Identification of D- and L-phenylalanine enantiomeric mixtures by employing deep neural network models

Nigdelioglu E., Toprak E., Guney Akkurt M., Erol Barkana D., Kazanci M., UYAVER Ş., ...Daha Fazla

Journal of Molecular Structure, cilt.1304, 2024 (SCI-Expanded) identifier

  • Yayın Türü: Makale / Tam Makale
  • Cilt numarası: 1304
  • Basım Tarihi: 2024
  • Doi Numarası: 10.1016/j.molstruc.2024.137628
  • Dergi Adı: Journal of Molecular Structure
  • Derginin Tarandığı İndeksler: Science Citation Index Expanded (SCI-EXPANDED), Scopus, Academic Search Premier, Chemical Abstracts Core, Chimica, Compendex, INSPEC
  • Anahtar Kelimeler: Deep learning, Enantiomer, Phenylalanine, Pre-trained models, Self-assembly
  • Marmara Üniversitesi Adresli: Evet

Özet

Phenylalanine is an aromatic essential amino acid that exhibits the tendency to self-aggregate into fibrillar structures in its enantiomerically pure form. This observation was indicated as the underlying mechanism of phenylketonuria, which is a genetic condition associated with various neurological, physical, and developmental issues, characterized with phenylalanine buildup in the brain. The presence of D-phenylalanine was demonstrated previously to inhibit the formation of fibrils by L-phenlyalanine, indicating its potential use in phenylketonuria treatment. In this study, several combinations of D and L-phenylalanine were examined with the help of state-of-the-art deep learning methods for their fibril forming capacity, demonstrating the usefulness and accuracy of deep learning methods in distinguishing between different self-assembled structures.