Lecturer creates app to help his non-coding students

What do you do when the tools you have at hand can’t quite deliver what you need them to? For Farhad Panah, postdoc at the Department of Food Science, the solution was to build something that could: MicroLoop.

It’s an interactive app for microbiome data analysis that can provide both general overviews of the data, but also in-depth, high quality and high-resolution plots and tables – and it can do it quickly and to users that do not necessarily have a great deal of coding knowledge.

“The main point of MicroLoop is to speed up microbiome analysis, and to make that type of data be more accessible to people that aren’t experts in coding and statistics. Using this app, you basically don’t need to code at all. The only thing you have to do is, essentially, switch on and off some buttons, and navigate some sliders. And with this, you can produce publication-quality results and tables or export the data for further analysis if you like. And it can do these things in a fraction of an hour as opposed to something closer to a week’s time,” says Farhad Panah.

Working from experience

The idea of creating MicroLoop has been brewing since Farhad was doing his Ph.D. in the early 2020’s, where the available tools then couldn’t do as much as he would have wanted.

“When you’re working with these types of data, the datasets are very large. The sequences that we can recover from humans microbiome goes up to 6000 species. Imagine you have a table with 4 patients in the column – that’s 6000 rows for each of them. Even scrolling through it is quite tedious. So, you need a machine that is intelligent enough to capture the patterns in the data and translate it,” explains Farhad.

Eventually, in 2022, Farhad became one of the lecturers on the Microbiome Analysis course, and in this setting, he wanted to be able to provide his students with a tool for bioinformatics that does not require them to be experts in coding.

“I wanted to remove the need for coding experience from the equation as much as possible for my course. So, last summer, while I was vacationing in Norway, I ended up coding and coding and coding to prepare this software. I really wanted to test it during the last year’s course. And, miraculously, I did it, and on the course, we finished all the analyses that we wanted to,” says Farhad.

Big plans for the future

All in all, MicroLoop does diversity-based analysis, compositional based analysis, and differential abundant analysis, explains Farhad. That means users can get an analysis of the microbial taxa in a sample, get clear overviews of the different strains of bacteria in them, and compare those results between samples. 

And while all of these things are what you’d expect from a good tool for microbial analysis, what sets MicroLoop apart is the speed with which it delivers high quality plots and tables, as well as its useability for non-coders. And this is what Farhad is hoping will make it an important tool for others in the future.

“I’m hoping that, in the future, it can be introduced in hospitals, in research facilities, in the industry and even in the context of decision-making. I have big plans to further develop it, such as implementing AI in it, so it’ll be able to immediately recognize patterns pointing to Parkinson’s or Irritable Bowel Syndrome, for example, even from unknown samples,” says Farhad.

For more resources on MicroLoop, which you can find here, he has made a tutorial, and you can find the app's github repository here.