Multi-purpose RNA language modelling with motif-aware pretraining and type-guided fine-tuning

Topics

RNAErnie · RNA language model · transformer architecture · motif-aware pretraining · RNA motifs · motif-level random masking · masked language modeling · RNA type tokenization · miRNA · Long non-coding RNA · type-guided fine-tuning · out-of-distribution generalization · RNA type prediction · post hoc feature embedding refinement · RNA sequence classification · RNA interaction prediction · RNA structure prediction · nucleotide sequence analysis · biological priors in pretraining