Welcome to the BiosBeat communication tool for the Department of Biostatistics at the UNC Gillings School of Global Public Health! Here you will find a collection of the latest department news, special features, dates to save, and much more. So, read on, enjoy, and be sure to regularly check back for updates!
Message from our Biostatistics Leadership – COVID19
We hope you are staying safe and healthy during this time. We encourage you to continue taking all the preventative measures recommended by UNC and the CDC.
Please visit the School of Public Health’s official Coronavirus Information Portal for the most recent information and updates: https://sph.unc.edu/global-health/2019-coronavirus-info-portal/
HIGHLIGHTS OF THIS ISSUE:
- PROFESSOR LAVANGE APPOINTED AS BIOS DEPARTMENT CHAIR
- FACULTY SHARE LESSONS FROM A VIRTUAL FALL SEMESTER
- BIOS STUDENT RECEIVES PRESTIGIOUS AWARD FROM NIH
- MACHINE LEARNING HELPS COMBAT CANCER
- BIOS PROFESSORS RESEARCH PUBLISHED IN BIOMETRICS
FACULTY REFLECTS ON THE POSITIVES OF A VIRTUAL SEMESTER
The Fall 2020 semester was anything but ordinary. With last semester consisting of entirely remote learning, it required an adjustment for both students and faculty. While it may not have been anyone’s idea of a perfect semester, there were certainly plenty of positives virtual environments have brought to biostatistics classrooms. Many lessons from teaching through Zoom can even be applied once we return to a normal classroom setting.
In some ways, online learning meant getting the opportunity to interact with students in new and interesting ways. Professor Jane Monaco shared her highlight of the remote semester, saying “one of the best things about last semester was asking the online students ‘share one positive thing – big or small.’ At a time when it was easy to focus on the many negative aspects of the pandemic, I enjoyed hearing about small and big victories! Whether it was “Starbucks is now serving the Pumpkin Spice Latte” or “my mom has recovered from Covid!” – it was satisfying to know about good things happening with the students.” She also mentioned enjoying getting to know her students in a different way than would be possible in a traditional classroom, stating “with remote learning, I got to know many students’ pets who joined the class and even know about students’ interests from wall posters, trophies and pictures in their bedrooms.”
Professor Todd Schwartz shared his perspective on the shift, mentioning the benefit of having recorded lectures for students to later refer to. “Recordings of our Zoom-based classes could be helpful to students, as they can review parts of the class session that may have been covered too quickly or that were unclear in their understanding of certain topics. It is also helpful in the case where a student needs to miss a class session for some reason. It seems important to realize that having such recordings and making them available to students can be relevant even when classes return to being in-person,” Professor Schwartz said.
Online courses also allow for increased accessibility for students, independent of their physical location. Students from different parts of the country, and even around the globe, could attend class meetings and office hours in a way never before made possible. Professor Michael Hudgens suggested he may continue the occasional use of Zoom lectures when in-person learning returns, for cases of snow days or work-related travel. Similarly, Professors David Couper and Xianming Tan mentioned plans to continue holding virtual office hours on Zoom, since the platform allows more students the ability to attend.
Research collaboration has also become easier than ever, allowing faculty to work together with those not in the Chapel Hill area. As Professor Schwartz said, “The remote environment helped me to realize that certain long-distance collaborations could be more feasible and sustainable than I would have thought pre-pandemic.” This has allowed research to continue to thrive at the Department of Biostatistics, despite the given challenges.
Looking forward upon the Spring semester, most courses will continue to be online only. While this presents some challenges, it also provides the opportunity for innovation and discovery, abilities Gillings School students and faculty excel at.
SPOTLIGHT ON FACULTY
Name: Michael R. Kosorok
Position: W.R. Kenan, Jr. Distinguished Professor of Biostatistics and Professor of Statistics and Operations Research. I was also chair of Biostatistics from May 1, 2006 – June 30, 2016, and again from July 1, 2017 – June 30, 2020.
Time at the Gillings School: Since May 1, 2006.
What I do (and why I love it): I work in machine learning and precision medicine mostly, and I enjoy it because it is extremely interesting and can lead to real improvements in human health. I also enjoy being a professor: I like to teach, help students progress, and do interesting and important research. My hobby (after hours) is music composition (modern classical).
First job or Internship I had was: My first job after my PhD (received in December, 1991), I worked as a post-doctoral fellow at the University of Washington where I studied infectious diseases.
BIOS STUDENT RECEIVES FELLOWSHIP AWARD FROM NIH
Tarek Zikry, a doctoral student currently pursuing a PhD in Biostatistics, recently received the prestigious F31 Fellowship Award from the National Institute of Health (NIH). The NIH F31 Fellowship, known as the Ruth Kirschstein National Research Service Award, was created to enable promising predoctoral students to develop into successful, independent research scientists, allowing them the opportunity to obtain mentored research training while conducting dissertation research.
In 2019, Tarek graduated with a BSPH in Biostatistics from the Gillings School of Global Public Health. “I’ve had the fortune of being in this department for many years from my time as an undergrad. I went to UNC for my undergraduate degree and started in the biostatistics BSPH my junior year. I didn’t come into UNC planning to major in biostatistics and bounced around a few majors, but I became passionate about biostatistics for the flexibility of being able to work on a plethora of topics, all of which seek to improve public health in their own way,” Tarek said.
He became involved in research within the Genetics department in May of 2017, which granted him the opportunity to focus on topics he was passionate about, as well as find mentors to learn from. Tarek shares, “I started undergraduate research with Dr. Jeremy Purvis in the Genetics department, which later morphed into a senior thesis under Dr. Purvis and Dr. Michael Kosorok in the Biostatistics department, and those ideas have segued into the forming of my doctoral research. I was excited to continue my education in this department not only because of the exceptional research opportunities, but also because of the amazing faculty to learn from in and out of the classroom.”
Tarek’s project, entitled ‘Deep Learning Models to Predict Primitive Streak Formation in Human Development’, will be a continuation of his work in the Purvis Lab, where they develop both experimental and computational models of human embryonic stem cells (hESCs) to understand the biological pathways and mechanisms governing cellular fates.
“For this project, I am aiming to use deep learning models to first track these cells as they differentiate for data on the single-cell level, and then predict and understand how hESCs ‘decide’ on particular fates, or developmental trajectories. These hESCs will be driven to differentiate towards primitive streak, hence modeling their fate ‘decisions’ will be crucial in advancing our understanding of human fetal development with the future goal of reducing congenital birth defects,” Tarek said. “In my undergraduate thesis I started learning more about tracking cells using machine learning algorithms and quickly became enamored with applying these methods to health-motivated data. This project not only offers me the opportunity to learn more about human cellular biology, but establishes a novel statistical and computational challenge I am very thrilled to meet.”
The F31 Fellowship award will be pivotal in allowing Tarek to continue enhancing his work. As he says, “The NIH F31 will allow me to continue to learn from and facilitate my research with Dr. Purvis and Dr. Kosorok as my supervising mentors. Furthermore, with the grant’s support, I will work with leading collaborators at UNC to develop my research skills. With Dr. Adriana Beltran, the director of the UNC Human Pluripotent Stem Cell Core, I will seek to understand the hESC developmental models I am analyzing in vitro. Dr. Michael Hudgens in the biostatistics department will also collaborate on this project. Under his guidance, I will apply and develop methods to causally infer the motivating factors behind hESC decisions. I am thankful for the support of my mentors, collaborators, and labmates both in developing this proposal and in my ongoing and future research.”
PROFESSOR LAVANGE APPOINTED AS BIOS CHAIR
Professor Lisa Lavange has been officially appointed to be the chair of the Department of Biostatistics. She takes on this role after serving as intern co-chair, along with Professor Jianwen Cai, for several months. Professor Lavange brings to the role a wide range of experience both in public and private sectors, including university, non-profit consulting, pharmaceutical, contract research organization, and government work. She has held leadership roles at RTI International, Quintiles, Inspire Pharmaceuticals, and the U.S. Food and Drug Administration (FDA). She was a professor of biostatistics and director of the Collaborative Studies Coordinating Center (CSCC) at the Gillings School from 2005 to 2011. She rejoined the Department of Biostatistics in 2018 as professor, associate chair, and director of the CSCC.
“Dr. LaVange is the ideal person to be chair now,” said Barbara K. Rimer, DrPH, Alumni Distinguished Professor and dean of the Gillings School. “She is extremely well-trained, with her undergraduate degree in mathematics from UNC-Chapel Hill, her master’s degree in mathematics from the University of Massachusetts and her doctoral degree in biostatistics from the Gillings School. She is committed deeply to the department and has demonstrated that in many ways.”
Read the full story about Professor Lavange’s appointment on the School website.
PROFESSOR HUDGENS ANNOUNCED AS BIOS ASSOCIATE CHAIR
Professor Michael Hudgens has been appointed to serve as Associate Chair of the Department of Biostatistics. In this role, Professor Hudgens will assist the chair with a broad range of administrative duties to help ensure the biostatistics department remains well-run. He will assist with managing committee appointments, overseeing the faculty mentoring program, and coordinating promotions, re-appointments and post-tenure reviews, to name just a few. He will also partner with the chair on strategic and financial planning for the department.
“We are facing many challenges at this moment in the life of the University and are fortunate to have Michael’s energy, experience, and positive attitude to help us stay on course as a top-ranked biostatistics department.” said our chair, Professor Lisa LaVange, “I am excited about working with Michael and look forward to sharing more on our initiatives in the coming months.”
HOW MACHINE LEARNING CAN HELP FIND OPTIMAL CANCER TREATMENT DECISIONS
A research team led by Biostatistics Professors Michael Kosorok and Naim Rashid have developed a computational framework to general evidence-based optimal cancer treatment decisions informed by a patient’s genomic biomarkers. These findings, which use machine learning to aid in the development of precision cancer treatments, are published in the Journal of the American Statistical Association.
Professor Rashid said the goal of the research was to develop and train new machine-learning methods to predict optimal treatments for cancer patients. This is based on big data from large scale preclinical screens in patient-driven xenografts, or PDXs.
Created by implanting part of a patient’s tumor into immunocompromised mice, a PDX line produces multiple models of the same tumor. This allows researchers to more efficiently test and evaluate how an individual patient’s tumor responds to different drugs. Professor Rashid and his colleagues analyzed data from a large PDX screen spanning five cancers, 1000 PDX lines, and 38 unique treatments evaluated.
“PDX studies represent an untapped resource to exploit for estimating optimal individualized treatment rules, which can be used to recommend best potential therapy in new patients,” Professor Rashid said. “This new machine learning method was tailored to address several unique aspects of PDX data, such as evaluating responses pertaining to a large number of treatments applied to the same tumor, and to search for predictive biomarkers from a large set of genomic features in this framework.”
Read the full story on the School’s website.
BIOS PROFESSORS PUBLISHED IN BIOMETRICS
Professors Bonnie Shook-Sa and Michael Hudgens recently had a paper published in Biometrics. Their paper, entitled ‘Power and Sample Size for Observational Studies of Point Exposure Effects’, is surrounding the consideration of the utility of a study’s design effect to quantify the effect of weighting on the precision of causal estimates.
Inverse Probability of Treatment Weights, IPTWs, are commonly used in studies to control for confounding when estimating causal effects of point exposures from observational data. When preparing for a study that uses IPTWs, it can be difficult to determine the required sample size for a given level of statistical power given the effect weighting may have on the variance of the estimated causal means. Professors Shook-Sa and Hudgens paper considers the utility of the design effect in order to quantify the effect of weighting on the precision of casual estimates.
To do this, they estimated the design effect by locating the ratio of the variance of the casual mean estimator divided by the variance of a naïve estimated, provided that no confounding had been present and weights were not needed. This calculation derives an approximation of the design effect that is outcome invariant and can be estimated during the study design phase. Their paper continues on to include simulations demonstrating the accuracy of the design effect approximation, and discusses practical considerations.
Read the full abstract of the paper from Biometrics.
PROFESSOR SCHWARTZ PROMOTED
Todd Schwartz has recently been promoted to Research Professor of Biostatistics. Professor Schwartz is a UNC Biostatistics graduate and has been with the faculty as an instructor since 2001. He has received several awards, including the prestigious 2020 Bernard G. Greenberg Alumni Endowment Award, which recognizes an outstanding full-time faculty of the Gillings School of Global Public Health for excellence in teaching, research and service. Congratulations, Professor Schwartz!
IMPORTANT STUDENT DATE REMINDERS
- February 26 – Minority Health Conference
If you have news or a story idea you feel would fit BiosBeat, please submit them to Jeff Oberhaus.
PLEASE NOTE: Given the recent events regarding COVID-19, the annual BiosRhythms is on hold. Issue 31 will not be sent until further notice