UF Statistics Student Seminar Schedule
Seminars are held on Tuesdays from 4:00 p.m. - 5:00 p.m. in Griffin-Floyd 100.
Refreshments will be provided!
|Feb 23 (Mon)||Samara Strauber||
Internship opportunities at DunnHumby
|Mar 3||Anthony Homer||
Internship opportunities at Sanofi Pasteur
|Sanofi Pasteur website|
|Mar 17||Tania Quesada|
|Mar 24||Vikneswaran Gopal|
|Mar 31||Vikneswaran Gopal|
|Apr 14||Doug Sparks|
|Apr 21||Dr. George Casella||
|Association Genetics the resistance of pitch canker disease in loblolly pine (Pinus taeda L.)|
Pitch canker is an economically important disease incited by the necrotrophic fungus Fusarium circinatum. It affects many pine species, causing resinous lesions, seedling mortality, and crown dieback.
Screening for resistance to this disease revealed that resistance is heritable and appears to be a complex trait. To explore the molecular genetic architecture of pitch canker resistance, we used an association genetic approach. Associations between phenotypes and genetic markers in populations that are composed of many unrelated individuals can reveal genetic loci that are causally related to the phenotype. As a first step toward association mapping of pitch canker resistance in loblolly pine,
498 largely unrelated genotypes (association population) were screened. A randomized incomplete block design was implemented, with four replicates (rooted cuttings) of each clone. Lesion length measurements were taken 4, 8, and 12 weeks after inoculation with F. circinatum microconidia.
Genotypes were ranked based on their resistance to pitch canker according to clonal estimates using best linear unbiased predictions (BLUP). We validated our population-level classification of genotypes by re-inoculating the 50 most resistant and susceptible clones; results confirmed the significant differences among resistant and susceptible
tails (p < 0.0001). Clonal repeatability (an estimate of trait
penetrance) ranged from 0.20 and 0.30, indicating genetic and environmental influence on expression of disease resistance. Association analysis of 3,938 SNP markers resulted in 24 potentially significant genes associated with pitch canker resistance, revealing potentially novel genes as well as previously identified genes for disease resistance.
|Linux and Efficiency|
There must be some reason as to why people who are comfortable with the Linux Operating System swear that they will use no other. In my experience, I have found that those who speak of Linux in a derisory manner are those who are unwilling to dive in and learn of its boundless capabilities. Rest assured, though, that my aim is not to proselytize, but merely to highlight some features that show the flexibility of Linux, and to introduce some of the built-in tools that can greatly increase one's productivity in the long run. I will talk about some common but powerful Linux commands, demonstrate connecting from home in order to run background jobs, and touch on the text editor "vi". The slides can be found at http://www.stat.ufl.edu/~viknesh/R_programming/seminar.html . You are welcome to peruse them and then decide to attend (or not).
The topics were chosen based on common questions that students from our department approached me with. In addition, if there is a burning question that you have, that has thus far prevented you from embracing Linux, feel free to email me and I will try to answer it during the talk. Finally, be forewarned that there will be no 'statistics' in this talk - there will not be a single greek symbol on any of my slides. In fact I could have made them in PowerPoint, but alas, that would have meant using Windoze....
|R programming tips|
In this talk I will demonstrate how to debug, profile and optimize R code using an extended example. After that I will attempt to explain the concept of lexical scoping in R. I believe knowing about this can lead to neater code, and can also prevent long frustrating hours spent in debugging. Time permitting, I will talk about Sweave/Stangle and a couple other random topics in R: using file connections and writing output to a console. The slides, example code and relevant links can be found at http://www.stat.ufl.edu/~viknesh/R_programming/seminar.html.
|Objective Bayesian Inference: Motivations and Methods|
Recent decades have seen a variety of new developments in the broad category of objective Bayesian inference. The benefits of using such methods go far beyond the simple fact that we can label our results as "objective." In this talk, I will discuss these benefits, give a brief history of the topic, and explain some of the issues that have motivated research in the field. More specifically, I hope to explain the reasons for the existence of the bewildering assortment of objective priors in the literature.