GraphTFactor: A graph-based approach for proteins transcription factor prediction
| Source: arXiv AI
Tags: GraphTFactor, transcription factors, graph neural networks, protein sequences, bioinformatics
GraphTFactor models protein sequences as graphs to classify transcription factors end to end, reports beating prior methods on benchmarks, and ships as open-source software with a web app.
Details
Identifying transcription factors, which regulate gene expression, is a core task in computational biology. GraphTFactor encodes protein sequences as graphs, which handles variable-length inputs without length limits while keeping relational information, and classifies transcription factors directly. The authors report it outperforms state-of-the-art methods on benchmark datasets and gives interpretable insight into which sequence regions drive predictions. They release an open-source package and an interactive web application for predictions on custom sequences.