When I first began speaking with the CEO of Pluton Biosciences, the company that I would eventually intern with, we started by defining the problems they try to solve. They develop revolutionary microbial products that aim to address some of the most relevant of society’s problems, such as lowering greenhouse gas build-up, increasing crop yield, and decreasing the spread of malaria-spreading insects. We decided to focus on a project to produce a novel carbon sequestration method through specialized, bacteria-enhancing soil amendments.

A “soil amendment” is an additive used to enhance the physical properties of soil; fertilizers are a common soil amendment used to replenish nitrogen lost to the previous season’s crops. Pluton aims to do more than replenish spent nutrients, however. Pluton envisions a future where commercial growers can spray bacteria onto their fields at planting and again at harvest, which will store carbon and nitrogen throughout the year. Carbon stored this way is removed from the atmosphere, where it would collect as the greenhouse gas carbon dioxide, reducing the spread of global warming. On the other hand, stored nitrogen means less use of traditional fertilizers, a win-win for growers, the environment, and Pluton.

However, before this product can be brought to market, the efficacy of a bacterial soil amendment must first be tested. And thus, questions of importance become: Can we detect the presence of this bacteria in fields after adding it? Can we detect the genes which are most important to nitrogen fixation? Will this differ between fields in different states, in various areas of the United States? Preliminary data suggested that this bacterium can be found in some but not all soil samples from both fields. I was tasked with building a computational pipeline that would look for bacterial DNA and identify the genetic composition, specifically looking for genes that give the bacterium of interest the ability to take nitrogen out of the air and “fix” it into the soil. While building this pipeline, I gained experience using cutting-edge computational tools in the Amazon Web Services (AWS) cloud computing environment.

It immediately became clear during my internship experience how sheltered I was as a graduate student at WashU, with access to a private computer server run by IT professionals. Every piece of software I could ever need was already downloaded, installed, and loaded with a single command. Not so when using AWS! Instead, I was in charge of creating my own virtual machines, essentially specialized minicomputers, which could then be loaded onto the cloud and run virtually. I oversaw ensuring that each machine was working correctly, communicating with the AWS service that stored the genetic data I was working with, as well as shutting off correctly; this last part was especially important as every second of computing time is billed. I learned how a linux operating system works, set up a virtual private gateway on AWS, and a host of other extremely technical and specialized tasks. Most importantly, however, I learned how to deal with the frustration of feeling like you have been hitting your head on a wall for 6 hours trying to find a bug in your pipeline but feel no closer to fixing the problem. I learned data science best practices for storing data and writing code. I also gained valuable experience working with a mentor in a small biotech company, where everyone wears many hats.

By the end of the internship, I had accomplished no actual computational biology; instead, I had gained experience working in an industry setting and developed my own fully functioning computational pipeline. It was, without a doubt, one of the most valuable experiences of my graduate career.

Winston Anthony is a Ph.D. student in Biomed at Washington University who wrote this blog as part of his Pivot 314 Fellowship.