Compact Artificial Neural Network Models for Predicting Protein Residue -- RNA Base Binding
Stanislav Selitskiy
TL;DR
The paper addresses the need for accessible protein–RNA binding prediction by testing small, shallow feed-forward ANNs with context-window encodings to replace heavy structure-aware models. It evaluates multiple activation functions, resampling strategies, and ensemble approaches on a dataset derived from prior works, achieving high test accuracy with models that run on modest hardware. Key findings show that ReLU, KGate, and Tanh converge, with best-performing configurations surpassing several SVM-based baselines on the same data; ensembles further reduce false positives while maintaining strong recall. The work demonstrates that practical, low-resource models can provide effective first-pass screening for RNA–protein interactions, while also outlining limitations like model forgetting and the potential for integrating geometric or graph-based information in future work.
Abstract
Large Artificial Neural Network (ANN) models have demonstrated success in various domains, including general text and image generation, drug discovery, and protein-RNA (ribonucleic acid) binding tasks. However, these models typically demand substantial computational resources, time, and data for effective training. Given that such extensive resources are often inaccessible to many researchers and that life sciences data sets are frequently limited, we investigated whether small ANN models could achieve acceptable accuracy in protein-RNA prediction. We experimented with shallow feed-forward ANNs comprising two hidden layers and various non-linearities. These models did not utilize explicit structural information; instead, a sliding window approach was employed to implicitly consider the context of neighboring residues and bases. We explored different training techniques to address the issue of highly unbalanced data. Among the seven most popular non-linearities for feed-forward ANNs, only three: Rectified Linear Unit (ReLU), Gated Linear Unit (GLU), and Hyperbolic Tangent (Tanh) yielded converging models. Common re-balancing techniques, such as under- and over-sampling of training sets, proved ineffective, whereas increasing the volume of training data and using model ensembles significantly improved performance. The optimal context window size, balancing both false negative and false positive errors, was found to be approximately 30 residues and bases. Our findings indicate that high-accuracy protein-RNA binding prediction is achievable using computing hardware accessible to most educational and research institutions.