Fig. 7
From: UTR-DynaPro: a CNN–transformer multimodal language model for decoding 5′UTR regulatory mechanisms
Prediction of MRL on the Human_Vary dataset ((human 5′ UTRs) by UTR-Dynapro and baseline models (UTR-Insight, Informer, UTR-LM, Optimus, and FramePool). Performance comparison across multiple evaluation metrics: (a) coefficient of determination, (b) Pearson correlation coefficient, (c) Spearman correlation coefficient, (d) root mean square error, and (e) mean absolute error. (f) Radar chart summarizing overall performance across all metrics. UTR-Dynapro consistently achieves the best or near-best results, highlighting its superior accuracy and robustness compared to existing methods.
