A benefit of introducing LISP in bioinformatics is the cultural element that tools are expected to work on the command line using simple linux streams for I/O. (As well as a systemic allergy to dependencies, preferring to do everything from scratch.) You can make drop-in Lisp replacements of shitty Perl tools and no one will be the wiser, as long as you are producing bit-compatible results. I imagine establishing trust in your tool will be harder if you provide them with Lisp vs. Perl source.

Other communities (e.g. statistics, machine learning) have a culture of ramming all their tools in a toolbox-library/framework (R, sklearn). This makes for a much tightly coupled environment, making it much harder to introduce Lisp work.