Skip to content

Acknowledgements

Citation

If ChopChopMF was useful for your work, please cite the repository - see CITATION.cff for the canonical, machine-readable entry (GitHub renders this as a "Cite this repository" button on the repo page).

Contributors

We Thank Christoph Büschl and Mathias Percipalle (Schur Group, Institute of Science and Technology Austria) for their insightful discussions and contributions toward the refinement of these tools. We are particularly grateful to Tom Goddard (University of California, San Francisco) for the release of UCSF ChimeraX version 1.10.dev202502272020, which addressed critical bugs within the Foldseek integration, and for his valuable feedback that helped shape the software's development. Further we want to thank Pim Huis in ‘t Veld (Huis Group, Max Perutz Labs), for discussing the idea of a ΔG Filter for the PDBePISA analysis.

We thank Mariana Pereira Guarda (de Bono Group, Institute of Science and Technology Austria) for her assistance in selecting the color schemes and for her conceptual ideas that directly influenced the implementation of new features. We also thank Ana Chegão for her suggestions regarding the nomenclature of the ChopChop plug-in and Philipp Bauer for his expert guidance in designing the intuitive graphical icons for the toolbar. Finally, we acknowledge the use of large language models, specifically ChatGPT-4, Gemini 1.5, and Claude (Anthropic), as computational copilots for scripting, bug identification, and feature development throughout the project.

Citations for Interface Scores

The "Interface Scores" reported in the PAE Analysis tool's Scores tab implement published, peer-reviewed methods. If you use them in your work, please cite the original sources:

  • pDockQ — Bryant, P., Pozzati, G. & Elofsson, A. Improved prediction of protein-protein interactions using AlphaFold2. Nat Commun 13, 1265 (2022). doi:10.1038/s41467-022-28865-w
  • mpDockQ — Bryant, P., Pozzati, G., Zhu, W., Shenoy, A., Kundrotas, P. & Elofsson, A. Predicting the structure of large protein complexes using AlphaFold and Monte Carlo tree search. Nat Commun 13, 6028 (2022). doi:10.1038/s41467-022-33729-4
  • pDockQ2 — Zhu, W., Shenoy, A., Kundrotas, P. & Elofsson, A. Evaluation of AlphaFold-Multimer prediction on multi-chain protein complexes. Bioinformatics 39, btad424 (2023). doi:10.1093/bioinformatics/btad424
  • LIS / cLIS / iLIS — Kim, A.-R., Hu, Y., Comjean, A., Rodiger, J., Mohr, S. E. & Perrimon, N. Enhanced Protein-Protein Interaction Discovery via AlphaFold-Multimer. bioRxiv (2024). doi:10.1101/2024.02.19.580970 (github.com/flyark/AFM-LIS)
  • ipSAE (d0chn, d0dom, d0res variants) — Dunbrack, R. L. Jr. github.com/DunbrackLab/IPSAE. The underlying d0(L) formula is the standard TM-score d0: Yang, Y. & Skolnick, J. Scoring function for automated assessment of protein structure template quality. Proteins 57, 702–710 (2004). All three of ipSAE's own d0 variants are implemented (fixed chain-length d0chn, adaptive-interface-size d0dom, per-residue d0res) - the adaptive full ipSAE score as originally packaged (which additionally selects/reports the best-scoring variant per chain pair) is not reproduced here.
  • Salt bridges — Barlow, D. J. & Thornton, J. M. Ion-pairs in proteins. J Mol Biol 168, 867–885 (1983). A charged Arg/Lys/His side-chain nitrogen within 4.0 Å of a charged Asp/Glu side-chain oxygen.
  • Buried interface area and hydrogen bond count use UCSF ChimeraX's own built-in measure buriedarea and hbonds commands rather than a separate implementation — please also cite ChimeraX itself: Pettersen, E. F. et al. UCSF ChimeraX: Structure visualization for researchers, educators, and developers. Protein Sci. 30, 70–82 (2021). doi:10.1002/pro.3943
  • PAE-based domains (Investigate's Health Check, chopchop domains) — Croll, T. pae_to_domains (github.com/tristanic/pae_to_domains). Graph-based community clustering of the PAE matrix (residue = node, edge weight 1/PAE^p for pairs with PAE below a cutoff, greedy modularity communities via networkx) — default parameters (5.0 Å cutoff, power 1, resolution 1) are the reference tool's own published defaults.
  • N-glycosylation sequons (chopchop glycosites, Glyco and Disulfide tool) — Bause, E. Structural requirements of N-glycosylation of proteins. Biochem J 209, 331–336 (1983). Asn-X-Ser/Thr with X not Pro; a sequon is only a candidate site, not proof of glycosylation.
  • Disulfide candidates (chopchop disulfides) — a plain geometric filter on the Cys SG–SG distance (default 2.5 Å, a deliberately generous heuristic of this tool; real disulfides are about 2.0–2.1 Å), plus whether the file carries a bond between the two atoms.
  • Homodimer asymmetry (chopchop asymmetry) — compares the C-alpha distance matrices of two copies of a chain, which needs no superposition; a simple measure written for this tool.
  • Structure Compare superposition uses UCSF ChimeraX's own built-in matchmaker command (sequence alignment + iterative-pruned Kabsch fit) rather than a separate implementation — see the ChimeraX citation above.

Citations for Cell Biology Motifs

The Phospho Sites, Signal Peptide, and TM Helix tools flag residues using hand-written regex approximations of published consensus motifs — a heuristic motif match, not a calibrated prediction (unlike SignalP/TMHMM/NetPhos, which are trained statistical predictors). Motifs marked (approximate) loosen the cited consensus (a wider spacer, a dropped positional constraint) rather than reproducing it exactly.

Phospho Sites (kinase consensus motifs, S/T/Y acceptor only):

  • ATM/ATR [ST]Q — Kim, S.-T. et al. J Biol Chem 274, 37538–37543 (1999).
  • PKA [RK]{2}.[ST] (approximate) — Kemp, B. E. & Pearson, R. B. Trends Biochem Sci 15, 342–346 (1990).
  • PKC [RK].{0,2}[ST].[RK] (approximate) — Nishikawa, K. et al. J Biol Chem 272, 952–960 (1997).
  • CDK [ST]P.[RK] — Songyang, Z. et al. Curr Biol 4, 973–982 (1994).
  • CK2 [ST].{2}[DE] — Pinna, L. A. Biochim Biophys Acta 1054, 267–284 (1990).
  • GSK3 [ST].{3}[ST] (approximate — omits the "primed", already-phosphorylated downstream residue the real consensus requires) — Fiol, C. J. et al. J Biol Chem (1987); Roach, P. J. (1990).
  • MAPK P..[ST]P (approximate) — Clark-Lewis, I. et al. J Biol Chem 266, 15180–15184 (1991).
  • AKT R.R..[ST] — Alessi, D. R. et al. FEBS Lett 399, 333–338 (1996).
  • AMPK [MFLIV]R..[ST]...[MFLIV] (approximate) — Gwinn, D. M. et al. Mol Cell 30, 214–226 (2008).

Signal Peptide (subcellular targeting-signal motifs):

  • NLS, monopartite [RK]{4,6} (approximate) — Dingwall, C. & Laskey, R. A. Trends Biochem Sci 16, 478–481 (1991).
  • NLS, bipartite [RK]{2}.{10,12}[RK]{3} — Robbins, J., Dilworth, S. M., Laskey, R. A. & Dingwall, C. Cell 64, 615–623 (1991).
  • NES, leucine-rich — Güttler, T. et al. Nat Struct Mol Biol 17, 1367–1376 (2010).
  • ER retention (KDEL/HDEL) — Munro, S. & Pelham, H. R. B. Cell 48, 899–907 (1987).
  • Peroxisomal PTS1 [SAC][KRH][LM]$ (approximate) — Gould, S. J. et al. J Cell Biol 108, 1657–1664 (1989).
  • Peroxisomal PTS2 (approximate) — Swinkels, B. W. et al. EMBO J 10, 3255–3262 (1991).
  • Mitochondrial presequence heuristic (N-terminal positive/negative charge count) — this project's own threshold, inspired by the amphipathic, net-positively-charged presequence property described in von Heijne, G. EMBO J 5, 1335–1342 (1986) and Roise, D. et al. EMBO J 5, 1327–1334 (1986); the specific "≥4 positive, ≤1 negative in the first 30 residues" cutoff is not itself a published rule.

TM Helix: candidate helices are DSSP-detected alpha-helices scored by mean Kyte & Doolittle (1982) hydrophobicity. When available, ChimeraX's own built-in mlp lipophilicity table additionally gives a SASA-weighted per-helix score (the rbvi/chimerax-recipes "helixmlp" method) — please cite:

  • Kyte & Doolittle hydrophobicity scale — Kyte, J. & Doolittle, R. F. J Mol Biol 157, 105–132 (1982).
  • Fauchère–Pliska atomic lipophilicity (MLP) — Fauchère, J.-L. & Pliska, V. Eur J Med Chem 18, 369–375 (1983), as implemented in ChimeraX's mlp command.

License note: ChopChopMF is licensed under the GNU AGPLv3.