Entering edit mode
15 months ago
claudiofr
▴
20
I'm a retired software engineer with an early career background in biochemistry and a basic understanding of bioinformatics looking to contribute to the bioinformatics space in retirement. I made a few contributions to the galaxy project but the work there did not involve bioinformatics per se.
While I am not a software developer I don't think I have come across projects which actively seek contributors.
If you are a software engineer then perhaps you may be able to make contributions to improving existing software. A significant number of software packages in bioinformatics space are written by people without formal engineering training and likely can use some polish in terms of making them more efficient.
This doesn't mean that contributors will be turned away. I would recommend reading a few issues of the bioinformatics journal or nucleic acids research (or just spend a few weeks digging through BioRxiv) to see what catches your interest; and go from there. "I can code and am looking for something interesting to do" isn't enough to garner specific recommendations.
thanks. I started looking BioRxiv but I have to admit that it is a lot to go through and absorb.
There are 400 outstanding issues on Biopython which you could work on :)
https://github.com/biopython/biopython/issues
I am also a software engineer and also went through the BioRvix grind by spending a couple hours per week reading papers. You will get comfortable in ~ 3-4 months time (proactively looking things up if you don't understand something).
Having said that I am thinking about creating an open source error logging service for bioinformatics to make it easier for people to understand their errors across tools. Think of it like Snyk.io or Sentry.io for bioinformatics. The early objective would be to help people setup their environment.
you can contact me on Twitter if you want to brainstorm about it (I am not that often on the Biostars forums).
https://twitter.com/vincent_hus
Some details regarding your interests may be helpful as well. "Bioinformatics" is a big space. For example, are you interested/experienced in algorithm development? Data viz? Processing pipelines? Modeling?
My experience has been mostly in backend development which I suppose would mean algorithm development, processing pipelines, modelling would be things that would appropriate. I have used a number of different programming languages over the years but my preference would be python which I used on the recent galaxy project contributions. As far as application domain is concerned something related to the application of bioinformatics to studying the genetic basis of disease, but I am open.
Well, it's not python, but there are thousands of packages on Bioconductor for R. Poke around, look at a few that seem interesting, check their Githubs for issues or planned features that you could help with. They also have a slack channel where you could potentially find some collaborators.
Python lacks the Bioconductor ecosystem and standardized data structures for bioinformatics, though there are some early efforts to port those to Python that are making great progress. You might consider looking at that project and getting in touch with the devs are digging through their various repos.
Thanks. I also came across biopython. Do you know the difference between it and biocpy which you reference above?
biocpy is attempting to reproduce Bioconductor's very useful R S4 classes and methods in python, which would be quite useful for other developers and enable more seamless R<->python conversions and analyses. It is new and has only a few developers working on it (though they've already made impressive progress).
biopython is a grab-bag of loosely connected modules for dealing with biological data and analyses in python. It is pretty expansive (some might say too expansive), and one of the first python package meant to handle biological data.
Personally, I feel biopython is starting to show its age to some degree, though it's undoubtedly still useful and popular. I am sure there are no shortage of issues/features for it that would welcome your help.