Hi @Henrik,
I can weigh in on this too.
First of all, there will be a need for people with statistics and programming backgrounds in many areas of biomedical research for the foreseeable future. This definitely includes microbiome research, and I would expect psychobiology research as well. Biomedical research now requires “data science” skills that haven’t typically been included in biology undergraduate or graduate training, so there is a gap that needs to be filled.
As @wasade mentioned, workshops like the Workshop on Genomics are generally great for getting hand-on experience with different bioinformatics tools. That’s a great way to get started with the software being used in a specific field. I haven’t been involved with that specific workshop, but I’ve heard very good things. @wasade didn’t mention this one, which is in Germany in December and is being taught by he and @antgonza. That one will be microbiome focused.
If you’re looking for something that is more focused on basic data science skills, like programming, working on a Unix/Linux command line, etc, you might see what Software Carpentry or Data Carpentry is offering in your area. There are also some great books on these topics - see for example the Reading List in An Introduction to Applied Bioinformatics.
A great way to learn, in my opinion, would be to work through some of those basic data science resources while at the same time starting to work with the bioinformatics tools that you want to learn, such as QIIME 2. This will give you a way to apply the skills you’re learning as you’re learning them, which typically helps me to learn quickly. Then, once you know which specific bioinformatics tools you want to learn more about, invest in attending a workshop that covers those tools. Having a little experience with those tools, while typically not required for a workshop, will probably let you get more out of a workshop.
Given your interest in psychobiology and longitudinal studies, you might be interested in our FMT tutorial which is based on a longitudinal study of children with autism who are treated with fecal microbiota transplant. The tutorial is based on a subset of the full data set, but you could start with that and then move on to the full data set once you’re comfortable. You could, for example, set a goal to reproduce the results in Figure 3 of the original study with QIIME 2 and the q2-longitudinal plugin - I expect that you’d be pretty comfortable with QIIME 2 and the command line when you met that goal. (If you’re going to go this route, you should still follow the order described in the QIIME 2 Getting Started guide.)
Hope this helps! Good luck and have fun!