I want everyone in the room, right now to take a deep breath in.  Now breathe out as if you’re blowing on a dandelion, pushing the air from deep in your belly.

Every breath you take, just like that one, is powered by a thin, dome-shaped muscle; called the diaphragm.  

We take it for granted but one of the most significant moments in our life as we enter the world is our first breath. 

And that first breath depends on the diaphragm. Within seconds after birth the diaphragm contracts forcing the lungs to expel the fluid that fills them and expand with air.

Unfortunately for some newborns, that first breath isn’t so easy. When the diaphragm doesn’t fuse correctly during pregnancy - a hole can be left that disrupts breathing. 

The consequences are devastating for families. In the US 30% of babies with this condition will not make it.

For the vast majority of cases we don’t know why this disease happens and this is in large part because we don’t really understand how a diaphragm develops normally during pregnancy. 

This is surprising. It’s an organ so essential for life, for all of us as we continue to breathe right now. 

To understand how the diaphragm develops we need to know how its blueprint, the DNA, is read and interpreted by each individual cell of the diaphragm.

Our research with collaborators at UC San Diego uses a powerful technology that allows us to map the DNA to look at not only the genes that are turned on in each cell, but also the instructions that switch those genes on and off.

Using mouse embryos as a model, we have successfully measured the genes and their DNA switches in approximately 10 thousand individual cells in the diaphragm.

Using this data, we have described key cells types including the nerves and the muscles that control the movement of the diaphragm as well as these really interesting cells that produce a lubricant that with every breath we take allows the lungs to slide smoothly against the diaphragm.

This blueprint is now helping us to understand why how and why diaphragm development can go wrong. 

We are now working with clinicians to discover mutations in patients that can lead to this disease in newborns.

This knowledge could one day help doctors diagnose the condition earlier, improve genetic counselling, or may even guide new treatment strategies.

Every first breath a newborn takes and every breath we take depends on the diaphragm, a muscle we rarely think of, but couldn’t live without.

Have you ever wondered, how to pamper that demanding diva in your life, without going bankrupt? Well, I have. But that diva currently in my life is none other than “Her Highness, The Microbes”. 

Yes, the very same microbes, we use to make biofuels. In return, they demand a princess treatment. They have to be inside this shiny, sterilized, air-conditioned tanks where they are constantly sipping a cocktail of sugar and other nutrients which has to be perfectly pH balanced. Very high maintenance, right? But also justifiable for the incredible work they do. However, the challenge is, some demands are trickier than others such as supplying oxygen. 

Oxygen is about 300 thousand times less soluble in water than the sugar is. So, to keep those microbes chilling in the sugar water breathing, we have to pump in a lot of oxygen and stir it vigorously to keep it circulating. This pumping and stirring consumes a lot of electricity, making the oxygenation step the top contributor to a facility’s total electricity cost.

Now, here’s the problem: too little oxygen, microbes choke. Too much of it, we go broke. Therefore, we need that “just right” oxygen zone where microbes thrive and our wallets survive.

That’s the sweet spot I’m trying to find. But for that, I can’t just play trial-and-error in a real facility. It’s way too costly & impractical. So instead, I built a computer model which I also sweetly like to call a BEE model- not a honeybee but a very buzzworthy blend of B-Biology, E-Engineering, and E-Economics. This holistic model integrates microbial growth, biofuel yield, oxygen flow, and all the associated cost factors, allowing me to simulate them simultaneously across countless scenarios on a computer. 

Using this model, I tested lower oxygen flow. As expected, the microbial drama queens threw a tantrum, and the yield dipped slightly. But after tweaking process setup and equipment design, I found a scenario that cuts oxygen cost and increases net profit by ~15%, even with the lower yield. That’s the sweet spot — not maximum oxygen, not maximum yield, but maximum value! 

Oxygen may be invisible, but its cost is not. My work makes it visible and optimal by bridging biology, engineering, and economics. Because, at the end of the day, I don’t only focus on pampering those microbial divas, I also care about scaling up a biofuel science without going bankrupt.

What do plastics, natural flavors, and jet fuel have in common? All of them can be manufactured with the help of bacteria! In nature, bacteria have evolved to eat whatever chemical compounds were available to them to make what they need to survive. Bacteria naturally produce thousands of unique and interesting chemical compounds to serve their own needs. One such compound is an energy-dense liquid, ideal for use as fuel. Another is one component for a stable but biodegradable plastic. To achieve our economic and environmental goals, we at the Joint Bioenergy Institute want these bacteria to produce as much of these desired products as possible, and we want them to do so while we feed them the cheapest carbon available, especially if that carbon would otherwise be waste, like peanut shells or sawdust. We want as much of our feed mass as possible to become product mass. There lies the core problem we have to overcome: bacteria don’t want to produce buckets of fuel. They want to produce more bacteria.

The biological patterns that direct where the mass from food goes are wired into the bacteria’s DNA. Fortunately, in 2025, we don’t have to accept that as a fact of life. We can reach into those cells’ genomes and make changes. We, as bioengineers, have genetic tools that can make individual genes more or less active, can remove genes or add entirely new ones from other species. With all of these options, across thousands of genes, we have so many options that we could never test all of them experimentally. That is where AI tools, and my own research, come in. AI models are exceptionally good at finding subtle patterns in data, patterns that our human brains would not notice or often even comprehend. By taking advantage of these tools along with experimental data, I can figure out what genetic edits are worth our attention, and what combinations of edits are likely to give us more product for the same amount of feed. The best part is, when we test the combinations the AI gives us, I get more data to inform the AI and make its predictions even better, giving us a positive feedback loop toward exactly what we want: as much product as is biologically possible.

Thanks to this process of AI-driven engineering, we are able to drive costs down and deliver biomanufactured products at ever cheaper price points. So, next time you see a biodegradable plastic, buy a palm oil-free lotion, or board a plane, you just might have AI and bacteria to thank for it.

Nature is brutal. Beneath our feet lies one of the fiercest arenas on the planet--The soil. Soil is complex, full of interactions among plants, animals, fungi, bacteria,......, and even viruses.

Phages are viruses that infect bacteria. Phages can kill their host to make more phages, or sneak in their DNA and stay hidden. But, bacteria are not just innocent victims. Some of these phage genes integrated into bacteria evolved into bacterial weapons that resemble phage tails. Bacteria use these weapons to stab their rivals and kill them. They can even acquire the DNA of the victim, gaining new functions. These invisible battles shape our world. Soil bacteria can make plants healthy or miserable; forests grow strong or struggle. Two huge challenges are that: we don’t know how widespread these weapons are—and how these competitions affect plant health.

My research is to understand how phages and these tail-like weapons shape plant-bacterial interactions. Team science at our lab providing massive experimental data, as well as the power of machine learning allows me to solve this mystery. I am building a machine learning tool that scans bacterial genomes and hunts down phages and these hidden weapons.

And the discoveries are striking. I have found hundreds of these weapons, especially in bacteria that help plants. We even caught plant-beneficial bacteria stealing plant colonization genes from their victims, turning themselves into better plant colonizers. This shows how these weapons allow more beneficial bacteria to grow, improving the health of our food or biofuel crops.

So next time you look at the forest on this hillside, remember: beneath the “peaceful” greenery lies one of the most brutal battlefields on the planet. Head or no head, phages and their legacy weapons are rewriting the rules of survival for bacteria, for plants, and ultimately, for us.