At first glance explaining my wife’s job is easy. She is a doctor, with a hard earned MD that required sleepless nights, a few traumatic rotations in residency, and a grueling fellowship period which felt like indentured servitude. It was probably hard on her as well. Once she earned her stripes she went to work in a lab for several years at Stanford, which enabled her to pursue her scientific interests, and medical training at the same time. My wife, Dr. Caitlin Pepperell, who I refer to in the rest of the post by her professional title, Dr. Pepperell, now runs a research lab at UW-Madison and practices clinical medicine through the University Hospital. Sometimes, when I’m explaining what she works on, things can get complicated if my conversation partner is not satisfied with an oversimplified answer that goes something like, “She works on infectious diseases, mostly tuberculosis and other scary things. She’s the doctor you never, ever, [long pause for effect] ever, want to see.”
Satisfying detailed inquiries into what my wife does, or more specifically, what kind of research she produces, is challenging. The problem mostly stems from the fact that she’s involved in a new style of science. He research world is less Frankenstein-style-test tubes and clean room labs, and more Google server farm big data computation. Her work requires her to be part sleuth as she seeks out scientists across the globe who possess valuable, unused data sets. She also needs to be part start up CTO as she pulls together groups of scientists, doctors, mathematicians, evolutionary theorists, computational biologists, and other assorted engineering types, which you would never think are working on detailed analysis of diseases like tuberculosis. A sobering moment for me was when she whizzed through a python course, and then hit me up to help get her lab squared away on Amazon S3. She had more engineering acumen than most of the product managers I knew at Yahoo! and AOL.
Pepperell’s science is quantitative and data modeling intensive, and as a result, tends to be cheaper than labs which need to invest in specialized equipment and staff. It’s been enabled by the dropping price of sequencing technology driven by companies like Silicon Valley based Pacific Biosciences. Researchers are now able to sequence a genome for under $1000, in less than a day. The falling cost of sequencing has meant a large influx of data, and only a handful of scientists, like Pepperell, are taking advantage of the data glut. Should Wall St. hit another bump in the road, the analytics professionals that have found themselves in finance may find a few use for their skills in the world of medical research.
Like most scientists, or academics for that matter, it’s a cause of celebration when a paper comes out. We send it around to friends and family, who politely pull it up on their screen, and then realize that the work will is all but impenetrable. I confess that I spent a night trying to read her latest work, ”The Role of Selection in Shaping Diversity of Natural M. tuberculosis Populations” (PLOS Pathogens - http://ow.ly/oIjgY), and quickly resorted to scanning the paper for sentences, and eventually words, which might make sense to me. Thus did go my introduction to the world of bioinformatics, genomics, and evolutionary biology. My inability to understand even the abstract of my wife’s work, an experience which was echoed by most of our friends and family, is what led me to tackle a layman’s explanation in this post and, a related podcast.
When I eavesdrop on conversations people are having with my wife at dinner parties or soccer games, I frequently hear our friends say, “I didn’t know tuberculosis was still a problem!” Dr. Pepperell then patiently walks through the basics, which usually has the effect of freezing our friends in the place while they wait for her to tell them that everything will be okay.
Tuberculosis (TB) infects 1/3 of the world’s population, and is only second to HIV as an infectious killer. To cap it off Multi-Drug Resistant strains of TB have emerged, commonly referred to as MDR-TB, which are resistant to two of the most effective treatments, isoniazid and rifampin. The good news is that only 10, 528 cases of TB were reported in the US in 2011. According to the CDC, that is the lowest number of reported cases since they started tracking in 1953. The bad news is that world wide 1.4 million people, a population size equal to Phoenix or San Diego, still died of TB. The CDC, and WHO websites, where all the preceding data originates, read like a mix of cautious optimism (infection rates are decreasing), and dire warnings, (funding constraints on TB “care and control”, and the steady rise of MDR-TB.)
Despite medical advances, and encouraging stats in the U.S., TB is still an effective and prolific killer world wide. Determining why it has been so successful for the last several hundred years is where Pepperell and her colleagues come in as they map out the disease’s evolutionary path. Unlike the deadly diseases we hear about in the mainstream entertainment media like “Contagion” or “The Andromeda Strain”, TB does not appear to evolve quickly. The DNA sequences of different TB strains are similar to each other. ”The evolutionary constraints on TB are strong,” Dr. Pepperell explained as I tried to sort out how her work fits into the bigger scientific picture of TB care and control. Mutations in TB are not tolerated well, and yet the biggest problem facing the bodies charged with controlling TB, are strains of the disease like MDR-TB, which are seemingly immune to common treatment protocols. Pepperell, and her collaborators, hope to refocus the problem by mapping out an accurate evolution model which can be used by other researchers. “Once you understand how TB evolves,” said Pepperell, “You can model it and experiment with different strategies on how to eradicate it.”
Studying the Evolution of the Second Deadliest Disease
To create their model the Pepperell lab, and their collaborators, analyzed 63 genomes of TB from across the globe. Their key finding from the analysis is that TB has evolved at its own rate, ignoring its human host’s evolutionary specialization. With the aid of computational detective work, the team of doctors, Ph.D’s, computer scientists, and mathematicians, created algorithms which rolled back how quickly TB has evolved. Their mathematical take on the history of TB’s evolution also offered up another important finding: there is a clear and direct relationship between world wide human population growth and the growth of TB. Like a lot of good science the end hypothesis seems simple and logical, but previous theories have contended that TB evolved at the same rate as humans, and the models placed the origination of TB as far back in time as the original migrations of humans out of Africa.
The detailed look at how multiples strains of TB evolved across the globe has provided a better picture on how the pathogen continues to be so successful. The similarity between the different threads of TB around the world has perplexed researchers until now. “Strains could appear similar to each other because they are evolving slowly, or it could be that strains are evolving quickly and most changes are “kicked out” of the population,” said Pepperell about the final comparison. “We showed that the latter was true, which resolves the paradox of most strains looking identical, yet seeming able to mutate rapidly and acquire drug resistance.”
The data produced by Pepperell and her collaborators, many of whom are a “who’s who” in the world of computational biology, puts these old theories to rest and starts a new vein of research to find a patient X, who can offer up an ancient TB sample to help trace the disease back to its origination point, both in historically, and geographically. With an eye on mapping how TB evolves, Dr. Pepperell aims at laying out the most basic understanding of one of the world’s deadliest diseases.
Below is an interview I conducted with the good doctor before putting this article together. It covers most of the above topics as well as a broader look at TB and the details of her research.
For additional info about Dr. Pepperell, The Pepperell Lab, and their collaborators, check the following resources
- Twitter: @Pepperell_Lab
- Pepperell Lab Website
- Pub Med Search on Dr. Caitlin Pepperell
- UW Madison Profile
Additional Links and Data Sources
- WHO Executive TB Report Summary : http://www.who.int/tb/publications/global_report/gtbr12_executivesummary.pdf
- CDC TB Trends Fact Sheet: http://www.cdc.gov/tb/publications/factsheets/statistics/TBTrends.htm
- Cost of sequencing Genomes - http://singularityhub.com/2011/03/05/costs-of-dna-sequencing-falling-fast-look-at-these-graphs/
- The Role of Selection in Shaping Diversity of Natural M. tuberculosis Populations (PLOS Pathogens - http://ow.ly/oIjgY)