Mixture Models

Plotting mixture models, part 2: sum of gaussians

Jun 6, 2019

In the previous post I plotted multiple fit gaussians onto a data set with ggplot2 - the final model is actually the sum of these fitted gaussians though, so how do I plot that? This is a more tricksy problem. I know I need to nest the mapply function inside another function, but my mind doesn’t immediately leap to the answer - I tried to come up with something quickly, but it inevitably failed.

Plotting mixture models in ggplot2

Jun 6, 2019

This post deals with something niche but practical - getting ggplot2 to plot multiple fitted gaussians from a model with different amplitudes. Google failed to provide the answer I was looking for - if you can’t find it with Google, it must need a blog post. We have a couple of different projects running at the moment that need us to think about mixture models - these are common in single molecule biophysics work as you frequently end up with two or more populations of molecules stretched out along a continuous variable; could be velocity, step size or something else.

Moving the blog

Mar 3, 2018

I moved my blog! Not just in terms of where it’s hosted, but how the whole thing is built. Previously I’d been using ghost, which was fine up until the moment I wanted to publish more code/analysis. The process of copy-pasting code and exporting images was taking more time than writing the code itself. I’m now using the blogdown package which builds static websites within RStudio using Hugo and lets you publish the results.

Turning a kymograph into tidy data using the tidyverse in R

Feb 2, 2018

I have a bit of a love hate relationship with kymographs. In the way that they compress data there’s no doubt that you loose information, but in the world of axonal transport and low signal:noise they have clear advantages in enabling quantification. I covered before a couple of strategies you can use to import image data into R. The next step in my workflow is usually to turn that image into a data table for further analysis.

Kinesin nomenclature

Dec 12, 2017

Andreas Prokop raised the issue of kinesin nomenclature on twitter over the weekend “I am tired of having to look up lists!”. I couldn’t agree more, it’s insanely confusing - especially when you’re trying to synthesise literature across several model organisms (shout out to Ustilago maydis and Dictyostelium discoideum here). Why is it confusing? The standardised kinesin nomenclature was defined in 2004, where the superfamily was divided into sub families. They even defined how motors should be referred to in the text.

Sci twitter like a ‘pro'

Nov 11, 2017

The science twitter community can be a hugely beneficial resource, it can also be overwhelming if your new to it. I asked the scientists of twitter what platform they were using to interact with twitter itself (‘for a thing’). This post is not the thing I was eluding too, but the poll threw up a few tips I thought it worth drawing attention too and I figured I’d chuck some of my own tips in while I was at it.

Importing images into R

Sep 9, 2016

In May my paper was finally released into the wild. Amongst other things it was the culmination of 4 years of me learning to use R for data analysis (and in my case data = images of cells). I’ve been meaning to blog some of this for a while, but I will try for my own sanity to keep everything in bite size chunks. Starting at the start: data import.

The pregnant/new mum guide to science conferences

Jun 6, 2015

Pregnancy and new motherhood has brought a whole swathe of new experiences to my conference outings. This is my definitive guide to science conferences for the expecting/new Mum to help you decide where to expend your energy….

Merry Christmas!

Dec 12, 2014

When I was on the microscope imaging EB3 comets in axons as a control experiment for a control experiment, I got distracted by the comets in the dendrites - so pretty! They reminded me of the icicle lights on the outside of people’s houses you see this time of year. So this is my science Christmas card to you - have a Merry one :)

Drawing microtubules

Jan 1, 2014

Sometime ago I got recruited by Jake to help him come up with a summary figure for his paper on p150 and microtubule catastrophe. I place high scientific value on well designed descriptive schematics that help you visualise and explain the process your trying to describe, so I took the task on and decided to use it as an opportunity to build myself a decent microtubule model for later use.

Plotting mixture models, part 2: sum of gaussians

Plotting mixture models in ggplot2

Ali Twelvetrees

Plotting mixture models, part 2: sum of gaussians

Plotting mixture models in ggplot2

Moving the blog

Turning a kymograph into tidy data using the tidyverse in R

Kinesin nomenclature

Sci twitter like a ‘pro'

Importing images into R

The pregnant/new mum guide to science conferences

Merry Christmas!

Drawing microtubules