%0 Journal Article
%A Breimann, Stephan
%A Kamp, Frits
%A Basset, Gabriele
%A Abou-Ajram, Claudia
%A Güner, Gökhan
%A Yanagida, Kanta
%A Okochi, Masayasu
%A Müller, Stephan A
%A Lichtenthaler, Stefan F
%A Langosch, Dieter
%A Frishman, Dmitrij
%A Steiner, Harald
%T Charting γ-secretase substrates by explainable AI.
%J Nature Communications
%V 16
%N 1
%@ 2041-1723
%C [London]
%I Springer Nature
%M DZNE-2025-00769
%P 5428
%D 2025
%X Proteases recognize substrates by decoding sequence information-an essential cellular process elusive when recognition motifs are absent. Here, we unravel this problem for γ-secretase, an intramembrane-cleaving protease associated with Alzheimer's disease and cancer, by developing Comparative Physicochemical Profiling (CPP), a sequence-based algorithm for identifying interpretable physicochemical features. We show that CPP deciphers a γ-secretase substrate signature with single-residue resolution, which can explain the conformational transitions observed in substrates upon γ-secretase binding. Using machine learning, we predict the entire human γ-secretase substrate scope, revealing numerous previously unknown substrates. Our approach outperforms state-of-the-art protein language models, improving prediction accuracy from 60
%K Amyloid Precursor Protein Secretases: metabolism
%K Amyloid Precursor Protein Secretases: chemistry
%K Amyloid Precursor Protein Secretases: genetics
%K Humans
%K Machine Learning
%K Substrate Specificity
%K Alzheimer Disease: metabolism
%K Algorithms
%K Protein Binding
%K Amyloid Precursor Protein Secretases (NLM Chemicals)
%F PUB:(DE-HGF)16
%9 Journal Article
%$ pmid:40593564
%2 pmc:PMC12219630
%R 10.1038/s41467-025-60638-z
%U https://pub.dzne.de/record/279438