## RNA type labelling

This repo has the code and associated docker/singularity images to run a heuristic labelling model over RNAcentral data.

We use the snorkel library, which builds a probabilistic model based on a set of labelling functions to heuristically label entries. This should be better than using the heuristics directly, which is (more or less) what happens now.

The whole thing is packaged into a nextflow pipeline, so it should in theory be reproducible.