Anomaly-Guided Folding Pathway Discovery for Multidomain Proteins

Build an “anomaly mining” pipeline that compares folding pathway predictions from a simple structure-based statistical mechanical model (e.g., Ooka & Arai, 2023) against DL-derived structural ensembles (AlphaFold2; Jumper et al., 2021) and remote-homolog-inferred trajectories (PAthreader; Zhao et al., 2023). Automatically flag and analyze regions where predicted contacts, intermediates, or order-of-domain folding disagree, and then design perturbations (mutations, disulfide toggling, domain swaps) to adjudicate which mechanism better explains experimental data. This approach treats disagreements between models as positive signals to mine conflicts as hypotheses about hidden nonlocal interactions, kinetic traps, or disulfide-dependent switches. It extends Ooka & Arai’s nonlocal-interaction model as a mechanistic prior and contrasts it with PAthreader’s evolutionary pathway conjecture, exploiting AF2’s uncertainty metrics as signatures of dynamics. The method targets multidomain and disulfide-rich proteins where mechanistic understanding is limited, enabling rapid validation of non-canonical couplings, cryptic intermediates, and refinement of energy functions. The impact is a general methodology for turning unexpected results into mechanistic discoveries and improved predictive models of folding pathways, especially relevant to complex multidomain proteomes.

References:

1. Accurate prediction of protein folding mechanisms by simple structure-based statistical mechanical models. Koji Ooka, M. Arai (2023). Nature Communications.
2. Protein structure and folding pathway prediction based on remote homologs recognition using PAthreader. Kailong Zhao, Yuhao Xia, Fujin Zhang, Xiaogen Zhou, Stan Z. Li, Guijun Zhang (2023). Communications Biology.
3. Highly accurate protein structure prediction with AlphaFold. J. Jumper, Richard Evans, A. Pritzel, Tim Green, Michael Figurnov, O. Ronneberger, Kathryn Tunyasuvunakool, Russ Bates, Augustin Žídek, Anna Potapenko, Alex Bridgland, Clemens Meyer, Simon A A Kohl, Andy Ballard, A. Cowie, Bernardino Romera-Paredes, Stanislav Nikolov, Rishub Jain, J. Adler, T. Back, Stig Petersen, D. Reiman, Ellen Clancy, Michal Zielinski, Martin Steinegger, Michalina Pacholska, Tamas Berghammer, Sebastian Bodenstein, David Silver, O. Vinyals, A. Senior, K. Kavukcuoglu, Pushmeet Kohli, D. Hassabis (2021). Nature.