Skip to content

Citation

If you use CETSAx–NADPH in your research, please cite the relevant methodological and data sources listed below.

Citing CETSAx–NADPH

If you use this pipeline or build upon it, cite the repository:

@software{cetsax_nadph,
  author = {Mishra, Abhinav},
  title = {CETSAx–NADPH: Explainable Protein–Metabolite Interaction Analysis},
  year = {2025},
  url = {https://github.com/yourusername/cetsax-nadph}
}
````
---

If you prepare a manuscript based on this work, include a brief description of the framework:

* CETSA-based dose–response modeling
* Composite sensitivity scoring
* Sequence-based prediction using protein language models
* Explainable AI for residue-level interpretation

---

## Primary Data Source

This work relies on CETSA data generated by:

```bibtex
@article{dziekan2020cetsa,
  author = {Dziekan, J. M. and Wirjanata, G. and Dai, L. and others},
  title = {Cellular thermal shift assay for the identification of drug–target interactions in the Plasmodium falciparum proteome},
  journal = {Nature Protocols},
  volume = {15},
  pages = {1881--1921},
  year = {2020},
  doi = {10.1038/s41596-020-0310-z}
}

This dataset forms the experimental basis for the modeling and analysis pipeline.

Protein Language Model (ESM-2)

Sequence modeling and embeddings are based on:

@article{lin2023esm,
  author = {Lin, Zeming and Akin, Hakan and Rao, Roshan and others},
  title = {Evolutionary-scale prediction of atomic-level protein structure with a language model},
  journal = {Science},
  volume = {379},
  pages = {1123--1130},
  year = {2023},
  doi = {10.1126/science.ade2574}
}

Methods and Algorithms

Logistic Dose–Response Modeling

The curve fitting approach is based on standard pharmacological dose–response models:

@book{motulsky2018pharmacology,
  author = {Motulsky, Harvey and Christopoulos, Arthur},
  title = {Fitting Models to Biological Data using Linear and Nonlinear Regression},
  year = {2018},
  publisher = {Oxford University Press}
}

Isotonic Regression

Monotonic smoothing is performed using isotonic regression:

@article{barlow1972isotonic,
  author = {Barlow, Richard and Bartholomew, David and Bremner, John and Brunk, H. Daniel},
  title = {Statistical Inference under Order Restrictions},
  year = {1972},
  publisher = {Wiley}
}

Principal Component Analysis (PCA)

@article{jolliffe2016pca,
  author = {Jolliffe, Ian T. and Cadima, Jorge},
  title = {Principal component analysis: a review and recent developments},
  journal = {Philosophical Transactions of the Royal Society A},
  year = {2016}
}

Factor Analysis

@book{harman1976factor,
  author = {Harman, Harry H.},
  title = {Modern Factor Analysis},
  year = {1976},
  publisher = {University of Chicago Press}
}

Community Detection (Modularity)

@article{newman2006modularity,
  author = {Newman, M. E. J.},
  title = {Modularity and community structure in networks},
  journal = {PNAS},
  year = {2006}
}

Software Dependencies

PyTorch

@article{paszke2019pytorch,
  author = {Paszke, Adam and Gross, Sam and Massa, Francisco and others},
  title = {PyTorch: An Imperative Style, High-Performance Deep Learning Library},
  journal = {NeurIPS},
  year = {2019}
}

scikit-learn

@article{pedregosa2011sklearn,
  author = {Pedregosa, Fabian and Varoquaux, Gaël and Gramfort, Alexandre and others},
  title = {Scikit-learn: Machine Learning in Python},
  journal = {JMLR},
  year = {2011}
}

NetworkX

@article{hagberg2008networkx,
  author = {Hagberg, Aric and Swart, Pieter and S Chult, Daniel},
  title = {Exploring Network Structure, Dynamics, and Function using NetworkX},
  journal = {SciPy Proceedings},
  year = {2008}
}

Notes

  • Cite only what is relevant to your use case.
  • If you use only the curve fitting and sensitivity scoring, citing CETSA + logistic modeling may be sufficient.
  • If you use sequence modeling, include the ESM reference.
  • If you use network or latent analysis, include the corresponding methodological references.

The goal is not to over-cite, but to accurately reflect the components you relied on.