When I first started studying biology, I thought the discipline was mostly about memorizing facts and figures about different organisms and their characteristics. In high school, I was more interested in physics and chemistry, which seemed to involve learning general principles and laws that could be applied to many problems. In other words, I appreciated the elegance and efficiency of the principles of physics, math, and chemistry.
However, during my first year of undergrad, I took a class titled “Bioinformatics Programming,” which was pitched to me as an introductory programming class in Python mixed with biological application. Intrigued, I took a risk with the class. Early on, we learned about the problem of DNA sequence alignment. Our professor explained how scientists use mathematical algorithms to compare different DNA sequences and understand how they evolved over time. For example, the fact that humans and chimpanzees share a substantial portion of their DNA is based on the principles of comparing the billions of DNA bases in our genomes and how they differ from other species because of accumulated insertions, deletions, substitutions, and rearrangements in the DNA sequence over time.
I found this concept fascinating, and once I learned more about how these comparison algorithms worked, I was amazed at how simple, elegant principles could reveal so much about the relationships between different species. I was finding in biology the thing I loved most about studying math, physics and chemistry.
When I took a deeper dive into the topic, I found that the problem of sequence alignment was just the tip of the iceberg. I learned that many of the most exciting discoveries in biology are made by applying mathematical and computational methods to understand the underlying principles of life.
Today, the frontier of innovation in artificial intelligence and machine learning is turning to problems in biology, specifically because the application of computational and mathematical methods to the discipline has been so successful in the past two decades.
Since its founding, the company 23andMe has sold more than 10 million DNA testing kits. After receiving this material, companies like 23andMe implement similar DNA comparison algorithms to assess genetic ancestry and disease risk. There are now more than 100 trillion DNA sequences that have been deposited into GenBank, the US National Institutes of Health public repository for DNA sequencing data of all organisms. Companies such as Google-owned DeepMind and others have used this DNA sequencing data and machine learning algorithms to predict the structures of nearly every single protein encoded by all known DNA sequences.
Biology and the core principles underlying genetic variation and evolution are increasingly relevant to our daily lives. My decision to become a biologist really began by recognizing as a first year in college how powerful and intriguing the union of computational and mathematical principles with the study of organisms could be. In truth, these principles also helped me learn the subject because it convinced me that I didn’t have to have a photographic memory to gain a deep understanding—I merely needed to learn how to apply broadly-applicable principles to diverse problems.
Comments