PSF: A Web Application Tool for Protein Scaffold Filling

The protein scaffold filling problem remains a significant challenge in computational proteomics, which is critical for accurate protein function prediction and drug design. Despite recent advancements, current sequencing methods often yield incomplete protein sequences, referred to as scaffolds, which require precise filling for further analysis. This paper presents a web-based application, implemented using the Django framework, adopting our previously developed machine learning and deep learning techniques for protein scaffold filling. The platform allows users to try our pre-trained models or train models on their datasets for new scaffolds. This system provides a versatile tool for researchers in computational proteomics, enhancing the efficiency of protein sequence prediction. The developed web application can be accessed through https://psf.ncat.edu/.