% IMPORTANT: The following is UTF-8 encoded.  This means that in the presence
% of non-ASCII characters, it will not work with BibTeX 0.99 or older.
% Instead, you should use an up-to-date BibTeX implementation like “bibtex8” or
% “biber”.

@ARTICLE{Breimann:279438,
      author       = {Breimann, Stephan and Kamp, Frits and Basset, Gabriele and
                      Abou-Ajram, Claudia and Güner, Gökhan and Yanagida, Kanta
                      and Okochi, Masayasu and Müller, Stephan A and
                      Lichtenthaler, Stefan F and Langosch, Dieter and Frishman,
                      Dmitrij and Steiner, Harald},
      title        = {{C}harting γ-secretase substrates by explainable {AI}.},
      journal      = {Nature Communications},
      volume       = {16},
      number       = {1},
      issn         = {2041-1723},
      address      = {[London]},
      publisher    = {Springer Nature},
      reportid     = {DZNE-2025-00769},
      pages        = {5428},
      year         = {2025},
      abstract     = {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
                      $60\%$ to $90\%,$ and achieves an $88\%$ success rate in
                      experimental validation. Building on these advancements, we
                      identify pathways and diseases not linked before to
                      γ-secretase. Generally, CPP decodes physicochemical
                      signatures-a concept that extends beyond sequence motifs. We
                      anticipate that our approach will be broadly applicable to
                      diverse molecular recognition processes.},
      keywords     = {Amyloid Precursor Protein Secretases: metabolism / Amyloid
                      Precursor Protein Secretases: chemistry / Amyloid Precursor
                      Protein Secretases: genetics / Humans / Machine Learning /
                      Substrate Specificity / Alzheimer Disease: metabolism /
                      Algorithms / Protein Binding / Amyloid Precursor Protein
                      Secretases (NLM Chemicals)},
      cin          = {AG Steiner / AG Lichtenthaler},
      ddc          = {500},
      cid          = {I:(DE-2719)1110000-1 / I:(DE-2719)1110006},
      pnm          = {352 - Disease Mechanisms (POF4-352)},
      pid          = {G:(DE-HGF)POF4-352},
      typ          = {PUB:(DE-HGF)16},
      pubmed       = {pmid:40593564},
      pmc          = {pmc:PMC12219630},
      doi          = {10.1038/s41467-025-60638-z},
      url          = {https://pub.dzne.de/record/279438},
}