Interview with Dr. Michael Shiloh - Part 2: The Changing Tools of Research

One of the things that interested me while working in the lab was how many different tools can be used to answer questions in infectious disease. I wanted to ask Dr. Shiloh how his own research has changed and how he thinks newer technologies, including AI, are affecting the field.

What tools or technologies do you use most often?

Michael Shiloh: Personally, the tool I probably use most often is a computer. Much of my job now involves writing: grants to fund our research, papers to communicate our findings, and protocols that allow us to conduct our work safely and appropriately. In the laboratory, though, we use a wide range of approaches. We do a lot of cell culture and genetics, as well as microbiology, where we infect cells with bacteria and other organisms. We also sometimes use animal models to understand how infections cause disease and how bacteria spread from infected to uninfected animals.

How has your work changed in the past few years?

Michael Shiloh: When I started my lab about 15 years ago, our work focused primarily on innate immunity, particularly how immune cells such as macrophages recognize bacteria and attempt to destroy them. Since then, we've expanded into two additional areas. One is mucosal immunology, where we're

interested in how bacteria enter the body. The other is neuroscience and neuroimmunology. We've become interested in how bacterial molecules activate the nervous system and trigger reflexes such as coughing, which can then contribute to disease transmission. So we haven't moved away from innate immunity as much as expanded outward from it, into questions at the intersection of immunology, microbiology, and neuroscience.

How is technology, including AI, changing the way scientists study infectious diseases?

Michael Shiloh: The AI revolution has only just begun, and I don't think we've yet seen its full impact on infectious disease research. Computational biology was already transforming the field before the recent growth of generative AI. It has allowed scientists to model interactions between pathogens and their hosts, predict protein structures, and use those structures to help design antibiotics and other therapeutics. With newer AI tools, I think we're still learning what they can do. AI can help generate hypotheses and predict outcomes, but ultimately someone still has to perform the experiment. A prediction isn't a scientific result until you test whether it's actually true.

What's a recent scientific development that excited you?

Michael Shiloh: My lab has developed a strong interest in neuroscience and neuroimmunology, and one of the most exciting developments over the past decade has been our growing understanding of how closely the nervous and immune systems communicate with one another. There's also a third participant in that conversation: microorganisms. Those can be organisms that normally live in our bodies as part of the microbiome, or pathogens that we encounter during an infection. We're only beginning to understand this three-way communication among microorganisms, the immune system, and the nervous system. I think it's an incredibly exciting area of biology.

What's one emerging technology students should be paying attention to?

Michael Shiloh: One technology that has already made major strides in infectious disease, particularly in diagnostics, is metagenomic sequencing. The basic idea is that you can take a clinical sample, such as blood or cerebrospinal fluid, collect the nucleic acids in that sample, sequence them, and then look for evidence of an infectious organism. Traditionally, diagnosing an infection often requires you to have some idea of what you're looking for, or to grow the organism in culture. Metagenomic sequencing can be much less targeted. You don't necessarily have to grow the organism or know in advance what you're looking for. In principle, that

allows you to identify both known pathogens and potentially novel ones. That ability to identify infections rapidly and without making as many assumptions in advance is an area of growing interest and potentially enormous value in infectious disease.