The Baserate Fallacy, revisited

In which I share some iPython code for an interactive demo to calculate and visualise the probability that a positive test implies a positive result. You can get it here, and preview it here. Then I get entangled in a topic outwith my expertise.

ANI are you okay? Are you okay ANI?

In which I describe Average Nucleotide Identity (ANI), which we can use to pigeonhole bacteria into conceptual boxes labelled as 'species'. And I share more code.

A Nice New Paradox

In which I work through a popular statistical puzzle/paradox (with potential implications for interpretation of large data studies). With example code.

Surely this has been done already...

In which it turns out that if it was done already, it was hiding somewhere. Also, I share a script that retrieves the corresponding nucleotide coding sequences from NCBI, given only a set of protein sequences.

KEGGWatch, part III

In which I finally get around to sharing some code, and give some examples of downloading and modifying KEGG pathway maps.

KEGGWatch, part II

In which I don't quite get around to writing a KGML parser and visualisation module (all very Tristram Shandy, this!), with a view to submitting to Biopython. This post describes some of the rationale and design choices - tune into part III for code and examples of use.

KEGGWatch, part I

In which I attempt to visualise metabolic maps for comparative genomics, and lead up to making a contribution to Biopython.

The Colours, Man! The Colours!

In which I take a short diversion into colour theory, and share some code to automate colour selection for class data.

23 And Me and Me: Part 3

In which I get a nice surprise.

On Reciprocal Best BLAST Hits

In which I narrowly avoid a rant. Reciprocal best BLAST hits can improve the quality of your searching, and are a good way to find candidate orthologues. There's evidence and everything.

23 And Me And Me: Part 1

In which I cave in to curiosity, and get myself genotyped.

Dead fish, and multiple-test correction

In which a salmon is resurrected, but not enough to really be significant.  Why finding 20 positive results when your P-value threshold suggests you should only see 10 isn't necessarily anything to be excited about.  And an introduction to Bonferroni and Benjamini-Hochberg multiple test correction.

The Base Rate Fallacy in Effector-Finding

In which an oft-overlooked bit of genome-mining statistics is considered, and your enjoyment of a holiday could depend heavily on other people's hygiene.

Jambiguity

In which a blog post is brought to you by the letter 'J', but should probably have been 'X'-rated.

A Pile of Sheets

In which a niggle with Excel files is easily crushed into nothingness.  If you have a Mac.