While the vast majority of well-structured single protein chains can now be predicted to high accuracy due to the recent AlphaFold [1] model, the prediction of multi-chain protein complexes remains a challenge in many cases. In this work, we demonstrate that an AlphaFold model trained specifically for multimeric inputs of known stoichiometry, which we call AlphaFold-Multimer, significantly increases accuracy of predicted multimeric interfaces over input-adapted single-chain AlphaFold while maintaining high intra-chain accuracy. On a benchmark dataset of 17 heterodimer proteins without templates (introduced in [2]) we achieve at least medium accuracy (DockQ [3]≥0.49) on 14 targets and high accuracy (DockQ≥0.8) on 6 targets, compared to 9 targets of at least medium accuracy and 4 of high accuracy for the previous state of the art system (an AlphaFold-based system from [2]). We also predict structures for a large dataset of 4,433 recent protein complexes, from which we score all non-redundant interfaces with low template identity. For heteromeric interfaces we successfully predict the interface (DockQ≥0.23) in 67% of cases, and produce high accuracy predictions (DockQ≥0.8) in 23% of cases, an improvement of +25 and +11 percentage points over the flexible linker modification of AlphaFold [4] respectively. For homomeric interfaces we successfully predict the interface in 69% of cases, and produce high accuracy predictions in 34% of cases, an improvement of +5 percentage points in both instances.
In “Discovery of complex oxides via automated experiments and data science”, a collaborative effort with partners at the Joint Center for Artificial Photosynthesis (JCAP), a Department of Energy (DOE) Energy Innovation Hub at Caltech, we present a systematic search for new complex crystalline metal oxides using a novel approach for rapid materials synthesis and characterization. Using a customized inkjet printer to print samples with different ratios of metals, we were able to generate more than 350k distinct compositions, a number of which we discovered had interesting properties. One example, based on cobalt, tantalum and tin, exhibited tunable transparency, catalytic activity, and stability in strong acid electrolytes, a rare combination of properties of importance for renewable energy technologies. To stimulate continued research in this field, we are releasing a database consisting of nine channels of optical absorption measurements, which can be used as an indicator of interesting properties, across 376,752 distinct compositions of 108 3-metal oxide systems, along with model results that identify the most promising compositions for a variety of technical applications.
DeepMind, the U.K.-based AI lab that seeks to develop artificial general intelligence, has finally become profitable, according to the company’s latest financial report.
Since being acquired by Google (now Alphabet Inc.) in 2014, DeepMind has struggled to break even with its growing expenses. And now, it is finally giving its parent company and shareholders hopeful signs that it has earned its place among Alphabet’s constellation of profitable businesses.
The Brookings Institution, Richard V. Reeves and Ember Smith
from
To Hanna Rosin, author of
The End of Men, it is “the strangest and most profound change of the century, even more so because it is unfolding in a similar way pretty much all over the world.”
Importantly, there is a gender gap not only in rates of college enrollment, as we described earlier in the year, and recently highlighted in the
Wall Street Journal and
The Atlantic, but also in rates of completion among those who do enroll.
The National Science Foundation has awarded a nearly $1-million grant to the University of Hawaiʻi Hawaiʻi Data Science Institute (HI-DSI) supporting cyberinfrastructure (CI) training for climate scientists and students in the Hawaiʻi-Pacific region. The three-year grant will help create workshop and curriculum models for undergraduate and graduate students to increase their cyberinfrastructure skills across the climate science domain.
The project, titled “Cyberinfrastructure Training to Advance Climate Science” (CI-TRACS), will help build essential CI skills for the Hawaiʻi-Pacific workforce through 36 undergraduate and 12 graduate student training experiences and 10 new publicly available workshops to increase CI awareness and skills.
Suicide is a leading cause of death in the United States that disproportionately affects minority and disenfranchised communities, including Black, Hispanic, and LGBTQ populations. Yet, these groups often are not included in suicide prevention research. In an effort to reduce disparities and increase the understanding of suicide, the National Institutes of Health awarded Penn Medicine researchers a grant of more than $14 million over the next five years to develop the Penn Innovation in Suicide Prevention Implementation Research (INSPIRE) Center.
Led by Maria A. Oquendo, MD, PhD, chair of Psychiatry in the Perelman School of Medicine at the University of Pennsylvania, and Gregory K. Brown, PhD, a research associate professor of Clinical Psychology in Psychiatry at Penn and director of the Penn Center for the Prevention of Suicide, the center brings together psychiatry, implementation science, health economics, machine learning, and other interdisciplinary research experts to apply innovative approaches to suicide prevention.
The University of Nebraska-Lincoln is seeking $75 million in federal relief funds to advance two proposals: a $50 million expansion of its Holland Computing Center and $25 million for a related facility to advance public-private agricultural partnerships.
The University of Nebraska-Lincoln was among dozens of other organizations requesting allocations from the Nebraska Legislature this past week, all seeking to advance major priorities through use of COVID-19 relief funds allocated through the American Rescue Plan Act of 2021, which came into effect in March of this year.
The Act is anticipated to allocate around a billion dollars to Nebraska in support of repairing the economic and societal damage caused by the pandemic. The University of Nebraska-Lincoln, for its part, is pitching its proposals as cornerstones in the advancement of Nebraskan industry – particularly its agricultural sector.
Carnegie Mellon University (CMU) and Emerald Cloud Lab (ECL) recently announced their plans to build a cloud lab at the university’s campus in Pittsburgh. A carbon copy of ECL’s lab in San Francisco, the CMU Cloud Lab will enable scientists to perform experiments remotely and give them access to nearly 200 types of scientific instruments.
To learn more about the CMU Cloud Lab, the motivation behind the project and the benefits it will bring, Technology Networks spoke to Rebecca Doerge, PhD, dean, Mellon College of Science, Carnegie Mellon University, and Toby Blackburn, head of business development and research, Emerald Cloud Lab.
In 2019, Geovation — the UK Ordnance Survey’s start-up accelerator — formed a partnership with the Omidyar network to raise awareness of the potential risks and harms that might arise from using location data, and to identify solutions that realise benefits while minimising negative impacts.
This program of work became known as the Benchmark Initiative, which has delivered a series of public thought leadership events and an entrepreneur program.
In parallel with Benchmark, the American Geographical Society (also in partnership with Omidyar Network) launched EthicalGEO in the USA, which has supported research fellows investigating issues at the intersection of ethics and geospatial sciences.
The Benchmark dialogues have investigated: bias, transparency, privacy, imbalances of power, anonymity, contact tracing, managing social distancing in public areas and workspaces, smart cities, designing for accessibility, data colonialism, tracking waste, and data relating to human migration and humanitarian crises.
The objective of Carlos’ project is to leverage magnetic-resonance imaging using machine learning to perform early detection of AD. Development of this technique could lead to automatic early diagnosis, which would improve clinicians’ ability to select appropriate care, enable quicker access to treatment for patients, and clarify treatment timeline and course of action for caregivers. It would also inform health agency decisions on how to best allocate resources and facilitate effective selection of participants in clinical trials.
Tom Erickson, the director of the new UW-Madison School of Computer, Data and Information Sciences, brought a message of statewide engagement with industry to a meeting of the Wisconsin Technology Council board of directors. From tourism to agriculture, and from manufacturing to transportation, computing and data is changing how those sectors perform while inventing new industries along the way.
“Five out of the largest 10 companies in the world are software companies,” said Erickson, who founded several private software companies himself before taking his first-ever job in academia to lead the new UW-Madison school.
Created through the merger of three departments, the School of Computer, Data and Information Sciences has an ambitious mission that goes beyond educating the next-generation of “data heads” — although that remains center stage. Computer science is already the largest single major on campus and the new data science major is the fastest growing. Combined with other sequences, CDIS includes about 3,700 students.
During the pandemic, California schools and nonprofits found themselves in desperate need of technical equipment to stay connected to their communities, clients and students.
Meanwhile, hundreds of thousands, if not millions, of laptops, monitors and keyboards began collecting dust in tech offices across the state after Gov. Gavin Newsom ordered office workers in non-essential roles to stay home.
Scope 1, 2 and 3 emissions are a way of categorising the different kinds of emissions that a company creates. They were first introduced by the greenhouse gas protocol (GHG protocol) in 2001.
The GHG protocol was created by the World Resource Institute and the World Business Council for Sustainable Development. The aim of the GHG protocol is to provide guidance for governments, businesses, NGOs and others in carbon accounting.
The number of reforestation projects worldwide is increasing. In many cases funding is obtained through the claimed carbon capture of the trees, presented as immediate and durable, whereas reforested plots need time and maintenance to realise their carbon capture potential. Further, claims usually overlook the environmental costs of natural or anthropogenic disturbances during the forest’s lifetime, and greenhouse gas (GHG) emissions associated with the reforestation are not allowed for. This study uses life cycle assessment to quantify the carbon footprint of setting up a reforestation plot in the Peruvian Amazon. In parallel, we combine a soil carbon model with an above- and below-ground plant carbon model to predict the increase in carbon stocks after planting. We compare our results with the carbon capture claims made by a reforestation platform. Our results show major errors in carbon accounting in reforestation projects if they (1) ignore the time needed for trees to reach their carbon capture potential; (2) ignore the GHG emissions involved in setting up a plot; (3) report the carbon capture potential per tree planted, thereby ignoring limitations at the forest ecosystem level; or (4) under-estimate tree losses due to inevitable human and climatic disturbances. Further, we show that applications of biochar during reforestation can partially compensate for project emissions.
Cross-Laminated Timber (CLT) is a material beloved of Treehugger because it is made from wood, a renewable resource with low upfront carbon emissions, compared to materials like concrete or steel, which emit a lot of carbon dioxide during their manufacture. 1 De Haro is the first CLT building in San Francisco, but the design by Perkins&Will demonstrates the advantages of mass timber construction go beyond just the storage or avoidance of carbon.
Online December 8-9. ” At this wide-ranging conference, Stanford University is convening experts and leaders from academia, industry, government, and clinical practice to explore critical and emerging issues related to impact of AI across the spectrum of health, healthcare, and related areas. Content will be relevant to practitioners, researchers, executives, policymakers, and professionals, with and without technical expertise.” [$$$]
“The roads of Discovery Park District become highly trafficked areas during certain events such as football games, concerts and festivals. Moreover, West Lafayette’s population nearly doubles every fall through spring when students return to campus. The West Lafayette Smart City Challenge asks participants to develop a scalable Internet of Things solution and/or software application to improve safety for vulnerable road users who navigate high-traffic corridors and intersections during major seasonal events.” Deadline for initial online applications is December 3.
SPONSORED CONTENT
The eScience Institute’s Data Science for Social Good program is now accepting applications for student fellows and project leads for the 2021 summer session. Fellows will work with academic researchers, data scientists and public stakeholder groups on data-intensive research projects that will leverage data science approaches to address societal challenges in areas such as public policy, environmental impacts and more. Student applications due 2/15 – learn more and apply here. DSSG is also soliciting project proposals from academic researchers, public agencies, nonprofit entities and industry who are looking for an opportunity to work closely with data science professionals and students on focused, collaborative projects to make better use of their data. Proposal submissions are due 2/22.
