A look at the Mojo language for bioinformatics
A few days ago, a blog post was posted on the website of Modular, the company behind the new high-performance programming language Mojo.
The post made the case for using Mojo in bioinformatics due to Mojo's dual features of being high-level language with high performance, and the blog author substantiated the case by presenting two benchmarks related to the processing of FASTQ files, showing impressive speed.
As a bioinformatician who is obsessed with high-performance, high-level programming, that's right in my wheelhouse! I decided to dig deeper into the benchmark, and this post is about what I found out.
The basic premise of the post is that the field of bioinformatics is struggling to handle its increasingly large datasets. These datasets are so large that they have to be processed programmatically, but programming is the field is split between high-level dynamic languages used to do the actual data analysis, and the high-performance, static languages that Python calls into to do the computation underlying the analysis. As the post states:
This creates a two-world problem where bioinformaticians who are not skilled in low-level languages, are prohibited from understanding, customizing, and implementing low-level operations.
