PAIRNet Predicts PIWI Cleavage Specificity Via Position-Aware Modeling

Researchers publish PAIRNet on February 19, 2026, a deep learning framework that predicts PIWI-mediated RNA cleavage rates by modeling guide-target interactions with positional embeddings and a hybrid CNN-Transformer. Validated across four PIWI-guide datasets, PAIRNet improves Pearson correlation by up to 34.7% (MILI) and 14.6% (MIWI) versus prior methods. The model recapitulates biochemical rules and aids piRNA guide design and mechanistic studies.