Article
Summary
According to Papin, as many as 10 million people can become fatalities to incurable drug-resistant infections sometime in our lifetimes. One may think the cure lies pureley in lab work, where new drugs, antibiotics, and treatments must be innovated. However, in Papin’s view, data science is the answer. To him, predictions about cell changes and reactions must be made to answer the core questions of the problem such as:
- Which genes are activated in certain environments?
- Which genes are not activated?
- Why are certain genes activated in certain situations while others are not?
To do this, Papin and his researchers at the University of Virginia plug in more and more information from the lab into developed computer simulations and algorithms to create the predictions. Then, they sort through the predictions, checking them against published literature, and return to the lab to find why they were correct or incorrect. After this, they go back to the algorithms and improve them. Papin states that “[they] can generate thousands and millions of different data points and see how all of those pieces connect to each other”. In doing so, they are identifying cells best equipped to respond to a particular drug, tracking bacterial evolution and adaptation over the course of an infection, and working to predict how antibiotic resistance develops or changes in the human body.
Per the article, 700,000 people die each year due to these incurable drug-resistant infections. By combining data science with traditional lab work, Jason Papin is working every single day at the University of Virginia to aid in decreasing that statistic.
My Thoughts
After reading this article, it truly hit me how much of a general use technology data science is. Although I knew there were applications in the biomedical engineering field (my undergraduate major), I didn’t know how far they extended. This is BIG research, and it is extremely exciting to me how it is developing at the same institution that I attend everyday. Dr. Papin is also my faculty advisor, and I could not be prouder.
Nevertheless, if he is able to succeed in finding or at the very least advancing information known about cures for these drug-resistant infections taking the lives of many around the world, the benefits would be insurmountable. It is just amazing and unexpected to me the role that data science plays in predicting the progression of disease and lab workers being able to better treatments in response to its models and algorithms.
Possible Coding Examples
The following are possible methods biomedical data scientists can use information about bacteria to model pathogenic and viral disease progression:
bacterial_distance = 23 #micrometers
time = 6 #seconds
bacterial_speed = bacterial_distance * time #micrometers per second. Can be useful to model pathogenic infection progression
str(bacterial_speed) ## num 138bacterial_radius = 0.03 #micrometers
bacterial_surface_area = 4*pi*(bacterial_radius)^2
str(bacterial_surface_area) #micrometers squared. Can be useful for bacterial progression## num 0.0113Above example used the following equation:
\[ Surface Area = (4)*\pi*r^2 \]
bacterial_volume = (4/3)*(pi)*(bacterial_radius)^3
str(bacterial_volume) #micrometers cubed. Useful to target bacteria with drugs and finding how resistant they are to existing drugs. ## num 0.000113Above example used the following equation:
\[ Volume = \frac{4}{3}*\pi*r^3 \]
