A long overdue set of links and quick thoughts:
Communicating research

Nat challenges us to make open data available in a useful way, by analogy to open source work. How often do we really try to make our results and raw data accessible, and how often do we shove them in an Excel spreadsheet as supplemental info? Two inspiring examples of nice data visualizations: The Broad’s Tumorscape for visualizing copy number changes in cancer and a Google motion chart showing the growth of data for vertebrate phylogenetic analysis.

Titus describes his experiences teaching nextgen analysis to biologists, while using Amazon EC2 to provide the computational power. It’s a great demonstration of how cloud infrastructure can be used to democratize the process of getting started on complex analyses.

Daniel Lemire has useful suggestions on how to write good research papers. A nice mix of practical suggestions along with more general thoughts on being productive in research and writing clearly.

Nice discussion of diversity and gender equality in computer science. Lots of interesting points in the comments as well.
Statistics

Bradford has two great posts about teaching yourself machine learning and statistics. He first provides some tips for learning on your own and then lays out a lifetime of machine learning reading giving some great pointers about where to get started.

Abraham links to a paper providing practical considerations for knowing when to stop a Markov Chain Monte Carlo simulation.

Good arguments why two tailed hypothesis tests are more realistic than one tailed.
Data analysis

Revolutions has advice for dealing with large data sets from the 2010 Workshop on Algorithms for Modern Massive Data Sets.

The larry package for manipulating tables in Python. This uses NumPy under the covers and is similar to dealing with data.frames in R.

Will describes how callbacks can drive an analysis pipeline. As analysis workflows get more complicated, your code can get to be a mess of special cases and become really fragile. Here he passes around functions through a standard runner to help generalize and abstract the process.
Phylogenetics

The wiki from the 2010 Bodega Phylogenetics workshop is a gold mine of tutorial style documentation for doing phylogenetics.

Practical advice on testing for nonlinear selection: summarizes a recent paper providing suggestions for improving the standard techniques.

Thomas provides a couple of awesome paper reviews: detecting selective sweeps with HMMs and integrating phylogenetics, speciation and selection into analysis of primate evolution.

Thomas again with lecture notes on estimating divergence times using molecular clocks.
Visualization

Stephen at getting genetics done provides R code to arrange multiple ggplot2 graphs into a single output.

The Protovis graphical toolkit provides an awesome javascript framework for web enabled graphics. Chris demonstrates building a genome browser with it.

R code and example graphics for fitting a loess curve to a scatterplot.

Nice discussion and examples from the Juice Analytics folks on how to make visualizations cool enough to remember and understandable enough to take action on.