eFold training data

Here's a breakdown of the data we used to train eFold. You can find these datasets on Hugging Face. The raw data in on this google drive.

Training stage	Name on HuggingFace	Source	Method	Number of sequences	Families	length [10, 199]	length [200, 499]	length [500, 999]	length [1000, 1999]	length [2000, inf]
Pre-training	rnacentral_synthetic	Sequences from RNA central	RNAstructure	226'729	All known families	176'486	49'463	780	0	0
Pre-training	ribo500-blast	Ribonanza Competition	RNAstructure + DMS and/or SHAPE	46'060	Unlabelled	46'049	11	0	0	0
Pre-training	bpRNA-1m	bpRNA-1m	Covariance analysis	66'715	Unlabelled, sRNA, tRNA	48'090	6'167	2'829	9'260	369
Fine-tuning	pri_miRNA	This work	RNAstructure + DMS	1'098	pri-miRNA	0	1'098	0	0	0
Fine-tuning	human_mRNA	This work	RNAstructure + DMS	1'456	mRNA	0	493	882	81	0
Testing	PDB	PDB	NMR, crystallography	355	Short non-coding RNA	343	6	6	0	0
Testing	viral_fragments	Peer-reviewed literature	RNAstructure + DMS	40	Viral RNA	12	17	11	0	0
Testing	lncRNA	Bugnon and al, 2022	RNAstructure + DMS	30	Long non-coding RNA	0	2	1	27	0
Testing	archiveII	Archive II	Covariance analysis	3'370	rRNA, tRNA, tmRNA, unlabelled	2'004	1'276	79	11	0

Notes

The normalisation, filtering and replicates matching of SEISMIC data is made using seismic__postprocessing.py.