Paper-reported evidence

Generalizable deep-learning-based mRNA-protein interaction prediction strongly depends on protein diversity

2026 · Peer-reviewed · version of record

Paper-reported results. The values below come from this source, not an independent rewire.it run.

Open primary paper →DOI: 10.1186/s13321-026-01197-3

Primary full text verified via Europe PMC fullTextXML; venue: Journal of Cheminformatics; PMC ID: PMC13235059. ProteinBERT encodes the protein side of an mRNA-protein task; score is not an RNA foundation-model result.

1 verified numerical rows · source checked 2026-09-15

Paper-reportedAuthor's model

ProteinBERT LLM-encoding model

mRNA-protein interaction prediction

Reported score: 71.5%
Metric: AUROC
Dataset / split: mRNA-RBP pairs · RBP-aware test set

LLM encoding of protein partner; RBP-aware partition tests generalization to unseen protein diversity

Table 2, RBP-aware test set row, auROC (%) columnVerify at source →