Skip to content

Instantly share code, notes, and snippets.

@willtownes
Last active August 29, 2015 14:06
Show Gist options
  • Save willtownes/625028178ae61d3bb863 to your computer and use it in GitHub Desktop.
Save willtownes/625028178ae61d3bb863 to your computer and use it in GitHub Desktop.
Research Interests

My background

I am interested in biological problems that require both computational and statistical methods. I started out doing tropical ecology. Then I was a software tester for a while, where I learned programming. I took a bunch of statistics classes at night (two of my favorites were Bayesian Statistics and Stochastic Simulation). I am a fan of open source collaboration and plan on using software engineering techniques like automated testing, version control and continuous integration in my future research. Here are some things I want to learn more about:

Statistics

  • Bayesian hierarchical modeling and probabilistic graphical models
  • Bayesian nonparametrics, including all the stochastic processes theory beneath it (Dirichlet processes, etc)
  • Nonparametric and semiparametric statistics in general
  • Variational Inference- I find this an intriguing alternative to MCMC
  • Spatial Statistics (I love looking at maps)
    • How does the choice of projection affect things? Is there a way to do 3-D spatial stats measuring distance in great circles, etc.?
  • Causal Inference- I don't know much about it currently but it seems interesting.
  • Missing Data- I am somewhat familiar with the EM algorithm and the Bayesian approach of using "latent parameters" but I'd like to learn more about the state-of-the-art, and especially how to address computational issues that arise.

Computation

  • building bioinformatics "pipelines"
  • High-dimensional data techniques
  • Machine Learning- I took a Coursera class on this and want to get a deeper, more theoretical understanding. I'd like to also trying some of the tools like elastic net, SVMs, etc. on different kinds of large datasets.
  • Graph data (eg, social networks), computing and stats on large, non-relational databases
  • The Julia programming language- I am fond of Python, but this new language is very exciting

Biology

  • Metagenomics- this seems like an exciting place for testing ecological hypotheses that could have relevance to health outcomes. For example, ecologists often debate whether diversity of a system affects its stability. I wonder if higher microbiome diversity is associated with stability? Or, in another example, is someone who eats a wide variety of types of food (such as an international traveler) more likely to have a diverse gut flora than someone who doesn't travel?
  • Epigenetics- how can we measure epigenetic modifications?
  • Mental Health
  • Cancer
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment