Bioinformatics, engineering and other disciplines driven by advances in computing were discussed by participants in an online seminar hosted by the São Paulo State Academy of Sciences to present the fifth chapter of FAPESP 60 Anos: A Ciência no Desenvolvimento Nacional (image: screenshot of the event)
Published on 11/01/2022
By Maria Fernanda Ziegler | Agência FAPESP – Two Nobel Prizes awarded this year are directly linked to advances in research on computation. The Nobel Prize in Physics went to three scientists for discoveries in the field of quantum information science that have significant applications in quantum computing. The Nobel Prize in Physiology or Medicine was given to Swedish geneticist Svante Pääbo for sequencing the Neanderthal genome and revealing genetic differences that help create a new understanding of human evolutionary history. He discovered a previously unknown hominin, Denisova, and that these extinct hominins coexisted with Homo sapiens for thousands of years, even interbreeding.
“It’s a marvelous discovery and also an example of how computing, engineering and bioinformatics can unleash new knowledge. The draft human genome was published in 2000 and the draft Neanderthal genome in 2009. As a result, it was possible to compare sequences of letters [corresponding to the nucleotides that make up DNA] each containing more than 3.2 billion nitrogen base pairs. But a volume of data as big as this can only be analyzed using bioinformatics, engineering and computing,” said Helder Nakaya, an immunologist affiliated with the Albert Einstein Jewish Hospital (HIAE), the Scientific Platform Pasteur-USP, and All for Health Institute. He also belongs to the Center for Research on Inflammatory Diseases (CRID). Hosted by the University of São Paulo (USP), CRID is one of the Research, Innovation and Dissemination Centers (RIDCs) funded by FAPESP.
Nakaya took part in an online seminar on “Computing: Science, Engineering and Art” held by the São Paulo State Academy of Sciences (ACIESP) on October 5, 2022, to discuss the fifth chapter of FAPESP 60 Anos: A Ciência no Desenvolvimento Nacional (“FAPESP 60 Years: Science in the Nation’s Development”).
“It took 14 years to sequence the human genome and required investment of USD 2.7 billion. The same thing can now be done in five hours for only USD 1,000. Besides the huge saving in cost and time, this progress has also brought us to the point we’re at in molecular engineering, where we can identify the spatial coordinates of each of the cells in a sample of human tissue, extract the RNA – which measures gene activity – and compare the expression profile of each of these cells individually. We’ve done this to investigate cancer and correlate it with the effects of sleep, for example,” Nakaya said.
The achievement recognized by the Nobel Prize in Physiology or Medicine, he added, is only one example of the impact of computing on people’s lives and the extent to which it has reconfigured social and cultural behavior while also contributing to discoveries in different areas of science.
Other examples include programs that exhibit the spread of the COVID-19 pandemic around the world, help create public safety solutions based on the Internet of Things (IoT) to detect and warn of potential disasters or crimes, and develop smart sensors for precision agriculture.
Data, algorithms and hardware
The fifth chapter of FAPESP 60 Anos: A Ciência no Desenvolvimento Nacional explores the interaction of data, algorithms and hardware in a virtuous circle that influences people and is influenced by them. The authors explain that the execution of algorithms using data generates new data, the analysis of data leads to the creation of models and new algorithms, data and algorithms are used to create executable hardware specifications, and this hardware in turn is used to run algorithms, all of which constitutes a constantly evolving cycle.
Another point discussed in the chapter is the novel challenges involving ethical questions about inappropriate use of data and algorithms with potentially negative social and economic consequences.
“Computing pervades and influences our lives. It’s in all knowledge domains. But at the same time as devices, algorithms and data influence our lives, we also influence the development and future of computing. With every discovery, this virtuous circle of interaction extends the knowledge we have of ourselves and the world around us,” said Claudia Bauzer Medeiros, a professor of computer science at the State University of Campinas’s Institute of Computer Science (IC-UNICAMP) and a member of the steering committee for the FAPESP Research Program on eScience and Data Science (eScience).
“Computing is science and engineering, but it also requires art for the development of algorithms, data and hardware,” Medeiros said to contextualize the title of the chapter.
Alongside their scientific and technological consequences, advances in computing also have human, social and economic repercussions, the authors note. The world is now permeated by data generated by the many devices we use and also by people. Many terms once limited to computing, such as IoT, big data, artificial intelligence (AI) and machine learning (a branch of AI), are becoming part of people’s vocabulary in other walks of life.
The authors stress, however, that the meaning given to these terms in everyday use is often misleading. They therefore set out to demystify the concepts concerned by clarifying their definitions and explaining how the terms should be used.
“At the same time as research in computing accelerates research in other areas, discoveries in other domains contribute to the advance of research and development in computing. It’s a virtuous circle. FAPESP has always recognized this from its inception. For example, it funded the first computers in Brazil. Now it’s funding a large number of Engineering Research Centers and the e-Science Program,” Medeiros said.
As the authors of the chapter note, many of the products and services that use models created by machine learning algorithms are designed to improve the quality of people’s lives or protect nature, giving rise to intelligent and sustainable environments.
“Much of what’s being done right now overlaps very significantly with data science and big data. Examples of this overlap include AI and machine learning. While AI can be seen as the science and engineering of making smart machines, machine learning investigates how computers can learn to solve problems on the basis of past experience,” said André de Carvalho, a professor at the University of São Paulo’s Institute of Mathematical and Computer Sciences (ICMC-USP).
Carvalho presented the Applied Research Center in Artificial Intelligence Recreating Environments (ARC-IARA), where he is principal investigator. This is one of six ARCs in AI selected under a joint call by FAPESP, the Ministry for Science, Technology and Innovation (MCTI) and the Brazilian Internet Steering Committee (CGI.br) (more at: agencia.fapesp.br/35896).
“ARC-IARA aims to stimulate the creation of inclusive and sustainable cities, functioning as a network of researchers from all over Brazil,” he said. The project will develop and implement solutions in healthcare, environmental management, education and mobility, starting with a copper mining town called Canaã dos Carajás, in Pará state (North Brazil), with about 38,000 inhabitants. “It’s grown very fast in the last ten years and is suffering from all the problems relating to lack of sufficient infrastructure and services. To guarantee a better future, the municipal government has decided to invest part of the royalties received from the copper mine in sustainable development,” he said.
Guarapuava, a city of 187,000 inhabitants in Paraná state (South Brazil), is also participating in ARC-IARA’s project. The mayor wants the region to be known as Genome Valley, and ARC-IARA is helping to plan and implement precision medicine solutions to make the city a health research laboratory. “We’ll be monitoring 5,000 inhabitants for 15 years with the aim of creating models to predict the emergence of rare and complex diseases,” Carvalho said. He also discussed ethical aspects of what he called responsible AI.
“The role of computing in health and agriculture should be emphasized, as it enables us to provide better services for citizens while also enhancing the quality and increasing the quantity of food,” said João Marcos Travassos Romano, Pro-Rector for Research at UNICAMP and principal investigator for the Brazilian Institute of Data Science (BI0S), one of the Engineering Research Centers supported by FAPESP via a cooperation agreement with MCTI.
Romano also highlighted the role of computing in decision-making support and advances in research on signal processing essential to the functioning of hardware and algorithms, as well as data storage.
The evolution of hardware and the development of new types of hardware were the focus of a presentation by Marcelo Knorich Zuffo, a professor at the University of São Paulo’s School of Engineering (POLI-USP) and principal investigator for the Interdisciplinary Center in Interactive Technologies (CITI). “Computing is evolving in faster and faster cycles of interaction among data, algorithms and hardware, with Moore’s law still valid nearly 60 years after it was formulated. Actually, Moore didn't formulate a law – he made the empirical observation that the complexity of electronics and hardware doubles every 18 months,” he said.
The proliferation of new devices is leading to unimagined advances, he added, and large volumes of data can be processed by new algorithms to complete the virtuous circle linking data, algorithms and hardware. He explained how his own research on hardware and devices is contributing to Project Tamar, which fights to protect sea turtles from extinction along much of the coast of Brazil.
In addition to Medeiros, Nakaya, Zuffo, Carvalho and Romano, the authors of the fifth chapter of FAPESP 60 Anos: A Ciência no Desenvolvimento Nacional also include Virgilio Almeida, a professor of computer science at the Federal University of Minas Gerais, UFMG.
Other participants in the online seminar included Ronaldo Pilli, Vice President of FAPESP; Vanderlan Bolzani, President of ACIESP; and Adriano Andricopulo, Executive Director of ACIESP.
A recording of the event can be watched at: www.youtube.com/watch?v=c-FF9afZ2vY.
Source: https://agencia.fapesp.br/39962
