In the past months I’ve always wanted to write a small Plasma applet to aid me in some boring tasks as a bioinformatician. One example (for the non-scientific crowd out there) is when I find a specific gene out of my analysis work which I want to take a look at. I am often lazy, so instead of firing up the browser to look at the online resources, I wanted to write something which could access said resources programmatically.

I’ve been wondering about why FOSS is often compared to the academic world, but at least in my limited experience, I see little people that grasp its concept in the world of research. On a quick look, developing FOSS in a research environment would be very good: not only you’d get publicly available results when you publish, but at the same time you can make sure that in an extreme case your application will be carried on by someone else should you not be able to continue development.

I’m often wondering why people only resort to R when working with microarrays. I can understand that Bioconductor offers a plethora of different packages and that R’s statistical functions come in handy for many applications, but still, I think people underestimate the impact of performance. R is not a performing language at all, it doesn’t parallelize well when using HPC (at least from the talks I’ve had with people studying the matter), and in general is a memory and resource hog.

While working today on an annotation class in Python I stumbled on a problem. Normally I work with lists of genes that are consistent, i.e. all Entrez Gene IDs (or RefSeq IDs, or Genome Browser IDs…), but today I had a list of mixed identifiers. The subsequent idea was “let’s implement auto-detection of common identifiers in the class”. The problem is… is there any actual documentation on how identifiers are made?

**Notice:**Just now I realized this has been linked to to a Stack Overflow question. I recently wrote a new post that uses a different technique and a combination of R and Python. [Check it out!]({{ site.url }}/2011/05/multiscale-bootstrap-clustering-with-python-and-r)

Following up my recent post, I’ve been looking for alternatives to TMeV. So far I’ve found the R package pvclust and the Pycluster library, part of BioPython. The first one also performs bootstrapping (I’m not sure if it’s similar to what support trees do, but it’s still better than no resampling at all). I’ve found another Python project but it is still too basic to perform what I need.

As people who read my science-related posts know already, I’m not a big fan of {{post id=“software-and-biological-research” text=“software made just to support a publication”}}. Recently I’ve stumbled again into similar software. Namely, I’m talking about the TIGR Multiexperiment Viewer (TMeV), a Java-based program which is often used for microarray analysis. It’s not exactly “fit for publication”, because it has reached version 4 last year, but shows some of the problems ({{post id=“genbugg” text=“mentioned already”}}) with releasing bioinformatics software.

I use TMeV mostly because I didn’t find any other implementation of the hierarchical clustering algorithm with support trees. However, I’ve stumbled upon a very annoying bug in the most recent version. Normally I use average linkage clustering and as the distance metric I employ the Pearson’s correlation, and with gene and sample bootstrapping: with certain files this makes TMeV report errors at random during the iterations.

Today I started working on a data set published on GEO. As the sample data were somehow inconsistent (they mentioned 23 controls when I found 28), I decided to parse the SOFT file from GEO in order to get the exact sample information. I did a grave mistake. First of all, Biopython’s SOFT parser is horribly broken (doesn’t work at all) and quite undocumented: I could work around the lack of documentation (API docs) but not with the fact that it wouldn’t work.

Today I was looking for an easy way to do some calculations of raw expression data on Affymetrix arrays, but I didn’t want to use R: I have already mentioned how I don’t like its design and implementation. While looking for some documentation, I stumbled upon this nifty little program called RMAExpress.