Suppressing self-looping in PAGA

I was wondering if there is any way to suppress self-looping in PAGA? I’m trying to replicate some RNA-seq analysis from a paper, and the trajectory observed for their dataset splits into two paths, but when I run PAGA, I don’t really get a good separation of paths because there is self-looping/cycles. Is there a way to deal with this?
I noticed that PAGA-tree does not infer loops, is there an implementation of this algorithm on Scanpy? Thanks!