Scientists at RUDN University, in collaboration with researchers from China, employed various machine learning models to identify a cluster of potential medications that hinder the enzyme responsible for unbridled cell division. Their findings were disclosed in the journal Biomedicines.

Potential Anticancer Medications

Cyclin-dependent kinase 2 (CDK 2) is an enzyme pivotal in regulating cell division. While it isn’t crucial in normal cells, it plays a critical role in the unchecked proliferation of cancer cells. Curtailing the activity of CDK 2 restrains tumor growth, making the search for effective CDK 2 inhibitors significant. RUDN University scientists and their Chinese counterparts utilized computational research techniques, combining machine learning and molecular modeling, ultimately uncovering several promising inhibitors.

“Cyclin-dependent kinase 2 holds great promise in the realm of cancer treatment, and the creation of inhibitors for it is crucial in the field of anti-tumor therapy. Although the enzyme’s role in tumor formation is not fully understood, inhibiting it has proven beneficial in cancer treatment. While some inhibitors have undergone clinical trials, a selective inhibitor specific to this enzyme is yet to be found,” explained Alexander Novikov, Ph.D. in Chemistry and senior researcher at RUDN University’s Joint Institute of Chemical Research.

To identify a potential drug candidate, scientists applied machine learning methods. Multiple models were constructed to pinpoint active CDK 2 inhibitors, and a molecular model was developed using the molecular docking method, identifying the most favorable molecular orientation for a stable complex formation.

Machine learning models accurately pinpointed 25 potential active CDK 2 inhibitors with a 98% success rate. Chemists then subjected each compound to molecular docking tests, revealing three substances with optimal performance. These top three compounds underwent computer simulations using the molecular dynamics method and were compared with the reference compound dalpiciclib. All three demonstrated greater stability and compactness than the control drug.

“Compared to the reference drug dalpiciclib, the three computed compounds displayed enhanced stability and a more compact nature. Despite these promising outcomes, our study has certain limitations. Comprehensive in vitro and in vivo clinical trials are necessary to verify inhibitory activity and potential therapeutic efficacy. The development of these drugs will require an examination of the compounds’ impact on off-target interactions and their toxicity,” added Alexander Novikov.

Other posts

  • AI-Powered Human Resources
  • Emotion Recognition with Machine Learning
  • Google Introduces The Open Source Framework Firebase Genkit
  • AI Innovations Presented at Google I/O
  • Researching Genetic Algorithms
  • Yelp Introduces an Innovative AI Assistant to Facilitate Business Connections
  • The Intersection of AI and Machine Learning in Financial Services
  • Anomaly Detection Using Machine Learning Algorithms
  • Meta Debuts The Latest AI Chip In A Competitive Technology Sprint
  • An Innovative Machine Learning Approach for Analyzing Material Surfaces