technique known as x-ray crystallography, which involves firing a beam of x-rays at a crystal containing millions of copies of a molecule lined up in a common orientation. Researchers then track the way x-rays bounce off the crystal to identify individual atoms and assign them positions in the molecule. The structures are invaluable for understanding how biological molecules behave and how drugs interact with them. But the technique requires growing crystals about the size of a grain of sand, which can be a major hurdle for some substances.

In recent years, researchers have modified the diffraction technique by replacing the x-rays with an electron beam. The electron beam is aimed at a sheetlike 2D crystal of the target biomolecule, usually a protein. But in some cases, those sheets stack atop one another, creating a 3D crystal that doesn’t work for ordinary electron diffraction and is too small for x-ray diffraction.

Two research teams—one in the United States, the other in Germany and Switzerland—found they could use such accidental crystals after all. They fired an electron beam at a tiny 3D crystal on a rotating stage and tracked how the diffraction pattern changed with each slight turn. The technique generated molecular structures in minutes—from microscopic crystals just one-billionth the size required for x-ray studies.

Well-suited for mapping small molecules such as hormones and potential drugs, the new technique should have a profound impact on fields ranging from the synthesis and discovery of new pharmaceuticals to the design of molecular probes to study and track diseases.


R. F. Service, Molecular CT scan could speed drug discoveryScience, Vol. 362, p. 389, 26 October 2018

Ice age impact

Computer visualization of fragments falling earthward, the impact creating Hiawatha Crater in Greenland.

A computer visualization of asteroid fragments falling toward Greenland. (NASA Scientific Visualization Studio)

The asteroid slammed into northwestern Greenland like a fusillade of nuclear bombs, instantly vaporizing rock and sending shock waves across the Arctic. The scar it left—a 31-kilometer-wide impact crater called Hiawatha—is big enough to hold Washington, D.C. Scientists reported the startling discovery in November, after aircraft radar revealed the crater lurking beneath the kilometer-thick ice sheet.

Hiawatha crater is one of the 25 largest on Earth. Though not as cataclysmic as the dinosaur-killing Chicxulub impact, which carved out a 200-kilometer-wide crater in Mexico 66 million years ago, the Hiawatha impact could have had a powerful effect on the global climate. Meltwater from the impact, pouring into the north Atlantic Ocean, could have sent temperatures plunging by halting a conveyor belt of currents that brings warmth to northwest Europe.

The radar images suggest Hiawatha is exceptionally fresh, dating from the past 100,000 years. And a disturbance in the crater’s deep ice hints that the asteroid may have struck as recently as 13,000 years ago. That would tie the impact to the Younger Dryas, a thousand-year global cooling event that began just as the world was thawing from the last ice age. It would also vindicate proponents of the controversial Younger Dryas impact theory. A decade ago, they proposed that extraterrestrial impacts could account for hints of mayhem in the archaeological and geological record. But they could never point to a crater.

The timing of this impact is far from settled. Ice cores elsewhere on Greenland, which record the past 100,000 years, contain no signs of impact debris. A firm answer will depend on painstaking work to tease dates from the radioactive clocks in tiny mineral crystals swept from under the ice.

If they show the Hiawatha impact did occur 13,000 years ago, it would have come just as humans were fanning across a new continent, chasing mastodons around North America. It is tempting to imagine their thoughts as they looked up to see the searing white orb of the impactor, four times brighter than the sun.


K. Kjaer et al., A large impact crater beneath Hiawatha Glacier in northwest GreenlandScience Advances, Vol. 4, 14 November 2018

#MeToo makes a difference

Illustration showing women speaking into a microphone.

(Daria Kirpach/@Salzmanart)

Sexual harassment in science has been underreported and largely ignored until recently. But this year brought signs of change.

In June, the U.S. National Academies of Sciences, Engineering, and Medicine released a landmark report on sexual harassment of women in academic science, engineering, and medicine that could prove to be a watershed. It concluded, based on recent data from two large university systems, that more than 50% of female faculty and staff and 20% to 50% of students, depending on stage and field, have endured sexual harassment, including the most pervasive form— sexist hostility both verbal and nonverbal: putdowns, not come-ons. And this year, several institutions took action.

Some, prodded by news exposés or by formal complaints from harassed students and staff, fired or forced out prominent scientists after investigations upheld allegations of wrongdoing. Others announced policy changes.

In September, the U.S. National Science Foundation (NSF) in Alexandria, Virginia, said that, going forward, universities must tell it when a grantee is placed on leave during a harassment investigation or found guilty of sexual harassment—with the potential for “targeted and serious consequences” from NSF as a result, said Director France Córdova. Bemoaning the community’s failure to protect harassment victims, Córdova declared: “This neglect must end.” That same month, AAAS, which publishes Science, adopted a policy under which AAAS fellows proved to be sexual harassers can be stripped of this lifetime honor. And the presidents of the National Academies promised in May to explore how proven harassers could be ejected from their prestigious ranks.

The pace of change is not nearly fast enough for critics. BethAnn McLaughlin, a neuroscientist at Vanderbilt University in Nashville who this year founded the advocacy group #metooSTEM, notes, for example, that the National Institutes of Health (NIH) does not require universities to report grantees under investigation, or even disciplined, for sexual harassment. McLaughlin opens her public talks with 46 seconds of silence, “1 second for every year that NIH has given money to scientists and doctors and not asked if they have violated Title IX,” a 1972 U.S. statute outlawing sexual harassment of students. She uses the silence, she says, “to honor the hundreds of women driven from our fields.”


National Academies of Sciences, Engineering, and Medicine, Sexual Harassment of Women, 2018

An archaic human ‘hybrid’

Bone fragment showing evidence of ancient interspecies mating.

A bone fragment found in a cave in the Denisova valley in Russia. (Thomas Higham, University of Oxford)

A fragment of bone from a woman who lived more than 50,000 years ago has revealed a startling connection between two extinct groups of archaic humans. Ancient DNA extracted from the bone, found in a cave in Siberia in 2012, showed the woman’s mother was a Neanderthal and her father was a Denisovan, the mysterious group of ancient humans whose remains were discovered in the same cave in 2011.

Researchers knew Denisovans, Neanderthals, and modern humans interbred, at least occasionally, in ice age Europe and Asia. The genes of both types of archaic human are present in Asian and European people today. And other fossils found in the Siberian cave have shown members of all three groups lived there at different times. But the new finding is intimate testimony of an encounter between a Denisovan and a Neanderthal.

After sequencing DNA from the bone, researchers at the Max Planck Institute for Evolutionary Anthropology in Leipzig, Germany, found it came from a female, and that her genome matched those of Denisovans and Neanderthals in roughly equal measure. That could have been because her parents themselves had mixed ancestry. But her chromosome pairs harbored different variants—so-called heterozygous alleles—of nearly half her genes, suggesting the maternal and paternal chromosomes came from different kinds of humans. Her mitochondrial DNA, which is almost entirely inherited from the mother, was uniformly Neanderthal, so the researchers concluded she was a first generation hybrid of a Denisovan male and Neanderthal female. A closer look at the genome suggested her father also had some Neanderthal ancestry.

In another telltale finding, the woman’s Neanderthal genes are closer to those of a Neanderthal found in Croatia than to those of earlier Neanderthal inhabitants of the Denisova Cave. That suggests, the authors say, that distinct groups of Neanderthals migrated back and forth between western Europe and Siberia multiple times. Along the way, apparently, they freely spread their genes to outsiders. Why did Denisovans and Neanderthals remain genetically distinct? Geographic barriers probably played a role, but researchers need more ancient DNA from different sites to understand the true influence of these prehistoric couplings.


V. Slon et al., The genome of the offspring of a Neanderthal mother and a Denisovan fatherNature, Vol. 561, p. 113, 22 August 2018

Forensic genealogy comes of age

Joseph James DeAngelo in jail. The former police officer is accused of being the Golden State Killer, suspected in at least a dozen killings and roughly 50 rapes in the 1970s and '80s.

Joseph James DeAngelo, the alleged Golden State Killer. (Paul Kitagaki Jr./The Sacramento Bee via AP/Pool)

In April, police announced they had arrested a suspect in one of the coldest of cold cases: a series of rapes and murders in California in the 1970s and 1980s. It was a stunning development, and so was the way investigators fingered the alleged Golden State Killer. They identified his relatives by uploading a profile of DNA recovered from one of the crime scenes to a public genealogy DNA database. Law enforcement agencies have since used this strategy to crack about 20 other cold cases, ushering in a new field: forensic genealogy.

Private DNA websites such as Ancestry and 23andMe contain millions of profiles that can be used to find a person’s relatives from bits of shared DNA, but police need a court order to search them. In the Golden State Killer case, authorities turned to a public, no-frills online database called GEDMatch, run by two amateur genealogists in Texas and Florida, to which anyone can submit DNA test results. Investigators uploaded a DNA profile from a rape kit to the database, and found several distant relatives of the perpetrator. Working with a genealogist, they then used public records to construct large family trees and homed in on 73-year-old Joseph James DeAngelo, whose age and location fit some of the crimes. When tests showed the crime scene DNA matched DNA from DeAngelo’s car door handle and a discarded tissue, they had their suspect.

This fall, geneticists reported that 60% of Americans of European descent (who make up most ancestry site users) would have a third cousin or closer match in a database with 1 million samples, about the size of GEDMatch. Once a database reaches 3 million profiles, more than 90% of white individuals could be found with similar methods—even if they have never had their DNA tested. All this has alarmed some ethicists and geneticists who see these familial searches as an invasion of privacy with a potential for misidentifying suspects.


J. Kaiser, We will find you: DNA search used to nab Golden State Killer can home in on about 60% of white Americans, Science, 11 October 2018

Gene-silencing drug approved

Computer visualization of RNA chains delivered into a cell, then translated to proteins.

Short RNA molecules attach to messenger RNA (blue), preventing translation into proteins. (Val Altounian/Science)

A drug based on a gene-silencing mechanism called RNA interference (RNAi) won regulatory approval this year. The long-awaited step could be the harbinger of a new class of drugs targeting diseasecausing genes.

Twenty years ago, two U.S. geneticists discovered that short RNA molecules can disrupt the translation of genes by attaching to the messenger RNA that carries a gene’s message to the cell’s proteinmaking machinery. The advance won them a Nobel Prize, but efforts to turn it into medicine quickly hit hurdles. Scientists struggled to keep the fragile RNA molecules intact and direct them to the right tissue. By 2008, researchers at Cambridge, Massachusetts–based Alnylam Pharmaceuticals thought they had a solution: a lipid nanoparticle that would protect the gene-silencing RNA and ferry it to the liver. There, they hoped, it could treat a rare disease called hereditary transthyretin amyloidosis by blocking the production of a misfolded protein that builds up and causes heart and nerve damage.

“We set off with great haste and enthusiasm,” says Akshay Vaishnaw, Alnylam’s president of R&D. But the new nanoparticle didn’t release enough RNA into liver cells to knock down the problem gene effectively in all patients. A more potent formulation worked in human trials and became the intravenous drug Onpattro, which won clearance from U.S. and EU regulators this year and hit the market with a $450,000-per-year list price.

The approval, along with the 2016 approval of a different class of RNA-based drug, has invigorated the field, says Frank Slack, a developmental biologist at Beth Israel Deaconess Medical Center in Boston who studies another type of small RNA molecule. Many RNAi researchers are now shifting their attention to a newer delivery method: hooking chemically stabilized RNA onto a sugar molecule that homes in on the liver. Alnylam has developed a similar approach to target tissues beyond the liver, such as the eye and the central nervous system. Getting RNA to accumulate in certain tissues, including the heart, will be a challenge, says Slack, but Alnylam’s success “has just opened the flood gates.”


H. Ledford, Gene-silencing technology gets first drug approval after 20-year waitNature, Vol. 560, p. 291, 10 August 2018

Molecular windows into primeval worlds

Analysis of a fossil of Dickensonia Tenuis, over 500 million years old suggests it is related to the Metazoa or animals group.

A fossil of Dickinsonia contained traces of cholesterol-like molecules, a signature of animal life. (D. Grazhdankin)

This year, scientists detected molecular traces from creatures that lived more than half a billion years ago, sharpening their picture of the mysterious world that gave rise to some of Earth’s first animals and pushing such molecular paleontology back several hundred million years. They detected the signatures of fat molecules in some of the strangest fossils known, the enigmatic life forms called the Ediacarans, and molecular evidence of sponges from long before they appear in the fossil record.

For more than 70 years, scientists have puzzled over the bewildering shapes of Ediacaran fossils. Some resemble leaves or fronds; others look like no other organisms that have ever lived on Earth. Were the ancient ocean dwellers plants? Animals? Or some entirely separate form of life that failed to survive?

Researchers at Australian National University in Canberra wondered whether they could extract chemical clues from some exceptional fossils that, despite being 550 million years old, still preserve a film of what looks like organic material. These fossils come from a cliff on the shore of the White Sea in northwestern Russia, where the rocks have escaped the heat and pressure that can obliterate such molecular traces.

Researchers first tested the idea on a collection of small, round Ediacaran fossils called Beltanelliformis. They removed the film from the rock, dissolved it, and used gas chromatography and mass spectrometry to look for preserved organic molecules. They found high levels of hopanes, molecules that suggested the balls were colonies of cyanobacteria, the researchers reported in January. That success gave them the nerve to try the technique on a fossil of a creature called Dickinsonia, one of the most famous Ediacaran species. Oval-shaped and about half a meter long, it resembles a quilted bath mat. In September, the team reported that the Dickinsonia fossil contained traces of cholesterol-like molecules, a signature of animal life. That fits with other evidence suggesting at least some Ediacarans were among the earliest animals on Earth.

In October, another team found traces of molecules that today are made only by sponges, in rock layers between 660 million and 635 million years old. The finding suggests sponges, another form of animal life, might have evolved 100 million years earlier than their oldest recognizable fossils.


I. Bobrovskiy et al.Molecular fossils from organically preserved Ediacara biota reveal cyanobacterial origin for BeltanelliformisNature Ecology & Evolution, Vol. 2, p. 437, 22 January 2018

I. Bobrovskiy et al.Ancient steroids establish the Ediacaran fossil Dickinsonia as one of the earliest animalsScience, Vol. 361, p. 1246, 21 September 2018

J. A. Zumberge et al.Demosponge steroid biomarker 26-methylstigmastane provides evidence for Neoproterozoic animalsNature Ecology & Evolution, Vol. 2, p. 1709, 15 October 2018

How cells marshal their contents

Millions of RNA molecules colliding in a cell assume specific shapes to recognize each other and condense.

Liquid droplets formed from protein and RNA are emerging as a new form of cellular organization. (E.M. Langdon et al., Science 2018)

How do multiple actors inside a cell get together at the right place and time to perform critical functions? Often the answer, biologists are coming to realize, is liquid droplets. Unseen until recently, they are showing up everywhere in cells, organizing (and sometimes gumming up) the works.

Tens of thousands of proteins and other molecules populate the cytoplasm, the thick liquid that surrounds the cell nucleus, often jostling each other and reacting to perform the tasks of life, from breaking down nutrients to liberating energy to recycling waste. Beginning in 2009, researchers discovered that many proteins separate, or condense, into discrete droplets, concentrating their contents, especially when the cell is responding to stress. This “liquid-liquid phase separation,” analogous to the “demixing” of oil and vinegar in a vinaigrette salad dressing, is now one of the hottest topics in cell biology, as evidence accumulates that it promotes critical biochemical reactions and appears to be a basic organizing principle of the cell.

Two 2017 papers in Nature had shown liquid protein droplets in the cell nucleus help compact regions of the genome, silencing the genes within. This year, three papers in Science pointed to an even bigger role for phase separation. They showed proteins that drive the transfer of the genetic code from DNA to RNA—the first step in making new proteins—can condense into droplets that attach to the DNA. Details remain to be worked out, but these studies reveal a role for phase separation in one of life’s fundamental mysteries, the selective expression of genes.

Biophysicists are working out how these droplets form. Certain classes of proteins trail spaghettilike tails that interact to trigger the condensation. But when the process goes awry, what should be a liquid can become a gel, and a gel can solidify, forming the kinds of aggregates seen in neurodegenerative diseases such as amyotrophic lateral sclerosis. A March Science paper showed this happening when such proteins are improperly excluded from the cell nucleus. In April, four papers in Cell revealed possible measures for dissolving the toxic aggregates, and several labs are now trying to exploit this knowledge to discover drugs for treating neurodegenerative diseases.


A. Klosin and A. Hyman, A liquid reservoir for silent chromatinNature, Vol. 547, p. 168, 13 July 2017

A. Plys and R. Kingston, Dynamic condensates activate transcriptionScience, Vol. 361, p. 329, 27 July 2018

M. Polymenidou, The RNA face of phase separationScience, Vol. 360, p. 859, 25 May 2018

S, Mikhaleva and E. Lemke, Beyond the Transport Function of Import Receptors: What’s All the FUS about?Cell, Vol. 173, p. 549, 19 April 2018


Climate-fueled disasters rise, political action stalls

Four hurricanes churn in the Atlantic Ocean in September, the first such lineup in a decade.

Four tropical cyclones were active in the Atlantic Ocean on 12 September: Florence (left), Helene (right), Isaac (bottom), and Joyce (middle). This was the first occurrence of four named storms at the same time since 2008. (National Oceanic and Atmospheric Administration)

Devastating wildfires in the western United States and northern Europe. A record heat wave in southern Europe. Hurricanes, cyclones, and flooding in the Americas and the eastern Pacific Ocean. For many, this was the year climate change hit home. Climate-influenced disasters have grown stronger and lasted longer. As the latest iteration of the U.S. National Climate Assessment put it in November: “The evidence of humancaused climate change is overwhelming ... [and] the impacts of climate change are intensifying across the country.”

Several modern records will be broken this year, as they have been inexorably year after year. The overall temperature of the world’s oceans—the best thermometer for the planet itself—is the highest it has been since record keeping began. Ocean levels are some 8 centimeters higher than in the 1990s—and the rise is accelerating. And global greenhouse gas emissions will again hit an all-time high, likely rising by more than 2% over last year.

Yet, as the evidence—enumerated in a series of alarming scientific reports this fall—has mounted, the gap between what the world needs to do and what it is doing seems wider and starker than ever. In the United States, President Donald Trump has disputed the science of human-driven climate change, sought to roll back most of the climate-focused policies that his predecessor enacted, and stood firm in his intent to pull the United States out of the Paris agreement, the international deal to curb greenhouse gas emissions. The White House even tried to downplay the National Climate Assessment, a report mandated by Congress and endorsed by government science agencies. “I do not believe it,” Trump said when asked about estimated economic impacts; his spokesperson Sarah Sanders called the report “extreme” and “not based on facts.” “The federal government is constructing an alternative reality,” says Phil Duffy, president of the Woods Hole Research Center in Falmouth, Massachusetts. “They’re in la-la land.”

The United States is not alone. “Each year that goes by with lack of action and leadership in the U.S., more and more countries around the world have an excuse for stepping back,” says Kelly Sims Gallagher, director of Tufts University’s Center for International Environment & Resource Policy in Medford, Massachusetts. For example, Brazil’s incoming president, Jair Bolsonaro, has promised to open Amazonian rainforest for development, potentially releasing a rush of carbon dioxide (CO2) emissions. China is once again focusing more on problems such as clean air rather than carbon emissions, and even the European Union is distracted by internal upheavals.

The costs of decades of little or no action are becoming manifest as the “natural” is slowly drained from natural disasters. The consequences are worst where human influence on the climate slams into the human predilection to live in risk-prone places. Take the record wildfires in California, such as the Camp Fire, which killed at least 86 people and reduced the town of Paradise to ash. Warming temperatures and a downturn in summer rainfall are drying out the western United States, prolonging torrid droughts that turn forests and brush to tinder. Large wildfires there now burn twice the area they did in 1970; by midcentury, the area burned by all wildfires in the region is projected to increase as much as sixfold. “These bigger fires, fires that move faster, and a longer fire season—it’s clearly, clearly here,” Duffy says.

On the East Coast, low-lying cities such as Norfolk, Virginia, are experiencing flooding at high tide thanks to a combination of sea-level rise and subsidence linked to the long-ago retreat of the ice sheets. And when there’s not sunny-day flooding, there are storms: This year brought no reprieve from 2017’s blockbuster hurricanes. Like Harvey, which devastated Houston, Texas, last year, this year’s Hurricane Florence exhibited many telltale signs of global warming’s influence: It intensified rapidly and dawdled over land, drowning the North Carolina coast with unprecedented rainfall.

Those effects don’t stop at U.S. shores. Super Typhoon Mangkhut, the year’s strongest storm, battered the Philippines, triggering landslides and killing at least 66 people. In the United Kingdom, human-driven warming has made debilitating summer heat waves 30 times more likely; by midcentury, such heat will grip the island once every 2 years. A similar heat wave in Canada this year killed more than 90 people. The recent spike in sea level, now at 3.9 millimeters a year, has put Pacific Island nations on edge, and studies this year suggested wave-driven overwash could make many of those islands uninhabitable within decades.

In October, the Intergovernmental Panel on Climate Change, a United Nations–sponsored group that includes hundreds of the world’s leading climate scientists, released a grim look at the effects of a global temperature increase just 1.5°C above preindustrial levels—not much more than the 1°C the planet has already warmed. Among the findings: After another half-degree of warming, many of the world’s coral reefs would be doomed. In some regions, drowning rains and scorching heat waves would grow more severe. Arctic sea ice would rapidly retreat. And holding the temperature increase to that level would require a stark drop in carbon emissions, along with steps to actively remove CO2 from the atmosphere, the report said.

“We reach 1.5° by 2040. We’ve only got 2 or 3 decades,” says Myles Allen, one of the report’s lead authors and a climate dynamicist at the University of Oxford in the United Kingdom. Perhaps it’s still theoretically possible, Duffy adds. “But to meet the goal, [the world] needs to change now. And I don’t see that happening.”

Even when global warming recaptures the world’s attention, the problem will not be easy to solve. The world needs to weigh the costs and benefits of keeping the warming to 1.5°C rather than 2°C or higher, Allen says. “Politically the conversation has to move from now to how much a burden we impose on the next generation,” he says. But Don Wuebbles, an atmospheric scientist at the University of Illinois in Urbana and a lead author of the National Climate Assessment, says the burden is already heavy. “I’ve been through Paradise,” he says, “which no longer exists.”


The National Climate Assessment, Fourth National Climate Assessment Volume II: Impacts, Risks, and Adaptation in the United States, 2018

Intergovernmental Panel on Climate Change, Global Warming of 1.5 ºC, 2018

An ethically fraught gene-editing claim

Artist depiction of egg cell being fertilized.

In vitro fertilization, the first step in gene editing. (MedicalRF/Science Source)

Humanity rewriting its own genetic code is no small feat. At another time, under different circumstances, germline gene editing might well have a shot at becoming Science’s Breakthrough of the Year. But a Chinese researcher’s claim in November that he had created twin baby girls resistant to HIV using the gene-editing technique CRISPR doesn’t qualify for that distinction.

Among scientists and ethicists, a consensus has emerged about the conditions under which such work might be acceptable in the future: if it is the only way to help parents conceive a healthy baby, if scientists have done everything possible to show the technique is safe, if the study has undergone careful ethical vetting, and if it is carried out transparently.

He Jiankui of the Southern University of Science and Technology in Shenzhen, China, appears to have met none of those criteria. As Science went to press, He had not published his findings, and there was no proof that two babies were born with altered genes for a protein exploited by HIV. It’s unclear whether the edits would truly shield Lulu and Nana from HIV infection, or why the potential benefits were worth the risks given that other, proven methods exist to prevent HIV infection and the girls were not facing an unusually high risk of exposure to the virus. The study’s ethical review seems murky at best, the work was shrouded in secrecy—a planned public relations campaign fell apart after a news leak—and He broke an international consensus on germline experiments as well as, it seems, Chinese regulations. Because of those many shortcomings, his claim counts as one of the science breakdowns of 2018.

J. Cohen, What now for human genome editing?, Science, Vol. 362, p. 1090, 7 December 2018

Brazilian science gutted

Gutted interior of Rio de Janeiro's National Museum after fire destroyed most of the contents.

Brazil’s National Museum, following a devastating fire. (Mauro Pimentel/AFP/Getty Images)

The fiery death of Brazil’s 200-year-old National Museum in Rio de Janeiro was painfully symbolic of what many researchers fear is a looming demise of Brazilian science. The museum burned down on the night of 2 September, following years of underfunding and neglect by authorities. Public funds for science in most of the country have followed a similar trajectory. The budget of the federal science ministry shrank by more than 50% in the past 5 years, and an additional 10% cut is expected for 2019, despite scientists’ many appeals to legislators in Brasília.

Fears deepened after the October election of far-right congressman Jair Bolsonaro as the next president of Brazil. Although he promises to triple the rate of investment in science, technology, and innovation to 3% of gross domestic product in the next 4 years—something many analysts say is not feasible—the former army captain is at odds with scientists on several issues. He has threatened to pull Brazil out of the 2015 Paris agreement on climate change and vowed to reduce funding for federal universities, where much of Brazil’s science is conducted, claiming that Brazilian academia is dominated by “left ideology” and that many universities in Brazil represent “wasted money.”


H. Escobar, Scientists, environmentalists brace for Brazil’s right turnScience, Vol. 362, p. 273, 19 October 2018