Welcome to EnrichRBP

Welcome to EnrichRBP, the first automated and interpretable platform designed for the prediction and analysis of RBP binding sites across circRNAs, linear RNAs and RNAs in various cellular contexts. EnrichRBP supports a total of 70 state-of-the-art models, including feature reconstruction, feature selection, machine learning, and deep learning modules, capable of characterizing encoders, model training, comparison, and evaluation in a fully automatic pipeline. Furthermore, EnrichRBP provides comprehensive result visualization analysis for predictive models, covering aspects such as model interpretability, feature analysis, and the discovery of functional sequence regions.

Search your task by jobID

Search

Non-Custom Prediction Module

In the non-custom prediction module, we provide nine pre-trained models for rapid identification, including recognition of CircRNA-RBP, Linear RNA-RBP, and RNA-RBP in different cellular contexts. To be specific, there are three main stages in this module: (i) Data input; (ii) Pre-trained model loading; and (iii) Visualization analysis. These ready-to-use classifiers greatly reduce the memory consumption and the computing time required.

Custom Prediction Module

In the custom prediction module, EnrichRBP allows 61 state-of-the-art methods including feature reconstruction, feature selection, machine learning, and deep learning. Specifically, we have three stages for this module: (i) Data input and selection; (ii) Custom method selection; and (iii) Result visualization. EnrichRBP implements personalized wrapper functions for each step to provide security and convenience of computation.

Result Report Module

In the Result Report module, in order to elucidate model performance and interpretability, EnrichRBP generates a total of 27 visual analyses for the two main modules described above, across five domains: datasets statistical, analysis of the training process, visualisation of the results, feature analysis and biological interpretability. Through these visualisation mechanisms, EnrichRBP provides insights into the model's decision-making process, enhancing trust and understanding.