Patrick Cahan is an assistant professor of biomedical engineering and a member of the Institute for Cell Engineering. He devises ways to find out whether lab-grown stem cells have differentiated into the desired cell type.

I see you spent time in Chile. Can you tell me about that?

CAHAN: When I graduated with my bachelor’s degree in computer science, I had no idea what I was going to do next. But my girlfriend knew what she was doing, and so I followed her to Chile, where she worked for a nongovernmental organization promoting public health. I met some expats who were starting a company making software for e-commerce websites — this was 2001. I came onboard, and the company ended up growing tremendously over the year I was there, which was exciting. And I enjoyed living in what was, to me, an exotic environment. But I continually felt like someone else had already solved the problems I was working on.

How did you find your way to your current field?

CAHAN: While I was still in Chile, a friend in medical school recommended I look into this new field called bioinformatics, which I did. But because I had almost zero background in biology, I moved back home, took a genetics course at Saint Louis University and was lucky enough to find a biology lab that took me on for a bioinformatics research project. One of the most important things I learned from that summer research project was that I really, really liked computational biology, so I went on to get a master’s degree and a Ph.D. in the field. First, I focused on gene expression analysis, profiling which genes are used, and to what extent, in certain cell types under certain conditions; later, I investigated links between the number of copies of certain genes and leukemia risk.

Then, as I transitioned from a graduate student to a postdoc, I became enthralled with the idea of converting one type of cell into another, such as making skin cells into pluripotent stem cells. So I joined a stem cell lab at Boston Children’s Hospital and Harvard, and that experience really set me on my current course.

And what is that?

CAHAN: The research group I was in works on one of the holy grails of the stem cell field: finding a way to push induced pluripotent stem cells to become what are called hematopoietic stem cells (HSCs). HSCs are the master blood cell and produce all of the cell types that make up our blood — B cells, T cells, macrophages, etc. The idea is that for patients who need a bone marrow transplant, you could make induced pluripotent stem cells, correct any genetic errors that caused their disease — for example, sickle cell disease — and then you would use those patient-matched, corrected stem cells to make HSCs for transplantation. Like I said, it is a holy grail.

One barrier to making that happen was that there wasn’t a definitive way of knowing where the goal post was — that is, whether stem cells grown in the lab had truly become HSCs. So I developed a computational platform that defines HSCs based on whether they regulate gene expression in the same way as real HSCs. I applied this to many other cell types as well — it is a generic system.

Apart from defining the goal posts, do these tools help bring them closer?

CAHAN: Exactly! When we run a comparison of the lab-grown stem cells to HSCs or other cell types, our software can generate likely combinations of transcription factors needed to dial the right genes up or down in the lab-grown cells and get them to the endpoint we’re after. Ultimately, rather than engineering the transcription factors directly, we’d like to find the developmental signaling pathways that control the transcription factors and tweak the cells’ environment to steer those pathways in the right direction.

In addition to custom hematopoietic stem cells for bone marrow transplant, what are other likely applications of specialized, lab-grown stem cells?

CAHAN: Potentially, being able to direct induced pluripotent stem cells to grow into the desired cell type would enable researchers to regenerate any diseased or damaged tissue. For example, right now, I’m looking at the cells that produce the cartilage inside joints, because replacing those cells in patients with osteoarthritis could be an effective treatment. But that and other medical applications are quite a ways off. On the other hand, there is tremendous enthusiasm right now for using lab-grown cells to model diseases, and I think that this, coupled with emerging genome engineering technologies, will revolutionize our understanding of many diseases and lead to more effective treatments in the years to come.