Bridging military service and engineering

For graduate students Kelsey Pittman and Jacqueline Orr, service in the U.S. military led to their interest in engineering, and to the MIT Department of Civil and Environmental Engineering (CEE).

Pittman’s first exposure to the military and engineering took place during her undergraduate years at the United States Military Academy West Point. 

“I remember back in high school, my dad kind of planted the seed of going to a military academy,” says Pittman.  While she admitted to feeling overwhelmed about the prospect of going to college at that time, her father’s rationale for West Point resonated with her. “I’m a structured person and I like routine,” she says — two aspects the environment at West Point provides.

While Pittman’s father hadn’t attended a military academy or served in the military, he was a member of the Federal Bureau of Investigation for 25 years, and her family connections provided Pittman with valuable perspectives on West Point. It ended up being the only undergraduate program Pittman applied to. “I just wanted to be part of something bigger than myself, and all the opportunity West Point could give was pretty incredible,” she says.

Pittman’s parents also recognized her passion for design and encouraged her to consider a career in architecture. Although West Point didn’t offer an architecture program, she chose civil engineering, a field that allowed her to combine her love of math and design.

After graduating, she was commissioned as an engineer officer in the U.S. Army and has served for over seven years. She is now pursuing her graduate education at MIT in structural engineering with advisor John Ochsendorf, professor of civil and environmental engineering and architecture. Pittman is researching Gothic-style infrastructure for its masonry resiliency and stability over time, specifically Beauvais Cathedral and its structural safety. One of the reasons she chose to pursue her graduate studies in CEE was the department’s openness to explore diverse research opportunities.

“I was really drawn to the ability to carve my own research niche and have the freedom to figure out what really interests me, rather than being presented with a limited set of research options,” says Pittman.

After receiving her master’s degree, Pittman will return to West Point as a faculty member for three years and then continue her service obligation in the Army. She credits her mentors at West Point as being instrumental in her academic and professional journey and hopes to play a role in shaping the lives of future generations of cadets. 

“I have incredible mentors that I still talk to, and I really wanted to be able to go back and give back to a place, and the people that gave me so much support and room to grow and find my passion. Every step has been made in my career so far to get back to West Point and teach in the civil engineering department.”

Pittman also acknowledges and values the Army for the opportunities it has provided her, particularly the chance to pursue her master’s degree at MIT, the relationships she has built along the way and career path it has opened.

“I’ve enjoyed getting to know the soldiers from all over the world and seeing them in this environment where you might give each other a hard time, but at the end of the day you know that you have each other’s back.”

Jacqueline Orr, also a U.S. Military Academy graduate, is currently pursuing a master’s degree in structural engineering under the guidance of Josephine Carstensen, the Gilbert W. Winslow Career Development Associate Professor for Civil and Environmental Engineering. Inspired by her father to pursue a strong foundation in math and science, she earned a bachelor’s degree in mechanical engineering. After graduation, she fulfilled her service obligation and served for six years as a member of the 173rd Airborne Brigade based in Vicenza, Italy — a unit renowned for its history, combat readiness, and crucial part of the Army’s joint integration with NATO. 

Reflecting on her experience, Orr says, “Airborne units, like many great units in the Army, require overcoming an additional litmus test — in this case, conquering the fear of jumping from high-performance aircraft, hundreds of feet above the ground.”

While she enjoyed her time in the Army, her experiences ultimately led her to pursue a career more closely aligned with her passion for engineering. “When I was studying mechanical engineering, I developed a strong interest in structures during my senior design project,” she says.

She particularly enjoyed learning how to model structures and analyze how they respond to various forces. She felt that the traditional methods taught in her classes lacked an optimization component, which sparked her interest in topology optimization as a potential solution.

This desire to further explore topology optimization — specifically in relation to structures and their behavior under different forces — motivated her to seek graduate programs specializing in this field. Orr applied for and was awarded a Department of Defense (DoD) SMART Scholarship that brought her to MIT to study topology optimization in the Carstensen Lab.

“MIT was the ideal institution to pursue this research due to Professor Carstensen’s expertise and innovative work happening in the civil and environmental engineering department,” Orr says.

Looking ahead, Orr plans to apply the knowledge gained at MIT to a research-oriented career as part of her obligation as a DoD SMART Scholar. But for now, she’s adjusting to life as a graduate student. “I’m really enjoying my classes and getting to know people in the lab — it’s been an amazing experience,” she adds.  

Startup turns mining waste into critical metals for the U.S.

At the heart of the energy transition is a metal transition. Wind farms, solar panels, and electric cars require many times more copper, zinc, and nickel than their gas-powered alternatives. They also require more exotic metals with unique properties, known as rare earth elements, which are essential for the magnets that go into things like wind turbines and EV motors.

Today, China dominates the processing of rare earth elements, refining around 60 percent of those materials for the world. With demand for such materials forecasted to skyrocket, the Biden administration has said the situation poses national and economic security threats.

Substantial quantities of rare earth metals are sitting unused in the United States and many other parts of the world today. The catch is they’re mixed with vast quantities of toxic mining waste.

Phoenix Tailings is scaling up a process for harvesting materials, including rare earth metals and nickel, from mining waste. The company uses water and recyclable solvents to collect oxidized metal, then puts the metal into a heated molten salt mixture and applies electricity.

The company, co-founded by MIT alumni, says its pilot production facility in Woburn, Massachusetts, is the only site in the world producing rare earth metals without toxic byproducts or carbon emissions. The process does use electricity, but Phoenix Tailings currently offsets that with renewable energy contracts.

The company expects to produce more than 3,000 tons of the metals by 2026, which would have represented about 7 percent of total U.S. production last year.

Now, with support from the Department of Energy, Phoenix Tailings is expanding the list of metals it can produce and accelerating plans to build a second production facility.

For the founding team, including MIT graduates Tomás Villalón ’14 and Michelle Chao ’14 along with Nick Myers and Anthony Balladon, the work has implications for geopolitics and the planet.

“Being able to make your own materials domestically means that you’re not at the behest of a foreign monopoly,” Villalón says. “We’re focused on creating critical materials for the next generation of technologies. More broadly, we want to get these materials in ways that are sustainable in the long term.”

Tackling a global problem

Villalón got interested in chemistry and materials science after taking Course 3.091 (Introduction to Solid-State Chemistry) during his first year at MIT. In his senior year, he got a chance to work at Boston Metal, another MIT spinoff that uses an electrochemical process to decarbonize steelmaking at scale. The experience got Villalón, who majored in materials science and engineering, thinking about creating more sustainable metallurgical processes.

But it took a chance meeting with Myers at a 2018 Bible study for Villalón to act on the idea.

“We were discussing some of the major problems in the world when we came to the topic of electrification,” Villalón recalls. “It became a discussion about how the U.S. gets its materials and how we should think about electrifying their production. I was finally like, ‘I’ve been working in the space for a decade, let’s go do something about it.’ Nick agreed, but I thought he just wanted to feel good about himself. Then in July, he randomly called me and said, ‘I’ve got [$7,000]. When do we start?’”

Villalón brought in Chao, his former MIT classmate and fellow materials science and engineering major, and Myers brought Balladon, a former co-worker, and the founders started experimenting with new processes for producing rare earth metals.

“We went back to the base principles, the thermodynamics I learned with MIT professors Antoine Allanore and Donald Sadoway, and understanding the kinetics of reactions,” Villalón says. “Classes like Course 3.022 (Microstructural Evolution in Materials) and 3.07 (Introduction to Ceramics) were also really useful. I touched on every aspect I studied at MIT.”

The founders also received guidance from MIT’s Venture Mentoring Service (VMS) and went through the U.S. National Science Foundation’s I-Corps program. Sadoway served as an advisor for the company.

After drafting one version of their system design, the founders bought an experimental quantity of mining waste, known as red sludge, and set up a prototype reactor in Villalón’s backyard. The founders ended up with a small amount of product, but they had to scramble to borrow the scientific equipment needed to determine what exactly it was. It turned out to be a small amount of rare earth concentrate along with pure iron.

Today, at the company’s refinery in Woburn, Phoenix Tailings puts mining waste rich in rare earth metals into its mixture and heats it to around 1,300 degrees Fahrenheit. When it applies an electric current to the mixture, pure metal collects on an electrode. The process leaves minimal waste behind.

“The key for all of this isn’t just the chemistry, but how everything is linked together, because with rare earths, you have to hit really high purities compared to a conventionally produced metal,” Villalón explains. “As a result, you have to be thinking about the purity of your material the entire way through.”

From rare earths to nickel, magnesium, and more

Villalón says the process is economical compared to conventional production methods, produces no toxic byproducts, and is completely carbon free when renewable energy sources are used for electricity.

The Woburn facility is currently producing several rare earth elements for customers, including neodymium and dysprosium, which are important in magnets. Customers are using the materials for things likewind turbines, electric cars, and defense applications.

The company has also received two grants with the U.S. Department of Energy’s ARPA-E program totaling more than $2 million. Its 2023 grant supports the development of a system to extract nickel and magnesium from mining waste through a process that uses carbonization and recycled carbon dioxide. Both nickel and magnesium are critical materials for clean energy applications like batteries.

The most recent grant will help the company adapt its process to produce iron from mining waste without emissions or toxic byproducts. Phoenix Tailings says its process is compatible with a wide array of ore types and waste materials, and the company has plenty of material to work with: Mining and processing mineral ores generates about 1.8 billion tons of waste in the U.S. each year.

“We want to take our knowledge from processing the rare earth metals and slowly move it into other segments,” Villalón explains. “We simply have to refine some of these materials here. There’s no way we can’t. So, what does that look like from a regulatory perspective? How do we create approaches that are economical and environmentally compliant not just now, but 30 years from now?”

The Friday Roundup – Teleprompter Tips and YouTube Videos

9 Tips On How To Use A Teleprompter I have a couple of clients at the moment who are using teleprompters because of the nature of the content they produce. It is important for them to ensure that when they are covering a subject they stay…

3 questions: Leveraging insights to enable clinical outcomes

Associate Professor Thomas Heldt joined the MIT faculty in 2013 as a core member of the Institute for Medical Engineering and Science (IMES) and the Department of Electrical Engineering and Computer Science. Additionally, Heldt is a principal investigator with MIT’s Research Laboratory of Electronics (RLE), and he directs the Integrative Neuromonitoring and Critical Care Informatics Group in IMES and RLE. He was recently named an associate director of IMES, where he will focus on internal affairs, among other duties. 

Heldt received his Medical Engineering and Medical Physics (MEMP) PhD from the Harvard-MIT Program in Health Sciences and Technology (HST) in 2004. Heldt’s research interests include signal processing, estimation and identification of physiological systems, mathematical modeling, model identification to support real-time clinical decision making, monitoring of disease progression, and titration of therapy, primarily in neurocritical and neonatal critical care. Here, Heldt describes how he collaborates closely with MIT colleagues and others at Boston-area hospitals, and how his research uses and analyzes physiologic data to aid clinical action.

Q: How does your research apply to solving clinical needs?

A: We look at current clinical environments and observe the volumes of multimodal physiologic waveform data that are collected on patients in critical care, peri-operative care, or even emergency care. Much of this data is typically visually reviewed by the clinicians and subsequently discarded after a holding period of just a few days. We thus lose the opportunity for more systematic analyses and for deriving patient-specific insights. Critical to such analyses of these data streams is a deep understanding of the relevant physiology at the time scales of interest. We leverage insights from physiology, formulated as reduced order mathematical models capturing the essential mechanisms that enable clinical action. We have applied this approach successfully to estimate intracranial pressure noninvasively, to make diagnostic decisions based on the analysis of the shape of the capnogram, and, are currently using ultrasound-based approaches to detect embolic events in patients on life support, such as ventricular assist devices or extracorporeal membrane oxygenation. 

Q: You work closely with colleagues across MIT, and with clinicians at Boston-area hospitals, including Boston Children’s Hospital (where you hold a courtesy research appointment in neurology), Boston Medical Center (neurosurgery), and Massachusetts General Hospital (emergency medicine). What has been the fruit of some of these collaborations — what is the impact on your research?

A: Boston is a fantastic place to conduct translational research that crosses from our laboratories at MIT into the clinical environments for validation in the actual target patient population! The collaborative disposition and forward-thinking mindset of our clinician colleagues have really been fundamentally enabling for our research and have provided amazing mentoring to our students, postdocs, and me. We have collected validation data in brain-injured patients in the ICUs [intensive care units] at Boston Medical Center, Boston Children’s Hospital (BCH), and Beth Israel Deaconess Medical Center (BIDMC); we have collected pilot and validation data for our capnography work in the emergency departments at BCH and BIDMC; we have collected data for our emboli work in the operating rooms and ICUs at BCH, and have analyzed the medical records of the neonatal ICU at BIDMC and the emergency department at Massachusetts General Hospital.

Our work with the neonatologist at BIDMC was focused on analyzing the monitoring alarm patterns in the neonatal ICU. We counted a staggering 177 alarms/baby/day, or one alarm every eight minutes on average, per baby. And this is a 54-bed neonatal ICU operating close to capacity every day! Such volumes of alarms contribute to noise pollution in an environment that should ideally be very calm. Additionally, since most of the alarms are nuisance alarms or do not require any clinical intervention, the clinical staff becomes desensitized to the alarm load and might end up ignoring truly important events. We analyzed the alarm patterns and alarm thresholds for a particular type of heart rate alarms and recommended a change in thresholds. This resulted in a 50 percent reduction in heart rate alarms per patient per day. Initially, the clinical staff had to file weekly reports to make sure the reduction in the alarm rate did not result in missed or adverse events. After about three months without a single reportable event, the hospital safety committee approved the change.

With colleagues from the MGH Department of Emergency Medicine, we developed and tested a triage rule to identify patients at risk of septic shock. At the time, the MGH ED [emergency department] saw more than 120,000 patients/year, and around 75 percent of patients ending up in the ICU with severe sepsis and septic shock came through the emergency department. Hence, ED triage was the first point of patient contact and the first opportunity to flag patients for possible sepsis and septic shock and initiation of early goal-directed therapy. One result of our work was a significant reduction in the time to appropriate antibiotic administration in the emergency department. The work was subsequently validated in other Partners hospitals and implemented in the electronic medical record system of Partners-affiliated hospitals. 

Q: Can you talk a bit about your background, and about how you became interested in systems-physiology and biomedicine? What are your goals for your research, and for your career?

A: That is a longer story! In short, I started out studying physics back in Germany. After a while, I got interested in applying concepts I learned in physics to physiology and medicine, so I designed my own MD/PhD program by picking up medicine as a second major. Through some fortuitous events, I ended up attending surgeries for congenital heart defects for about a term. This was a very formative experience, and almost pushed me toward dropping physics and going all-out on becoming a surgeon. However, I had also always wanted to spend part of my education abroad and had applied to various universities in the U.S. I ended up getting admitted to the graduate physics program at Yale and spent a couple of years doing nonlinear optics. While I loved the work at Yale and had a fantastic mentor, I missed the clinical exposure and application of my work to medicine. I had heard about the HST program and decided to send in an application. I joined the MEMP program in 1997 and have been at MIT ever since.

In our current research, we are very interested in providing better monitoring modalities for patients with brain injuries. We are developing novel algorithmic and device approaches so we can replace the current invasive monitoring modalities with entirely noninvasive ones and provide additional clinically actionable information that gives insights on the physiology of the injured brain and can help guide treatment decision. I want to see some of these technologies through to routine deployment at the bedside.

The great thing about being in IMES and MIT is that we everybody is very collaborative. What I am looking forward to is much of the same, working with colleagues in IMES on important problems that none of us is be able to tackle alone, but that together we have a real chance of tackling — and having fun along the way! 

Connecting the US Coast Guard to MIT Sloan

Jim Ellis II SM ’80 first learned about a special opportunity for members of the U.S. Coast Guard while stationed in Alaska.

“My commander had received a notice from headquarters about this opportunity. They were asking for recommendations for an officer who might be interested,” says Ellis.

The opportunity in question was the MIT Sloan Fellows program, today known as the MIT Sloan Fellows MBA (SFMBA) program. Every year for 50 years, the Coast Guard has nominated a service member to apply to the program. Fifty Sloan Fellows and two Management of Technology participants have graduated since 1976, and the 53rd student is currently enrolled.

With his tour nearly over, Ellis followed his commander’s recommendation to apply. The Coast Guard nominated him and his application to MIT Sloan School of Management was accepted. In 1980, Ellis became the fifth-ever Coast Guard Sloan Fellow to graduate due to the special arrangement.

“My experience at MIT Sloan has been instrumental throughout my entire career,” says Ellis, who, with his wife Margaret Brady, established the Ellis/Brady Family Fund to support the MIT Sloan Sustainability Initiative and graduate fellowships through the MIT Sloan Veterans Fund.

“The success of the people who have been through the program is a testament to why the Coast Guard continues the program,” he adds.

The desire to change the world

Throughout its 163-year history, MIT has maintained strong relationships with the U.S. military through programs like the MIT Reserve Officers’ Training Corps, the 2N Graduate Program in Naval Architecture and Marine Engineering, and more.

The long-standing collaboration between MIT Sloan and the Coast Guard adds to this history. According to Johanna Hising DiFabio, assistant dean for executive degree programs at MIT Sloan, it demonstrates the Coast Guard’s dedication to leadership development, as well as the unique benefits MIT Sloan has to offer service members.

This is especially evident in the careers of the 52 Coast Guard Sloan Fellow alumni, many of whom the program often invites to speak to current students. “It is inspiring to hear our alumni reflect on how this education has significantly influenced their careers and the considerable impact they have had on the Coast Guard and the global community,” says DiFabio.

Captain Anne O’Connell MBA ’19 says, “It is very rewarding to be able to pay it back, to look for those officers coming up behind you who should absolutely be offered the same opportunities, and to help them chart that course. I think it’s hugely important.”

One of the most notable Coast Guard Sloan Fellows is Retired Admiral Thad Allen SM ’89, who served as commandant of the Coast Guard from 2006 to 2010. One of the service’s youngest-ever flag officers, Allen is a figure beloved by current and former guardsmen. As commandant, he embraced new digital technologies, championed further arctic exploration, and solidified relations with the other armed services, federal partners, and private industry.

“When you leave MIT Sloan, you want to change the world,” says Allen.

Inspired by his father, who enlisted after the attack on Pearl Harbor, Allen attended the U.S. Coast Guard Academy and subsequently held various commands at sea and ashore during a career spanning four decades.

A few years before the end of his second decade, Allen learned about the Sloan Fellows Program through a service-wide solicitation. “The people I worked for believed this would be a great opportunity, and that it would match with my skill set,” says Allen. With the guidance of his senior captains, he applied to MIT Sloan.

Allen matriculated with a cohort whose members included Carly Fiorina SM ’89, former CEO of Hewlett-Packard; Daniel Hesse SM ’89, former CEO of Sprint; and Robert Malone SM ’89, former chair and president of BP America. Though he initially felt a sharp disconnect between his national service experience and their global private sector knowledge, Allen realized everyone in the cohort were becoming his peers.

Strong bonds with global perspectives

Like Allen, many of the Coast Guard Sloan Fellows acknowledge just how powerful their cohorts were when they matriculated, as well as how influential they have remained since.

“I have classmates with giant perspectives and unique expertise in places all over the world. It’s remarkable,” says Retired Commander Catherine Kang MBA ’06, who served as deputy of financial transformation for Allen.

The majority of SFMBA candidates come to Cambridge from around the world. For example, the 2023–24 cohort comprised 76 percent international citizens.

For Coast Guard Sloan Fellows with decades of domestic experience, their cohort’s global perspectives are as novel as they are informative. As Retired Captain Gregory Sanial SM ’07 explains, “We had students from 30 to 40 different countries, and I had the opportunity to learn a lot about different parts of the world and open up my mind to many different experiences.”

After the Coast Guard, Sanial pursued a doctoral degree in organizational leadership and a career in higher education that, professionally, has kept him stateside. Yet the bonds he built at MIT Sloan remain just as strong and as international as they were when he first arrived.

Many Coast Guard Sloan Fellows attribute this to the program’s focus on cooperation and social events.

“What impressed me most when I first got there were the team-building exercises, which made a difference in getting a group of diverse people to really gel and work together,” says Retired Captain Lisa Festa SM ’92, SM ’99. “MIT Sloan takes the time at the beginning to invest in you and to make sure you know the people you’re going through school with for the next year.”

The most recent Coast Guard Sloan Fellow alumnus, Commander Mark Ketchum MBA ’24, says his cohort’s connections are still fresh, but he believes they will last a lifetime. Considering the testimonies of his predecessors, this may very well be the case.

“My cohort made me stronger, and I would like to think that I imparted my strengths onto my classmates,” says Ketchum.

Big challenges with high impacts

Before earning the Coast Guard’s nomination and an acceptance letter from the SFMBA program, potential Sloan Fellows have already served in various leadership positions. Once they graduate, the recognition and distinction that comes with an MIT Sloan degree is quick.

So, too, are the more challenging leadership tracks.

After graduation, Allen served as deputy program manager for the Coast Guard’s shipbuilding program at the behest of the then-commandant. “For the agency head to say, ‘This is a bad problem, so I’m picking the next graduate from MIT Sloan,’ is indicative of the program’s cachet value,” he says. Allen then served in the office of budget and programs, a challenging and rewarding post that has become a hub for Coast Guard Sloan Fellows past, present, and future.

Like Rear Admiral Jason Tama MBA ’11 and Captain Brian Erickson MBA ’21, both of whom credit the office with introducing them to the vigorous work ethic necessary for both obtaining an MIT Sloan education and for becoming an effective leader.

“Never in a thousand years would I have gone on the resource management path until a mentor told me it would be one of the most challenging and high-impact things I could do,” says Tama. “You can never be fully prepared for the Sloan Fellows experience, but it can and will change you for the better. It changed the way I approach problems and challenges.”

“I owe MIT for the senior-level opportunities I’ve had in this organization, and I will probably owe them for some of the opportunities I may get in the future,” adds Erickson. “You should never, ever say no to this opportunity.”

From the early cohorts of Ellis, Allen, and Festa, to more recent alumni like O’Connell, Kang, and Ketchum, Coast Guard Sloan Fellows from the past half-century echo Erickson and Tama’s sentiments when asked about how MIT Sloan has changed them. Words like “challenge,” “opportunity,” and “impact” are used often and with purpose.

They believe joining the SFMBA program as up-and-coming senior leaders is an incredible opportunity for the individual and the Coast Guard, as well as the MIT community and the world at large.

“I am excited to see this tradition carry on,” says Tama. “I hope others who are considering it can see the potential and the value, not only for themselves, but for the Coast Guard as well.”

Participation by U.S. Coast Guard members in this highlight of prior MIT Sloan Fellows is not intended as, and does not constitute an endorsement of, the MIT Sloan Fellows MBA program or MIT by either the Department of Homeland Security or the U.S. Coast Guard.

3 Steps for Retailers to Generate and Capture Value from AI Investments

The retail sector is growing and increasingly competitive as companies vie for consumers’ attention and wallets. According to the National Retail Federation, core sales rose 3.2% year-over-year in the first half of 2024, and total sales are forecast to eclipse 2023 by between 2.5% and 3.5%….

Dhiren Bhatia, Co-Founder & CEO of Inventive AI – Interiew Series

Dhiren Bhatia is the Co-Founder & CEO of Inventive AI,  an AI-powered RFP and questionnaire response management platform. RFP stands for Request for Proposal, a formal document issued by organizations to invite vendors or service providers to submit proposals for specific projects or services. The RFP…

How Google’s AlphaChip is Redefining Computer Chip Design

The evolution of artificial intelligence (AI) is rapidly changing how we work, learn, and connect, transforming industries around the globe. This shift is primarily driven by AI’s advanced ability to learn from larger datasets. While bigger models boost AI’s data processing power, they also require more…

A causal theory for studying the cause-and-effect relationships of genes

By studying changes in gene expression, researchers learn how cells function at a molecular level, which could help them understand the development of certain diseases.

But a human has about 20,000 genes that can affect each other in complex ways, so even knowing which groups of genes to target is an enormously complicated problem. Also, genes work together in modules that regulate each other.

MIT researchers have now developed theoretical foundations for methods that could identify the best way to aggregate genes into related groups so they can efficiently learn the underlying cause-and-effect relationships between many genes.

Importantly, this new method accomplishes this using only observational data. This means researchers don’t need to perform costly, and sometimes infeasible, interventional experiments to obtain the data needed to infer the underlying causal relationships.

In the long run, this technique could help scientists identify potential gene targets to induce certain behavior in a more accurate and efficient manner, potentially enabling them to develop precise treatments for patients.

“In genomics, it is very important to understand the mechanism underlying cell states. But cells have a multiscale structure, so the level of summarization is very important, too. If you figure out the right way to aggregate the observed data, the information you learn about the system should be more interpretable and useful,” says graduate student Jiaqi Zhang, an Eric and Wendy Schmidt Center Fellow and co-lead author of a paper on this technique.

Zhang is joined on the paper by co-lead author Ryan Welch, currently a master’s student in engineering; and senior author Caroline Uhler, a professor in the Department of Electrical Engineering and Computer Science (EECS) and the Institute for Data, Systems, and Society (IDSS) who is also director of the Eric and Wendy Schmidt Center at the Broad Institute of MIT and Harvard, and a researcher at MIT’s Laboratory for Information and Decision Systems (LIDS). The research will be presented at the Conference on Neural Information Processing Systems.

Learning from observational data

The problem the researchers set out to tackle involves learning programs of genes. These programs describe which genes function together to regulate other genes in a biological process, such as cell development or differentiation.

Since scientists can’t efficiently study how all 20,000 genes interact, they use a technique called causal disentanglement to learn how to combine related groups of genes into a representation that allows them to efficiently explore cause-and-effect relationships.

In previous work, the researchers demonstrated how this could be done effectively in the presence of interventional data, which are data obtained by perturbing variables in the network.

But it is often expensive to conduct interventional experiments, and there are some scenarios where such experiments are either unethical or the technology is not good enough for the intervention to succeed.

With only observational data, researchers can’t compare genes before and after an intervention to learn how groups of genes function together.

“Most research in causal disentanglement assumes access to interventions, so it was unclear how much information you can disentangle with just observational data,” Zhang says.

The MIT researchers developed a more general approach that uses a machine-learning algorithm to effectively identify and aggregate groups of observed variables, e.g., genes, using only observational data.

They can use this technique to identify causal modules and reconstruct an accurate underlying representation of the cause-and-effect mechanism. “While this research was motivated by the problem of elucidating cellular programs, we first had to develop novel causal theory to understand what could and could not be learned from observational data. With this theory in hand, in future work we can apply our understanding to genetic data and identify gene modules as well as their regulatory relationships,” Uhler says.

A layerwise representation

Using statistical techniques, the researchers can compute a mathematical function known as the variance for the Jacobian of each variable’s score. Causal variables that don’t affect any subsequent variables should have a variance of zero.

The researchers reconstruct the representation in a layer-by-layer structure, starting by removing the variables in the bottom layer that have a variance of zero. Then they work backward, layer-by-layer, removing the variables with zero variance to determine which variables, or groups of genes, are connected.

“Identifying the variances that are zero quickly becomes a combinatorial objective that is pretty hard to solve, so deriving an efficient algorithm that could solve it was a major challenge,” Zhang says.

In the end, their method outputs an abstracted representation of the observed data with layers of interconnected variables that accurately summarizes the underlying cause-and-effect structure.

Each variable represents an aggregated group of genes that function together, and the relationship between two variables represents how one group of genes regulates another. Their method effectively captures all the information used in determining each layer of variables.

After proving that their technique was theoretically sound, the researchers conducted simulations to show that the algorithm can efficiently disentangle meaningful causal representations using only observational data.

In the future, the researchers want to apply this technique in real-world genetics applications. They also want to explore how their method could provide additional insights in situations where some interventional data are available, or help scientists understand how to design effective genetic interventions. In the future, this method could help researchers more efficiently determine which genes function together in the same program, which could help identify drugs that could target those genes to treat certain diseases.

This research is funded, in part, by the MIT-IBM Watson AI Lab and the U.S. Office of Naval Research.