Functional annotation of proteins for signaling network inference in non-model species

• Published in: Nature communications Vol. 14; no. 1; p. 4654
• Main Authors: Van den Broeck, Lisa, Bhosale, Dinesh Kiran, Song, Kuncheng, Fonseca de Lima, Cássio Flavio, Ashley, Michael, Zhu, Tingting, Zhu, Shanshuo, Van De Cotte, Brigitte, Neyt, Pia, Ortiz, Anna C., Sikes, Tiffany R., Aper, Jonas, Lootens, Peter, Locke, Anna M., De Smet, Ive, Sozzani, Rosangela
• Format: Journal Article
• Language: English
• Published: London Nature Publishing Group UK 03.08.2023; Nature Publishing Group; Nature Portfolio
• Subjects: 631/114/2397; 631/449/1659; 631/449/2661/2665; 82/58; Amino acid sequence; Annotations; Bayesian analysis; Glycine max; Humanities and Social Sciences; Kinases; Molecular biology; multidisciplinary; Multilayers; Neural networks; Oryza sativa; Phosphatase; Phosphorylation; Predictions; Proteins; Regulatory proteins; Science; Science (multidisciplinary); Sorghum; Sorghum bicolor; Soybeans; Statistical inference; Substrates; Triticum aestivum; Wheat; Zea mays
• ISSN: 2041-1723; 2041-1723
• DOI: 10.1038/s41467-023-40365-z

Molecular biology aims to understand cellular responses and regulatory dynamics in complex biological systems. However, these studies remain challenging in non-model species due to poor functional annotation of regulatory proteins. To overcome this limitation, we develop a multi-layer neural network that determines protein functionality directly from the protein sequence. We annotate kinases and phosphatases in Glycine max . We use the functional annotations from our neural network, Bayesian inference principles, and high resolution phosphoproteomics to infer phosphorylation signaling cascades in soybean exposed to cold, and identify Glyma.10G173000 (TOI5) and Glyma.19G007300 (TOT3) as key temperature regulators. Importantly, the signaling cascade inference does not rely upon known kinase motifs or interaction data, enabling de novo identification of kinase-substrate interactions. Conclusively, our neural network shows generalization and scalability, as such we extend our predictions to Oryza sativa , Zea mays , Sorghum bicolor , and Triticum aestivum . Taken together, we develop a signaling inference approach for non-model species leveraging our predicted kinases and phosphatases. An artificial-intelligence network is used to generate highly accurate predictions of proteins’ functionality. The predictions on the identity of regulatory proteins is used to create regulatory networks and make discoveries about complex biological systems.