研究実績の概要 |
To find new natural product-derived antibiotics, the sizable empirical data generated from thousands of years of anti-infection practice of Traditional Chinese Medicine (TCM) should be fully utilized. By mapping the relevant networks including involving bacteria, antibiotics, TCM syndromes, formulae, natural products ingredients and metabolites, TCM formulae significant in the treatment of lung diseases were screened out and divided into binary groups according to whether their corresponding TCM syndromes reveal potential effects or not. Overall, a natural product ingredient-level TCM formula dataset were constructed, and supervised learning are carried out on whether the formula samples own the potential to fight infections in bacterial pneumonia. Regression and classification algorithms verified the validity of the dataset, and the features of natural product ingredient that are significant for efficacy were extracted from the maching learning models. The candidate natural products as new antibacterical agent were identified and validated by published literature and KNApSAcK Metabolite Activity database. The remains for which we predict novel associations with antimicrobial efficacy might be considered as candidates in the early step of the new antibiotic discovery cycle. In parallel, we developed a method for Antibacterial Activity Prediction of Plant Secondary Metabolites Based on a Combined Approach of Graph Clustering and Deep Neural Network utilizing data from KNApSAcK database and published our work in a journal
|
次年度使用額が生じた理由 |
Because of Covid-19, we could not attend an international conference last year so we transferred the money to current year. We are planning to attend The 44th Annual International Conference of the IEEE Engineering in Medicine and Biology Societye (EMBC), Glasgow, UK this year. Also the money will be utilized for publication costs of three journal papers we are planning to publish this year,
|