EUVIMEDEUVIMED entdecken
Open Access · CC BY 4.0

Combining Molecular Dynamics and Machine Learning to Predict Drug Resistance Causing Variants of BRAF in Colorectal Cancer

Longsheng Xie, Christopher Lockhart iD, Dmitri K Klimov, Mohsin Saleet Jafri iD

Molecules · 2025

DOI: 10.3390/molecules30173556 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Molecules, 2025. Verfasst von Longsheng Xie und 3 weiteren Autor:innen. Beteiligte Einrichtung: School of Systems Biology, George Mason University, Fairfax, VA 22020, USA. Hauptthemen laut MEDLINE: Proto-Oncogene Proteins B-raf, Colorectal Neoplasms, Machine Learning, Molecular Dynamics Simulation. Die Arbeit wurde bisher 5-mal zitiert (Crossref) und nennt 36 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0).

AbstractSchlagwörterBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

The BRAF protein regulates cell growth and division through key signaling pathways. Mutations in BRAF, particularly the V600E variant, are frequently observed in colorectal cancer (CRC) and are associated with poor prognosis and therapeutic challenges. Tumors harboring certain BRAF mutations often exhibit primary resistance to BRAF inhibitor monotherapies. Over time, these tumors can also develop acquired resistance, further complicating treatment. In this study, we employed replica exchange molecular dynamics simulations combined with machine learning techniques to investigate the structural alterations induced by BRAF mutations and their contribution to drug resistance. Our analyses revealed that conformational changes in mutant BRAF proteins associated with dabrafenib residues psi494, phi600, phi644, phi663, psi675, and phi677 were sufficient for classifying drug-resistant vs. drug-sensitive variants. Similarly, for vemurafenib, residues psi450, phi484, phi495, phi518, psi622, and phi622 were the key residues that influence drug binding and resistance mechanisms. These residues are located in the N-lobe of CR3, which is responsible for ATP binding and the regulation of BRAF kinase activity. These findings offer deeper insights into the molecular basis of BRAF-driven resistance and provide predictive models for phenotypic outcomes of various BRAF mutations. The study underscores the importance of targeting specific BRAF variants for more effective, personalized therapeutic strategies in drug-resistant CRC patients.

Schlagwörter

MeSH-Schlagwörter (MEDLINE)

Proto-Oncogene Proteins B-rafColorectal NeoplasmsMachine LearningMolecular Dynamics SimulationDrug Resistance, NeoplasmHumansMutationProtein Kinase InhibitorsVemurafenibAntineoplastic AgentsImidazoles

Stichwörter der Autor:innen

BRAFcolorectal cancermachine learningmolecular dynamic simulation

Substanzen

Proto-Oncogene Proteins B-rafBRAF protein, humanProtein Kinase InhibitorsVemurafenibAntineoplastic AgentsImidazoles

Autor:innen und Institutionen

  • School of Systems Biology, George Mason University, Fairfax, VA 22020, USA.
  • Center for Biomedical Engineering and Technology, University of Maryland School of Medicine, Baltimore, MD 21201, USA.

ORCID-Profile: Christopher Lockhart, Mohsin Saleet Jafri

Zitieren

APA

Xie, L., Lockhart, C., Klimov, D. K., & Jafri, M. S. (2025). Combining Molecular Dynamics and Machine Learning to Predict Drug Resistance Causing Variants of BRAF in Colorectal Cancer. Molecules. https://doi.org/10.3390/molecules30173556

Vancouver

Xie L, Lockhart C, Klimov DK, Jafri MS. Combining Molecular Dynamics and Machine Learning to Predict Drug Resistance Causing Variants of BRAF in Colorectal Cancer. Molecules. 2025. doi:10.3390/molecules30173556

BibTeX
@article{Xie2025,
  author = {Xie, Longsheng and Lockhart, Christopher and Klimov, Dmitri K and Jafri, Mohsin Saleet},
  title = {Combining Molecular Dynamics and Machine Learning to Predict Drug Resistance Causing Variants of BRAF in Colorectal Cancer},
  journal = {Molecules},
  year = {2025},
  doi = {10.3390/molecules30173556}
}
Kostenlos starten →