SUN’IY INTELLEKT YORDAMIDA TIBBIY KIMYONI O‘QITISH VA O‘RGANISH JARAYONINI INTEGRATSIYALASHGAN MODELI

Komilov S. N. & Tojiboyeva I. M.


Annotatsiya

Ushbu maqolada farmatsevtika ta’limi va dori vositalarini loyihalash sohasining asosi bo‘lgan tibbiy kimyo fanining zamonaviy o‘qitish va o‘rganish modeli tahlil qilinadi.Tadqiqotning maqsadi — sun’iy intellekt (SI) texnologiyalarini tibbiy kimyo fanini o‘qitish tizimiga integratsiya qilishning yangi modelini ishlab chiqish va asoslashdir.Maqolada retsidiv neyron tarmoqlari (RNN), generativ-adversal tarmoqlar (GAN) va Monte-Karlo daraxt qidiruvi (MCTS) kabi zamonaviy algoritmlarning talabalar tomonidan molekulyar dizayn va retrosintezni o‘rganishdagi o‘rni yoritiladi. Shuningdek, model tarkibida kvant-kimyoviy hisob-kitoblar va molekulyar modifikatsiyalarning (masalan, timin analoglari misolida) organizmga ta’sirini bashorat qilish metodikasi taklif etilgan. Tadqiqot natijalari shuni ko‘rsatadiki, SI yordamida integratsiyalashgan o‘qitish modeli talabalarda dori vositalarining ADMET (so‘rilish, taqsimlanish, metabolizm, ekskretsiya va toksiklik) xossalarini “in silico” sharoitida aniqlash ko‘nikmalarini shakllantiradi. Ushbu model farmatsevtika sanoatining "Design-Make-Test-Analyze" (DMTA) siklini o‘quv jarayoniga tatbiq etish orqali kadrlar tayyorlash sifatini oshirishga xizmat qiladi.


Adabiyotlar

  1. Olivecrona M, Blaschke T, Engkvist O, Chen H. Molecular de-novo design through deep reinforcement learning. J. Cheminform. 9(1),1–14 (2017).
  2. Artificial Intelligence in Drug Discovery" (Muallif: Nathan Brown) "Applied Chemoinformatics: Achievements and Future Opportunities" (Mualliflar: Thomas Engel, Johann Gasteiger). Nature Machine Intelligence va Journal of Chemical Information and Modeling jurnallaridagi so'nggi maqolalar.
  3. Vamathevan, J., et al. (2019). "Applications of machine learning in drug discovery and development". Nature Reviews Drug Discovery. (Sohadagi eng nufuzli maqolalardan biri).
  4. Duvenaud, D., et al. (2015). "Convolutional networks on graphs for learning molecular fingerprints". NIPS. (GCNlarning asos soluvchi maqolasi).
  5. Tsubaki, H., et al. (2019). "Compound–protein interaction prediction with end-to-end learning of neural networks for graphs and sequences". Bioinformatics.
  6. Gayvert, K. M., et al. (2016). "A Data-Driven Approach to Predicting Toxicity of Compounds in Development". PLOS Computational Biology.
  7. Kocsis, L., & Szepesvári, C. (2006). "Bandit based Monte-Carlo Planning". ECML. (UCT algoritmini MCTS ga integratsiya qilgan asosiy ilmiy ish).
  8. Segler, M. H. S.,Preuss, M.,&Waller, M. P. (2018). Planning chemical syntheses with deep neural networks and symbolic AI. Nature, 555(7698), 604-610.
  9. Xiong, Z., et al. (2021). ADMETlab 2.0: an integrated platform for predictions of ADMET properties. Nucleic Acids Research, 49(W1), W5-W14.
  10. Jiménez-Luna, J., Grisoni, F., & Schneider, G. (2020). Drug discovery with explainable artificial intelligence. Nature Machine Intelligence, 2(10), 573-584.