Multivariate time series (MTS) forecasting is an important problem in many fields. Accurate forecasting results can effectively help decision-making. Variables in MTS have rich relations among each other and the value of each variable in MTS depends both on its historical values and on other variables. These rich relations can be static and predictable or dynamic and latent. Existing methods do not incorporate these rich relational information into modeling or only model certain relation among MTS variables. To jointly model rich relations among variables and temporal dependencies within the time series, a novel end-to-end deep learning model, termed Multivariate Time Series Forecasting via Heterogenous Graph Neural Networks (MTHetGNN) is proposed in this paper. To characterize rich relations among variables, a relation embedding module is introduced in our model, where each variable is regarded as a graph node and each type of edge represents a specific relationship among variables or one specific dynamic update strategy to model the latent dependency among variables. In addition, convolutional neural network (CNN) filters with different perception scales are used for time series feature extraction, which is used to generate the feature of each node. Finally, heterogenous graph neural networks are adopted to handle the complex structural information generated by temporal embedding module and relation embedding module. Three benchmark datasets from the real world are used to evaluate the proposed MTHetGNN and the comprehensive experiments show that MTHetGNN achieves state-of-the-art results in MTS forecasting task.

The ability of intelligent agents to learn and remember multiple tasks sequentially is crucial to achieving artificial general intelligence. Many continual learning (CL) methods have been proposed to overcome catastrophic forgetting which results from non i.i.d data in the sequential learning of neural networks. In this paper we focus on class incremental learning, a challenging CL scenario. For this scenario, generative replay is a promising strategy which generates and replays pseudo data for previous tasks to alleviate catastrophic forgetting. However, it is hard to train a generative model continually for relatively complex data. Based on recently proposed orthogonal weight modification (OWM) algorithm which can approximately keep previously learned feature invariant when learning new tasks, we propose to 1) replay penultimate layer feature with a generative model; 2) leverage a self-supervised auxiliary task to further enhance the stability of feature. Empirical results on several datasets show our method always achieves substantial improvement over powerful OWM while conventional generative replay always results in a negative effect. Meanwhile our method beats several strong baselines including one based on real data storage. In addition, we conduct experiments to study why our method is effective.

Every human genome contains tens of thousands of rare genetic variants—which include single nucleotide changes, insertions or deletions, and larger structural variants—and some may have a functional effect. Ferraro et al. examined data from individuals in the Genotype-Tissue Expression (GTEx) project for outliers across tissues caused by gene expression, splicing, and allele-specific expression. Single rare variants were observed that affected the expression and allele-specific expression of multiple genes and, in the case of a gene fusion event, splicing. Experimental and computational validation suggest that many individuals carry more than 50 rare variants that affect transcription in some way. Although most variants were predicted to not affect an individual's phenotype, a small percentage showed likely disease-related associations, emphasizing the importance of studying the impact of rare genetic variation on the transcriptome. Science , this issue p. [eaaz5900][1] ### INTRODUCTION The human genome contains tens of thousands of rare (minor allele frequency <1%) variants, some of which contribute to disease risk. Using 838 samples with whole-genome and multitissue transcriptome sequencing data in the Genotype-Tissue Expression (GTEx) project version 8, we assessed how rare genetic variants contribute to extreme patterns in gene expression (eOutliers), allelic expression (aseOutliers), and alternative splicing (sOutliers). We integrated these three signals across 49 tissues with genomic annotations to prioritize high-impact rare variants (RVs) that associate with human traits. ### RATIONALE Outlier gene expression aids in identifying functional RVs. Transcriptome sequencing provides diverse measurements beyond gene expression, including allele-specific expression and alternative splicing, which can provide additional insight into RV functional effects. ### RESULTS After identifying multitissue eOutliers, aseOutliers, and sOutliers, we found that outlier individuals of each type were significantly more likely to carry an RV near the corresponding gene. Among eOutliers, we observed strong enrichment of rare structural variants. sOutliers were particularly enriched for RVs that disrupted or created a splicing consensus sequence. aseOutliers provided the strongest enrichment signal when evaluated from just a single tissue. We developed Watershed, a probabilistic model for personal genome interpretation that improves over standard genomic annotation–based methods for scoring RVs by integrating these three transcriptomic signals from the same individual and replicates in an independent cohort. To assess whether outlier RVs identified in GTEx associate with traits, we evaluated these variants for association with diverse traits in the UK Biobank, the Million Veterans Program, and the Jackson Heart Study. We found that transcriptome-assisted prioritization identified RVs with larger trait effect sizes and were better predictors of effect size than genomic annotation alone. ### CONCLUSION With >800 genomes matched with transcriptomes across 49 tissues, we were able to study RVs that underlie extreme changes in the transcriptome. To capture the diversity of these extreme changes, we developed and integrated approaches to identify expression, allele-specific expression, and alternative splicing outliers, and characterized the RV landscape underlying each outlier signal. We demonstrate that personal genome interpretation and RV discovery is enhanced by using these signals. This approach provides a new means to integrate a richer set of functional RVs into models of genetic burden, improve disease gene identification, and enable the delivery of precision genomics. ![Figure][2] Transcriptomic signatures identify functional rare genetic variation. We identified genes in individuals that show outlier expression, allele-specific expression, or alternative splicing and assessed enrichment of nearby rare variation. We integrated these three outlier signals with genomic annotation data to prioritize functional RVs and to intersect those variants with disease loci to identify potential RV trait associations. Rare genetic variants are abundant across the human genome, and identifying their function and phenotypic impact is a major challenge. Measuring aberrant gene expression has aided in identifying functional, large-effect rare variants (RVs). Here, we expanded detection of genetically driven transcriptome abnormalities by analyzing gene expression, allele-specific expression, and alternative splicing from multitissue RNA-sequencing data, and demonstrate that each signal informs unique classes of RVs. We developed Watershed, a probabilistic model that integrates multiple genomic and transcriptomic signals to predict variant function, validated these predictions in additional cohorts and through experimental assays, and used them to assess RVs in the UK Biobank, the Million Veterans Program, and the Jackson Heart Study. Our results link thousands of RVs to diverse molecular effects and provide evidence to associate RVs affecting the transcriptome with human traits. [1]: /lookup/doi/10.1126/science.aaz5900 [2]: pending:yes

Understanding how human genetic variation affects phenotype requires tissue- or even cell type–specific measurements. Kim-Hellmuth et al. used computational methods to identify cell-type proportions within bulk tissues in the Genotype-Tissue Expression (GTEx) project dataset to identify cell-type interaction quantitative trait loci and map these to genetic variants correlated with expression or splicing differences between individuals. By characterizing the cellular context, this study illustrates how genetic variants that operate in a cell type–specific manner affect gene regulation and can be linked to complex traits. This deconvolution and analysis of cell types from bulk tissues allows greater precision in understanding how phenotypes are linked to genetic variation. Science , this issue p. [eaaz8528][1] ### INTRODUCTION Efforts to map quantitative trait loci (QTLs) across human tissues by the GTEx Consortium and others have identified expression and splicing QTLs (eQTLs and sQTLs, respectively) for a majority of genes. However, these studies were largely performed with gene expression measurements from bulk tissue samples, thus obscuring the cellular specificity of genetic regulatory effects and in turn limiting their functional interpretation. Identifying the cell type (or types) in which a QTL is active will be key to uncovering the molecular mechanisms that underlie complex trait variation. Recent studies demonstrated the feasibility of identifying cell type–specific QTLs from bulk tissue RNA-sequencing data by using computational estimates of cell type proportions. To date, such approaches have only been applied to a limited number of cell types and tissues. By applying this methodology to GTEx tissues for a diverse set of cell types, we aim to characterize the cellular specificity of genetic effects across human tissues and to describe the contribution of these effects to complex traits. ### RATIONALE A growing number of in silico cell type deconvolution methods and associated reference panels with cell type–specific marker genes enable the robust estimation of the enrichment of specific cell types from bulk tissue gene expression data. We benchmarked and used enrichment estimates for seven cell types (adipocytes, epithelial cells, hepatocytes, keratinocytes, myocytes, neurons, and neutrophils) across 35 tissues from the GTEx project to map QTLs that are specific to at least one cell type. We mapped such cell type–interaction QTLs for expression and splicing (ieQTLs and isQTLs, respectively) by testing for interactions between genotype and cell type enrichment. ### RESULTS Using 43 pairs of tissues and cell types, we found 3347 protein-coding and long intergenic noncoding RNA (lincRNA) genes with an ieQTL and 987 genes with an isQTL (at 5% false discovery rate in each pair). To validate these findings, we tested the QTLs for replication in available external datasets and applied an independent validation using allele-specific expression from eQTL heterozygotes. We analyzed the cell type–interaction QTLs for patterns of tissue sharing and found that ieQTLs are enriched for genes with tissue-specific eQTLs and are generally not shared across unrelated tissues, suggesting that tissue-specific eQTLs originate in tissue-specific cell types. Last, we tested the ieQTLs and isQTLs for colocalization with genetic associations for 87 complex traits. We show that cell type–interaction QTLs are enriched for complex trait associations and identify colocalizations for hundreds of loci that were undetected in bulk tissue, corresponding to an increase of >50% over colocalizations with standard QTLs. Our results also reveal the cellular specificity and potential origin for a similar number of colocalized standard QTLs. ### CONCLUSION The ieQTLs and isQTLs identified for seven cell types across GTEx tissues suggest that the large majority of cell type–specific QTLs remains to be discovered. Our colocalization results indicate that comprehensive mapping of cell type–specific QTLs will be highly valuable for gaining a mechanistic understanding of complex trait associations. We anticipate that the approaches presented here will complement studies mapping QTLs in single cells. ![Figure][2] Detection of cell type–specific effects on gene expression. The enrichment of seven cell types is calculated across GTEx tissues, enabling mapping of cell type–interaction QTLs for expression and splicing by testing for significant interactions between genotypes and cell type enrichments. Linking these QTLs to complex trait associations enables discovery of >50% more colocalizations compared with standard QTLs and reveals the cellular specificity of traits. The Genotype-Tissue Expression (GTEx) project has identified expression and splicing quantitative trait loci in cis (QTLs) for the majority of genes across a wide range of human tissues. However, the functional characterization of these QTLs has been limited by the heterogeneous cellular composition of GTEx tissue samples. We mapped interactions between computational estimates of cell type abundance and genotype to identify cell type–interaction QTLs for seven cell types and show that cell type–interaction expression QTLs (eQTLs) provide finer resolution to tissue specificity than bulk tissue cis-eQTLs. Analyses of genetic associations with 87 complex traits show a contribution from cell type–interaction QTLs and enables the discovery of hundreds of previously unidentified colocalized loci that are masked in bulk tissue. [1]: /lookup/doi/10.1126/science.aaz8528 [2]: pending:yes

Telomeres are DNA-protein complexes that protect chromosome ends. Their length is of great interest because short telomeres are associated with specific diseases and with aging. Demanelis et al. measured telomere length from 952 Genotype-Tissue Expression (GTEx) project donors across tissues, of which 24 tissue types have measurements for more than 25 samples. This dataset shows that telomere length is not constant but is correlated across tissues. Most tissue telomeres shorten with age, but some, such as those in the testis and cerebellum, do not. In African Americans, telomeres are longer on average than those from individuals of primarily European descent across many tissue types. This observation is consistent with variability being passed from germ cells to zygote to differentiated cells during development. Science , this issue p. [eaaz6876][1] ### INTRODUCTION Telomeres are DNA-protein complexes located at the end of chromosomes that protect chromosome ends from degradation and fusion. The DNA component of telomeres shortens with each cell division, eventually triggering cellular senescence. Telomere length (TL) in blood cells has been studied extensively as a biomarker of human aging and risk factor for age-related diseases. The extent to which TL in whole blood reflects TL in disease-relevant tissue types is unknown, and the variability in TL across human tissues has not been well characterized. The postmortem tissue samples collected by the Genotype-Tissue Expression (GTEx) project provide an opportunity to study TL in many human tissue types, and accompanying data on inherited genetic variation, gene expression, and donor characteristics enable us to examine demographic, genetic, and biologic determinants and correlates of TL within and across tissue types. ### RATIONALE To better understand variation in and determinants of TL, we measured relative TL (RTL, telomere repeat abundance in a DNA sample relative to a standard sample) in more than 25 tissue types from 952 GTEx donors (deceased, aged 20 to 70 years old). RTL was measured for 6391 unique tissue samples using a Luminex assay, generating the largest publicly available multitissue TL dataset. We integrated our RTL measurements with data on GTEx donor characteristics, inherited genetic variation, and tissue-specific expression and analyzed relationships between RTL and covariates using linear mixed models (across all tissues and within tissues). Through this analysis, we sought to accomplish four goals: (i) characterize sources of variation in TL, (ii) evaluate whole-blood TL as a proxy for TL in other tissue types, (iii) examine the relationship between age and TL across tissue types, and (iv) describe biological determinants and correlates of TL. ### RESULTS Variation in RTL was attributable to tissue type, donor, and age and, to a lesser extent, race or ethnicity, smoking, and inherited variants known to affect leukocyte TL. RTLs were generally positively correlated among tissues, and whole-blood RTL was a proxy for RTL in most tissues. RTL varied across tissue types and was shortest in whole blood and longest in testis. RTL was inversely associated with age in most tissues, and this association was strongest for tissues with shorter average RTL. African ancestry was associated with longer RTL across all tissues and within specific tissue types, suggesting that ancestry-based differences in TL exist in germ cells and are transmitted to the zygote. A polygenic score consisting of inherited variants known to affect leukocyte TL was associated with RTL across all tissues, and several of these TL-associated variants affected expression of nearby genes in multiple tissue types. Carriers of rare, loss-of-function variants in TL-maintenance genes had shorter RTL (based on analysis of multiple tissue types), suggesting that these variants may contribute to shorter TL in individuals from the general population. Components of telomerase, a TL maintenance enzyme, were more highly expressed in testis than in any other tissue. We found evidence that RTL may mediate the effect of age on gene expression in human tissues. ### CONCLUSION We have characterized the variability in TL across many human tissue types and the contributions of aging, ancestry, genetic variation, and other biologic processes to this variability. The correlation observed among TL measures from different tissues highlights the existence of host factors with effects on TL that are shared across tissue types (e.g., TL in the zygote). These results have important implications for the interpretation of epidemiologic studies of leukocyte TL and disease. ![Figure][2] TL in human tissues. Using a Luminex-based assay, TL was measured in DNA samples from >25 different human tissue types from 952 deceased donors in the GTEx project. TL within tissue types is determined by numerous factors, including zygotic TL, age, and exposures. TL differs across tissues and correlates among tissue types. TL in most tissues declines with age. Telomere shortening is a hallmark of aging. Telomere length (TL) in blood cells has been studied extensively as a biomarker of human aging and disease; however, little is known regarding variability in TL in nonblood, disease-relevant tissue types. Here, we characterize variability in TLs from 6391 tissue samples, representing >20 tissue types and 952 individuals from the Genotype-Tissue Expression (GTEx) project. We describe differences across tissue types, positive correlation among tissue types, and associations with age and ancestry. We show that genetic variation affects TL in multiple tissue types and that TL may mediate the effect of age on gene expression. Our results provide the foundational knowledge regarding TL in healthy tissues that is needed to interpret epidemiological studies of TL and human health. [1]: /lookup/doi/10.1126/science.aaz6876 [2]: pending:yes

In humans, the inheritance of the XX or XY set of sex chromosomes is responsible for most individuals developing into adults expressing male or female sex-specific traits. However, the degree to which sex-biased gene expression occurs in tissues, especially those that do not contribute to characteristic sexually dimorphic traits. is unknown. Oliva et al. examined Genotype-Tissue Expression (GTEx) project data and found that 37% of genes in at least one of the 44 tissues studied exhibit a tissue-specific, sex-biased gene expression. They also identified a sex-specific variation in cellular composition across tissues. Overall, the effects of sex on gene expression were small, but they were genome-wide and mostly mediated through transcription factor binding. With sex-biased gene expression associated with loci identified in genome-wide association studies, this study lays the groundwork for identifying the molecular basis of male- and female-based diseases. Science , this issue p. [eaba3066][1] ### INTRODUCTION Many complex human phenotypes, including diseases, exhibit sex-differentiated characteristics. These sex differences have been variously attributed to hormones, sex chromosomes, genotype × sex effects, differences in behavior, and differences in environmental exposures; however, their mechanisms and underlying biology remain largely unknown. The Genotype-Tissue Expression (GTEx) project provides an opportunity to investigate the prevalence and genetic mechanisms of sex differences in the human transcriptome by surveying many tissues that have not previously been characterized in this manner. ### RATIONALE To characterize sex differences in the human transcriptome and its regulation, and to discover how sex and genetics interact to influence complex traits and disease, we generated a catalog of sex differences in gene expression and its genetic regulation across 44 human tissue sources surveyed by the GTEx project (v8 data release), analyzing 16,245 RNA-sequencing samples and genotypes of 838 adult individuals. We report sex differences in gene expression levels, tissue cell type composition, and cis expression quantitative trait loci (cis-eQTLs). To assess their impact, we integrated these results with gene function, transcription factor binding annotation, and genome-wide association study (GWAS) summary statistics of 87 GWASs. ### RESULTS Sex effects on gene expression are ubiquitous (13,294 sex-biased genes across all tissues). However, these effects are small and largely tissue-specific. Genes with sex-differentiated expression are not primarily driven by tissue-specific gene expression and are involved in a diverse set of biological functions, such as drug and hormone response, embryonic development and tissue morphogenesis, fertilization, sexual reproduction and spermatogenesis, fat metabolism, cancer, and immune response. Whereas X-linked genes with higher expression in females suggest candidates for escape from X-chromosome inactivation, sex-biased expression of autosomal genes suggests hormone-related transcription factor regulation and a role for additional transcription factors, as well as sex-differentiated distribution of epigenetic marks, particularly histone H3 Lys27 trimethylation (H3K27me3). Sex differences in the genetic regulation of gene expression are much less common (369 sex-biased eQTLs across all tissues) and are highly tissue-specific. We identified 58 gene-trait associations driven by genetic regulation of gene expression in a single sex. These include loci where sex-differentiated cell type abundances mediate genotype-phenotype associations, as well as loci where sex may play a more direct role in the underlying molecular mechanism of the association. For example, we identified a female-specific eQTL in liver for the hexokinase HKDC1 that influences glucose metabolism in pregnant females, which is subsequently reflected in the birth weight of the offspring. ### CONCLUSION By integrating sex-aware analyses of GTEx data with gene function and transcription factor binding annotations, we describe tissue-specific and tissue-shared drivers and mechanisms contributing to sex differences in the human transcriptome and eQTLs. We discovered multiple sex-differentiated genetic effects on gene expression that colocalize with complex trait genetic associations, thereby facilitating the mechanistic interpretation of GWAS signals. Because the causative tissue is unknown for many phenotypes, analysis of the diverse GTEx tissue collection can serve as a powerful resource for investigations into the basis of sex-biased traits. This work provides an extensive characterization of sex differences in the human transcriptome and its genetic regulation. ![Figure][2] Sex affects gene expression and its genetic regulation across tissues. Sex effects on gene expression were measured in 44 GTEx human tissue sources and integrated with genotypes of 838 subjects. Sex-biased expression is present in numerous biological pathways and is associated to sex-differentiated transcriptional regulation. Sex-biased expression quantitative trait loci in cis (sex-biased eQTLs) are partially mediated by cellular abundances and reveal gene-trait associations. TT, AT, and AA are genotypes for a single-nucleotide polymorphism; TF, transcription factor. Many complex human phenotypes exhibit sex-differentiated characteristics. However, the molecular mechanisms underlying these differences remain largely unknown. We generated a catalog of sex differences in gene expression and in the genetic regulation of gene expression across 44 human tissue sources surveyed by the Genotype-Tissue Expression project (GTEx, v8 release). We demonstrate that sex influences gene expression levels and cellular composition of tissue samples across the human body. A total of 37% of all genes exhibit sex-biased expression in at least one tissue. We identify cis expression quantitative trait loci (eQTLs) with sex-differentiated effects and characterize their cellular origin. By integrating sex-biased eQTLs with genome-wide association study data, we identify 58 gene-trait associations that are driven by genetic regulation of gene expression in a single sex. These findings provide an extensive characterization of sex differences in the human transcriptome and its genetic regulation. [1]: /lookup/doi/10.1126/science.aba3066 [2]: pending:yes

The Genotype-Tissue Expression (GTEx) project was established to characterize genetic effects on the transcriptome across human tissues and to link these regulatory mechanisms to trait and disease associations. Here, we present analyses of the version 8 data, examining 15,201 RNA-sequencing samples from 49 tissues of 838 postmortem donors. We comprehensively characterize genetic associations for gene expression and splicing in cis and trans, showing that regulatory associations are found for almost all genes, and describe the underlying molecular mechanisms and their contribution to allelic heterogeneity and pleiotropy of complex traits. Leveraging the large diversity of tissues, we provide insights into the tissue specificity of genetic effects and show that cell type composition is a key factor in understanding gene regulatory mechanisms in human tissues.

Flush with money and a hard-won respectability, alien hunters are deploying new telescopes and tactics. In 2015, Sofia Sheikh was at loose ends. Her adviser at the University of California (UC), Berkeley, with whom she studied hot, giant exoplanets, had left for a new job. Browsing reddit, she saw a post about a lavishly funded new search for extraterrestrial intelligence (SETI) and noticed that its leader was also at UC Berkeley: astrophysicist Andrew Siemion. She asked her former adviser for an introduction and met with Siemion when he was still unpacking boxes in a new office. “Everything's kind of history from there,” says Sheikh, who became the team's first undergraduate student. Sheikh is now a Ph.D. student at Pennsylvania State University (Penn State), University Park, where she led a radio survey of 20 nearby star systems aligned with Earth's orbital plane. If an intelligent civilization inhabited one of these systems and pointed a powerful telescope our way, they would see Earth passing in front of the Sun, and they might detect signs of life in our atmosphere. They might even decide to send us a message. The results, published in February in The Astrophysical Journal , were unsurprising. “Spoiler alert: no aliens,” Sheikh jokes. SETI researchers are used to negative results, but they are trying harder than ever to turn that record around. Breakthrough Listen, the $100 million, 10-year, privately funded SETI effort Siemion leads, is lifting a field that has for decades relied on sporadic philanthropic handouts. Prior to Breakthrough Listen, SETI was “creeping along” with a few dozen hours of telescope time a year, Siemion says; now it gets thousands. It's like “sitting in a Formula 1 racing car,” he says. The new funds have also been “a huge catalyst” for training scientists in SETI, says Jason Wright, director of the Penn State Extraterrestrial Intelligence Center, which opened this year. “They really are nurturing a community.” Breakthrough Listen is bolstering radio surveys, which are the mainstay of SETI. But the money is also spurring other searches, in case aliens opt for other kinds of messages—laser flashes, for example—or none at all, revealing themselves only through passive “technosignatures.” And because the data gathered by Breakthrough Listen are posted in a public archive, astronomers are combing through it for nonliving phenomena: mysterious deep-space pulses called fast radio bursts and proposed dark matter particles called axions. “There are untapped possibilities here,” says axion searcher Matthew Lawson of Stockholm University. Perhaps the most important consequence of Breakthrough Listen is that it has nudged SETI, once considered fringe science, toward the mainstream. “Journals are relaxing and letting good technosignature papers be published,” says astrobiologist Jacob Haqq-Misra of the Blue Marble Space Institute of Science. “The giggle factor is reducing.” After nearly 3 decades of eschewing SETI, NASA organized a technosignature workshop in 2018. In June, it awarded a grant to model the detectability of possible technosignatures in the atmospheres of exoplanets, its first ever SETI-related grant not involving radio searches. But some astronomers worry the funding boon is distorting science. Fernando Camilo, chief scientist of the South African Radio Astronomy Observatory, says Breakthrough Listen's voracious appetite for time on large telescopes leaves him uncomfortable. “It leaves less time to do astronomy.” Others say SETI's high-risk, rush-for-the-prize approach could distract funders from a more rational, stepwise search for extraterrestrial life. “We do have a really thoughtful process on what gets funded and what doesn't,” says Harvard University astronomer David Charbonneau. “That doesn't happen with rich individuals.” But SETI proponents don't see themselves as separatists. They are increasingly working hand in hand with those searching for exoplanets and studying astrobiology. “Looking for intelligence is the logical conclusion of this search for life,” says astronomer David Kipping of Columbia University. SETI STARTED SMALL. In 1960, astronomer Frank Drake pointed a 26-meter radio telescope in Green Bank, West Virginia, at two nearby Sun-like stars. He scanned frequencies around 1.42 gigahertz, which correspond to wavelengths of about 21 centimeters—the part of the spectrum where clouds of interstellar hydrogen emit photons. This 21-centimeter glow is ubiquitous, and Drake supposed it might be a universal channel on the cosmic dashboard, a natural place for a clarion “We are here!” But his targets, Tau Ceti and Epsilon Eridani, were expressionless. The survey, called Project Ozma, saw no sign of artifice, such as an intense spike squeezed into a narrow frequency band. With funding from NASA and the National Science Foundation (NSF), however, searches continued, with bigger telescopes to listen for fainter signals and hardware that could scan thousands and eventually millions of narrow frequency channels at once. Drake devised his now famous, eponymous equation that estimates how many communicative extraterrestrial civilizations may exist in the Milky Way. It depends on seven variables, from the rate of star formation to the average lifetime of a civilization. Even though only one of the seven factors—star-formation rate—was known with any certainty, alien hunters were on the prowl. In 1992, NASA decided to look harder, only to quickly reverse course. It embarked on the Microwave Observing Project, a 10-year, $100 million SETI search using several large telescopes. But the following year, the project was ridiculed and cut by lawmakers focused on reducing the federal budget deficit. Ever since, NASA has mostly shied away from SETI. ![Figure][1] CREDITS: (GRAPHIC) N. DESAI/ SCIENCE (DATA) JASON WRIGHT/PENN STATE Even as federal funding shriveled, the 1990s gave SETI an unexpected gift. Until then no one had detected an exoplanet, much less a potentially hospitable one, but that decade brought a host of discoveries. Since then, missions such as NASA's Kepler telescope have suggested that planetless stars are rare, and that about one in five Sun-like stars has potentially habitable Earth-size planets—two more factors in the Drake equation that have fueled optimism among SETI advocates. The turn-of-the-century tech boom offered another boost: newly minted billionaires with a taste for space. A high point came in 2007 with the inauguration of the Allen Telescope Array, a SETI observatory in California kick-started with $11.5 million from Microsoft cofounder Paul Allen. Then the field took another plunge. The 2008 financial crisis struck and within a few years, with federal and state funding tight, UC Berkeley withdrew from the project. The array was put into hibernation for 8 months. A planned expansion from 42 to 350 dishes never materialized. “SETI was entirely decimated,” Siemion says. “I was one of maybe two or three in the whole world working on SETI.” That was when Yuri Milner called. BORN AND EDUCATED in Moscow, Milner worked as a particle physicist at the Lebedev Physical Institute. In 1990, as the Soviet Union collapsed, he left to study business at the University of Pennsylvania, and in 1999 he founded an internet investment fund. The fund was an early backer of Facebook and Twitter, and later Spotify and Airbnb. Forbes magazine puts Milner's net worth at $3.8 billion. “I made some lucky investments,” he tells Science . Milner says he's always felt a connection with space and SETI. He was born in 1961, days after Drake convened the first SETI conference. He is named after Yuri Gagarin, the first cosmonaut. Once he had built up a fortune, “I discovered that now I can give back to science,” he says. He knew of SETI's dire financial straits, and he believed his money and knowledge of the tech industry could help speed up the search. Siemion's UC Berkeley center, across the San Francisco Bay from Milner's home in Silicon Valley, became the beneficiary. Breakthrough Listen set out ambitious goals ( Science , 24 July 2015, p. [357][2]). It would survey 1 million of the closest stars to Earth and 100 nearby galaxies using two of the world's most sensitive steerable telescopes, the 100-meter Green Bank Telescope in West Virginia and the 64-meter Parkes radio telescope in Australia. Buying up about 20% and 25% of the time on those telescopes, Breakthrough Listen promised to cover 10 times more sky than previous surveys and five times more of the radio spectrum, and gather data 100 times faster. Achieving these goals required new hardware. The key electronic component is a digital backend, which chops telescope data into ultrathin frequency slices and records it. Siemion says Breakthrough Listen's backends are “orders of magnitude more powerful than anything else on site.” The instruments are available for 100 hours every year to other astronomers interested in such fine frequency resolution. That allocation is often oversubscribed at Green Bank, Siemion says, ever since the backend helped characterize the first repeating fast radio burst. The project is adding a major new telescope to its mix of collaborations: MeerKAT, a South African array of 64 dishes each 13.5 meters across ( Science , 22 June 2018, p. [1285][3]). Instead of buying time on the array, Breakthrough Listen is tapping into the data stream while the telescope observes its regular targets—a procedure known as commensal observing. “You take what you can get,” Camilo says. “When it works, it's fantastic.” Commensal observing will also be added to the Karl G. Jansky Very Large Array in New Mexico, the workhorse of U.S. radio astronomy, in a project led by the privately funded SETI Institute. Gathering data sets is one thing; scouring heaps of them for alien messages is another. SETI researchers have long looked for energy packed into narrow frequency signals—something that is hard for nature to replicate, although astronomers need to exclude humanmade signals. One test is to see whether the signal's frequency drifts over time: An alien transmitter would be on a moving planet, causing a Doppler shift. If the frequency is rock steady, it's likely to be earthly interference. Similarly, if the signal persists when the telescope moves from its target, it's noise from Earth. But aliens might send something more complex than a single loud note. How do you scan SETI data for something that just seems anomalous or weird? Researchers have been trying to enlist artificial intelligence (AI), but it hasn't been easy. One species of AI, natural language algorithms, can recognize key words in the flow of human speech—think of Amazon's Alexa, or eavesdroppers at the National Security Agency—after being trained on vast speech data sets. But the huge number of narrow frequency channels in SETI data overwhelms these algorithms. Converting the data stream into 2D diagrams that resemble images works better, at least in tests, in which machine vision algorithms picked out strange pictures from a torrent of similar ones. “We have to guess what an anomaly might look like and train the algorithm to look for this, or look for things that look similar,” says Steve Croft of UC Berkeley's SETI Research Center. THE FOCUS OF SETI searches tends to reflect the technology of the times. Radio was in its heyday when Drake started out. But as lasers have grown in power and sophistication, so have efforts to spot alien laser signals with so-called optical SETI. Astronomers have carried out optical searches with modest telescopes since the 1990s. Breakthrough Listen is doing its own, with time on the 2.4-meter Automated Planet Finder (APF) telescope at the Lick Observatory in California. APF has been scanning a sample of stars to distances up to 160 light-years but will now work through a new list: stars with potentially habitable planets identified by NASA's Transiting Exoplanet Survey Satellite ( Science , 30 March 2018, p. [1453][4]). Others are developing telescopes that wouldn't need to target individual stars. The LaserSETI project, funded by the SETI Institute, is a collection of $30,000 mini-observatories, made up of an off-the-shelf fisheye lens, two cameras, and electronics that would gather light from the entire sky. The first was installed last year on an observatory roof north of San Francisco. Eventually, the institute wants to install 60 instruments around the world for 24/7 coverage. LaserSETI's small telescopes would only pick up an especially bright flash from a nearby source. Shelley Wright of UC San Diego hopes to see much farther with the Pulsed All-sky Near-infrared Optical SETI (PANOSETI), an all-sky telescope able to detect ultrashort laser pulses across all optical wavelengths. PANOSETI's design includes lightning-fast photon counters sensitive to pulses less than one-billionth of 1 second long. “It's hard for nature to make that,” Shelley Wright says. It relies on a Fresnel lens, a type used in lighthouses to focus light into a narrow beam. Flipped over, a Fresnel can gather light from a 10°-wide patch of sky onto the photon counters. The team is building two observatories, each an array of 80 telescopes with lenses 50 centimeters across, bunched together in a fly's eye arrangement. The plan is to site the pair 1 kilometer apart—to help root out false positives—at the Palomar Observatory in California. Funded by Qualcomm co-founder Franklin Antonio, the project has built five telescopes but has been stalled by the COVID-19 pandemic. THEN AGAIN, even intelligent aliens might be too busy or too shy to send messages to the stars. So SETI researchers also hope to detect passive signs of technology. People's ideas about what to look for often reflect their time: Consider the 19th century “discovery” of canals on Mars when canals were still a common form of transport on Earth. In 1960, amid rapid economic growth and concerns about energy shortages, physicist Freeman Dyson imagined an advanced society might build a megastructure surrounding a star to capture its energy ( Science , 3 June 1960, p. 1667). Such “Dyson spheres” continue to fascinate and were suggested as an explanation for the strange dimmings of the star KIC 8462852, known as Tabby's Star. In 2015, Jason Wright led a search for the glow of Dyson spheres in 100,000 nearby galaxies, using data from NASA's Wide-field Infrared Survey Explorer satellite. Technosignatures could be more subtle. In the not-too-distant future, ultrasensitive radio telescopes might be able to pick up the beams of a radar, like the ones used for air traffic control, from a distant exoplanet. Future optical telescopes might reveal the glow of a city's lights or its infrared warmth. Heavy industry or geoengineering might leave fingerprints in a planet's atmosphere. These efforts chime with searches for biosignatures, detectable marks that organic life might leave on an exoplanet ( Science , 3 November 2017, p. [578][5]). “The line between technosignatures and biosignatures is blurring,” Sheikh says. “It makes sense to observe both.” In deciding to fund the 2018 workshop on technosignatures, NASA felt that they could be discussed “on a firmer scientific foundation than before,” says Michael New, the agency's deputy associate administrator for research. After the workshop, the wording in NASA funding calls that had for some years excluded SETI-related proposals quietly disappeared. In June, Jason Wright and his colleagues benefited from the new openness when they were awarded a grant to model exoplanet atmospheres and put together a “library” of potential technosignatures, which astronomers can refer to when observing exoplanets. The team will first model chlorofluorocarbons—a pollutant that isn't produced naturally—and vast solar power arrays, because they would leave an obvious cutoff in the ultraviolet part of the spectrum. “What we should look for is things that can't be avoided, civilization's manifestations in the biosphere,” says Adam Frank, lead investigator on the grant at the University of Rochester. BUT EVEN AFTER the fanfare of Breakthrough Listen, SETI remains far from a central concern for most astronomers. In 2018, panels of researchers convened by the National Academies of Sciences, Engineering, and Medicine (NASEM) drew up strategies for NASA on astrobiology and exoplanets. They made scant mention of technosignatures and didn't advise NASA to spend any money on the topic, or, more generally, SETI. SETI enthusiasts say they are trying to avoid being shut out of an even bigger NASEM effort: its decadal survey of astrophysics, a once-a-decade priority setting exercise that is influential with funding agencies and legislators. The survey is due to report early next year. “We've made a big push to get the decadal survey … to explicitly say that NASA and the NSF need to nurture this field,” Jason Wright says. He and colleagues made nine submissions, known as white papers, to the survey, compared with a single white paper in the previous survey. Sheikh says: “There are signs the winds are starting to shift.” But many astronomers think the more important hunt is for alien life of a more basic kind, not the higher risk search for technological societies. “We have to invest in general questions,” says Charbonneau, who co-chaired the NASEM panel that developed the NASA exoplanet hunting strategy. “If we just go for the prize and don't find anything, what have we learned from that?” Mainstream astrobiologists hope the decadal survey will give a thumbs up to the Large UV/Optical/IR Surveyor, or LUVOIR, a proposed NASA space telescope as much as six times wider than the Hubble Space Telescope ( Science , 14 December 2018, p. [1230][6]). It would scrutinize habitable planets for biosignatures and estimate the fraction of them that support life—another term in the Drake equation. “The progress we've made as scientists follows the terms of the Drake equation in order,” says astrobiologist Shawn Domagal-Goldman of NASA Goddard Space Flight Center. “That progress could lead to a search for technosignatures. I could see LUVOIR being used to do that, even though it wasn't designed for such a search.” Jason Wright, however, thinks the potential payoff of SETI is just too tempting to put off the search. In July, he and his colleagues reported the “discovery space”—all the possible locations, frequencies, sensitivities, bandwidths, timings, polarizations, and modulations—that SETI radio surveys have so far explored. The result: If the entire discovery space is represented by the world's oceans, SETI has so far searched the volume of a hot tub. Milner seems ready to support at least a few more SETI hot tubs. He says he wants Breakthrough Listen to continue past 2025, when his initial funding runs out. “It's one of the most existential questions in our universe,” he says. “Just knowing we are not alone … is something that can bring us together here on Earth.” [1]: pending:yes [2]: http://www.sciencemag.org/content/349/6246/357 [3]: http://www.sciencemag.org/content/360/6395/1285 [4]: http://www.sciencemag.org/content/359/6383/1453 [5]: http://www.sciencemag.org/content/358/6363/578 [6]: http://www.sciencemag.org/content/362/6420/1230

All the below links and tweets are in English. A test run of a robot using artificial intelligence has begun in Hamamatsu City, central Japan. The robot is designed to help prevent the spread of the coronavirus by identifying people not wearing masks. Norway's Minister for Research and Higher Education Iselin Nybø is visiting Japan, including AISTs Artificial Intelligence R&D Center. NEC Corporation signed an LoI to support R&D activities in the areas of Artificial Intelligence and Robotics at the proposed India Japan Centre for Artificial intelligence and Robotics (IJCAIR) at Indian Institute of Technology Bombay.

But the potential for 5G in business leaves plenty of room for excitement, too, and organizations should also start thinking about how 5G could improve processes and production. The time to dream is now.