Cancer. Cardiovascular disease. Neurodegenerative disorders. Andrew Emili, a professor of biochemistry and biology who leads the Center for Network Systems Biology, is looking for solutions to large-scale problems at a tiny level: subcellular. Understanding the fundamental biological processes that drive human health and disease requires the systematic mapping of the dynamic molecular network inside cells and tissues. To that end, Emili maps these networks using high-throughput experimental methods, centered on precision mass spectrometry. To process the results, he works with computing and data sciences faculty to develop and apply innovative algorithms and associated software. These partnerships lead to groundbreaking approaches in causal inference, integrative machine learning, and deep learning that support interdisciplinary research into some of our greatest challenges.

Explore systems biology and molecular networks with Andrew Emili at the Emili Lab.

Social networks have been around long before Facebook and Instagram, and they take far more diverse forms. Assistant Professor Neha Gondal in our Department of Sociology has studied the gamut. From moneylending in Renaissance Florence to academic hiring to health in public housing, she explores the relationship between networks and culture, and its effect on social inequalities. That’s a lot of ground. To cover it, she frequently employs cutting-edge statistical techniques for modeling social networks, including varieties of exponential random graph models and agent-based modeling to study the diverse contexts. One finding: the cultural basis of social networks helps produce, legitimize, and preserve social inequalities.

Learn more about Neha Gondal’s fascinating work and the data sciences that make it possible.

A leader in both education and research, Eric Kolaczyk, a professor of Mathematics & Statistics who leads the Hariri Institute for Computing and Computational Science & Engineering, works at the critical point where statistical theory and methods support human endeavors enabled by computing and engineering systems. He studies principles of design, representation, modeling, inference, prediction, and uncertainty quantification that are foundational to new paradigms for data measurement and analysis and, ultimately, key to gaining insight into everything from health and science to business and society. In addition to his research work, Eric founded the Masters in Statistical Practice Program (MSSP) – a pedagogically innovative, practice-centric degree program that exposes students to how statistics is applied in real-world scenarios.

Whether research or education, learn more about Eric Kolaczyk’s thinking on statistics here.

A global population of more than 7.5 billion offers the potential for staggering amounts of data we could harness in the interest of better health for all. Spearheading that effort is Assistant Professor Elaine Nsoesie at our School of Public Health. She applies novel data science methodologies to global health problems. In one example, she applied machine learning and statistical modeling to 291,443 Twitter posts about miscarriages to glean important insights into how social media may be used to seek and share health-related information. In another, she used artificial intelligence to analyze over one million reviews of foods on Amazon.com, finding that far more unsafe foods may need to be recalled. Both studies are data-intense, requiring large-scale resources and advanced data sciences, but the potential health benefits are enormous.

Global health is your health. Discover other ways Elaine Nsoesie is working to improve it.

Heart disease is the leading cause of death in the US and worldwide, so it is difficult to overstate the importance of the work being done by Professor Vasan Ramachandran, MD, at our School of Medicine. He studies heart health and the risk of heart disease within individuals and within populations. Employing harmonized and integrative analysis of high-dimensional and multilevel data across multiple cohorts, he is working to understand how health and risk of disease evolve in dynamic fashion under the influence of multiple factors, such as genes, family, community, geospatial location, and macrosocial factors. Involving many thousands of subjects, the research requires interdisciplinary cooperation with data sciences and computing on a scale that only an institution like BU can offer.

This just scratches the surface. Dig deeper into Vasan Ramachandran’s groundbreaking work.

Humans are making machines that “think,” but we barely understand their thought processes. Associate Professor Kate Saenko in our Department of Computer Science is making machine learning more human-like by developing artificial intelligence methods that allow machines to see, talk, and interact with the world in ways humans better understand. Her contributions include state-of-the-art deep-learning (neural network) models for object recognition in images, activity detection in video, automatic video captioning, and vision-based robotic object manipulation. Heady stuff, but ultimately practical for such applications as teaching robots to sort recyclables (one of her recent projects).

Explore more next-generation projects and AI methods at Kate Saenko’s website.

How do cells work together to perform complex behaviors, such as assembling into tissue, forming a biofilm, healing a wound, or developing into different cell types? Assistant Professor Allyson Sgro in our Department of Biomedical Engineering is working on ways to “eavesdrop” on cells’ internal information processing by integrating advances in biotechnology to visualize molecular processes, microscopy to capture images, and image processing and machine learning to extract the information. This research will allow us to identify the natural information-processing algorithms groups of cells use, potentially leading to new computational and synthetic living systems based on this system.

Learn more about Allyson Sgro’s breakthrough research and methods at her website.

You might say peace of mind is the research focus of Professor Adam Smith in our Department of Computer Science & Engineering. He works in the areas of data privacy and cryptography, providing techniques to analyze sensitive data sets, while guaranteeing the privacy of individuals whose data they contain. While privacy is his subject, he doesn’t keep his findings to himself. “Differential privacy,” a standard he invented, is the basis for deployed systems in the 2020 US Census and mobile data collections systems at Google, Apple, and Microsoft. Overall, his work has led to novel lines of inquiry within statistical inference, machine learning, algorithmic game theory, and legal scholarship. And, of course, a better night’s sleep for us all.

Visit Adam Smith’s website to find more—very public—information about his work.

Before we can make our complex media system more beneficial to the public, and democratic governance, we have to understand it. That’s what makes the work of Assistant Professor Chris Wells in our Division of Emerging Media Studies important. An innovator in the use of time-series analysis in communications, Wells found that, in 2016, social media—especially retweets of Donald Trump’s Twitter posts—drove coverage by traditional broadcast outlets, contributing to Trump’s meteoric rise. Working closely with colleagues from our Department of Mathematics & Statistics, his team developed a version of spectral clustering that further determined that those retweets were largely the work of small populations, particularly the alt-right.

What else is Chris Wells learning at the intersection of media and data sciences? He invites you to find out.

Are there actually solutions to seemingly intractable social problems such as rural suicide, religious violence, and sexual exploitation of children? Professor Wesley Wildman at our School of Theology is seeking answers. Using computer modeling, he creates populations of cognitively, emotionally, and behaviorally complex AI agents to deepen our understanding of such problems. These replica societies can then provide a platform for virtual experimentation, allowing evaluation of the likely costs, benefits, and unintended consequences of public policy proposals. The simulations are data-hungry and depend on careful theoretical integration across numerous disciplines—just the sort of thing BU is built for.

Wesley Wildman’s research center offers more examples of how he brings together the humanities and data sciences.

As the internet continues to grow, so does interest in the research of Associate Professor Georgios Zervas at BU’s Questrom School of Business. He investigates how online marketplaces, especially shared economies such as Yelp and Airbnb, affect consumer and firm behavior. This places him squarely at the intersection of marketing, data science, and economics, assembling novel data sets that he analyzes through a variety of methods such as causal inference, structural modeling, and machine learning. The insights generated by his work help shape the future of online business and contribute to our broader understanding of the digital economy.

What has data revealed about online marketplaces? Find out at Georgios Zervas’ website.