ProRefiner: an entropy-based refining strategy for inverse protein folding with global graph attention

• Published in: Nature communications 2023-11, Vol.14 (1), p.7434-12, Article 7434
• Main Authors: Zhou, Xinyi, Chen, Guangyong, Ye, Junjie, Wang, Ercheng, Zhang, Jun, Mao, Cong, Li, Zhanwei, Hao, Jianye, Huang, Xingxu, Tang, Jin, Heng, Pheng Ann
• Format: Article
• Language: English
• Subjects: 631/114/1305; 631/114/2397; 631/114/469; 631/114/794; 631/535/1267; Algorithms; Benchmarks; Context; Critical components; Design; Entropy; Folding; Genetic modification; Genome editing; Humanities and Social Sciences; multidisciplinary; Protein Folding; Proteins; Proteins - chemistry; Proteins - genetics; Residues; Science; Science (multidisciplinary); Transposase
• ISSN: 2041-1723; 2041-1723
• DOI: 10.1038/s41467-023-43166-6

Inverse Protein Folding (IPF) is an important task of protein design, which aims to design sequences compatible with a given backbone structure. Despite the prosperous development of algorithms for this task, existing methods tend to rely on noisy predicted residues located in the local neighborhood when generating sequences. To address this limitation, we propose an entropy-based residue selection method to remove noise in the input residue context. Additionally, we introduce ProRefiner, a memory-efficient global graph attention model to fully utilize the denoised context. Our proposed method achieves state-of-the-art performance on multiple sequence design benchmarks in different design settings. Furthermore, we demonstrate the applicability of ProRefiner in redesigning Transposon-associated transposase B, where six out of the 20 variants we propose exhibit improved gene editing activity. Inverse Protein Folding is a critical component of protein design. Here, authors introduce ProRefiner, a deep-learning model for IPF that exhibits both high performance and memory efficiency, thereby contributing to advancements in protein design.