It’s crazy to me that it’s already Thursday; this week has seemed to fly by! Today started with a demo on how to set up a professional website on WordPress. I’m not going to link my site yet because I’ve barely started on it, but stay tuned for updates. The process of designing a website is daunting, but I really just need to jump in and get started without worrying about how it looks in the beginning. Easier said than done, I know.

Once that session concluded, we learned a bit about machine learning from Dr. Natarajan. He is a very good presenter and I found myself both laughing and learning throughout his presentation, which lasted roughly 2 hours. Near the end, he gave a peek into what REU students would be doing on his project.

There is data online from researchers who have tracked a large sample of people’s health and lifestyle since 1985. Dr. Natarajan and Nandini Ramanan, the graduate student mentor, have been using machine learning on this data to accurately predict how likely a person is to have a heart attack. Interestingly, if the machine learning algorithm only had access to one year’s data and had to predict the likelihood of that person having a heart attack at a later age, the accuracy was higher using data from the first years of the study. Dr. Natarajan said that exhibiting unhealthy behaviors such as smoking at a young age was a major factor in the likelihood of an individual experiencing heart issues.

After his presentation, I asked Dr. Natarajan if he had read “What’s Killing Poor White Women?”, an article that his recent work brought to mind. It is eye-opening journalism detailing how the life expectancy of uneducated rural white women has gone down by 5 years in the last generation, a worrying abnormality for a medically advanced country such as the United States. The article was close to home for me – literally. The family that the author focused on was distantly related to me and lived in the same town I grew up in.

Though he hadn’t read the article, Dr. Natarajan asked me a related question: why are people who live on the highway more likely to have a heart attack than those who live elsewhere? I guessed that it could have to do with the noise pollution in the area, which is shown to raise background stress levels. Rather, he informed me, it was due to the lack of accessibility of healthy food in those areas; usually only fast-food restaurants were close to their homes.

I had already planned on putting down Nandini and Dr. Natarajan‘s project as my first choice to work on this summer; the presentation this morning just sealed the deal. I was absolutely thrilled when, this evening, I got an email saying that I would be on their project alongside Robert Long. I think that this summer experience is going to be a lot of work, but also a lot of fun. Furthermore, it will let me know if I enjoy the machine-learning side of bioinformatics, which is my current career aspiration.

Since I have only dipped by toes into the sea of machine learning, I have a lot to learn in a relatively short time span. As I finished my CITI training this evening, I took breaks between modules with videos from this this Google Developers Machine Learning Playlist about decision trees, which helped refresh me on the subject.

Unknown Animal of the Day: Pacu Fish

Pacu, a species of fish in South America, have disturbingly human-like teeth. Like us, they are omnivores. This diet has led them to evolve flat teeth unlike their cousin, the Piranha.

Conservation Status: Though over-fished in it’s native regions, this fish has become an invasive species on other continents.