Multi-manufacturer drug identification based on near infrared spectroscopy and deep transfer learning

Near infrared(NIR)spectrum analysis technology has outstanding advantages such as rapid,nondestructive,pollution-free,and is widely used in food,pharmaceutical,petrochemical,agricultural products production and testing industries.Convolutional neural network(CNN)is one of the most successful methods in big data analysis because of its powerful feature ex-traction and abstraction ability,and it is especially suitable for solving multi-classification problems.CNN-based transfer learning is a machine learning technique,which migrates para-meters of trained model to the new one to improve the performance.The transfer learning strategy can speed up the learning efficiency of the model instead of learning from scratch.In view of the difficulty in acquisition of drug NIR spectral data and high labeling cost,this paper proposes three simple but very effective transfer learning methods for multi-manufacturer identification of drugs based on one-dimensional CNN.Compared with the original CNN,the transfer learning method can achieve better classification performance with fewer NIR spectral data,which greatly reduces the dependence on labeled NIR spectral data.At the same time,this paper also compares and discusses three different transfer learning methods,and selects the most suitable transfer learning model for drug NIR spectral data analysis.Compared with the current popular methods,such as SVM,BP,AE and ELM,the proposed method achieves higher classification accuracy and scalability in multi-variety and multi-manufacturer NIR spectrum classification experiments.

Identification of a dihydroorotate dehydrogenase inhibitor thatinhibits cancer cell growth by proteomic profiling

Dihydroorotate dehydrogenase(DHODH)is a central enzyme of the de novo pyrimidine biosynthesis pathway and is a promising drug target for the treatment of cancer and autoimmune diseases.This study presents the identification of a potent DHODH inhibitor by proteomic profiling.Cell-based screening revealed that NPD723,which is reduced to H-006 in cells,strongly induces myeloid differentiation and inhibits cell growth in HL-60 cells.H-006 also suppressed the growth of various cancer cells.Proteomic profiling of NPD723-treated cells in ChemProteoBase showed that NPD723 was clustered with DHODH inhibitors.H-006 potently inhibited human DHODH activity in vitro,whereas NPD723 was approximately 400 times less active than H-006.H-006-induced cell death was rescued by the addition of the DHODH product orotic acid.Moreover,metabolome analysis revealed that H-006 treatment promotes marked accumulation of the DHODH substrate dihydroorotic acid.These results suggest that NPD723 is reduced in cells to its active metabolite H-006,which then targets DHODH and suppresses cancer cell growth.Thus,H-006-related drugs represent a potentially powerful treatment for cancer and other diseases.

Exploring of drug leads from diversity-oriented Michael-acceptor library derived from natural products

A potential strategy for drug lead identification and in-active natural products re-discovery is elaborated.Starting from fifteen structurally diverse natural products,a focused library featured by Michael acceptors is constructed with IBX mediated oxidation.Biological assay on five tumor cell lines indicates that four Michael acceptors,8a,11a,12a,14a,are with improved cytotoxicity(3-10 folds more potent than the parent compounds),which merit further investigations.Further thiol-sensitive assay of the active hit 8a revealed that it was an irreversible Michael acceptor.The results suggest that the strategy is not only effective and relatively high discovery rate(28%),but also resource saving.

Study on the Quality Standard of Heracleum moellendorffii Hance.

[Objectives]This study was conducted to investigate the quality standard of Heracleum moellendorffii Hance.,so as to provide data support for the 2003 version of Quality Standards for Chinese Medicinal Materials and Ethnic Medicinal Materials in Guizhou Province.[Methods]H.moellendorffii was subjected to character identification and microscopic identification.Referring to Chinese Pharmacopoeia,TCL thin-layer identification and extract determination were carried out on 10 batches of H.moellendorffii from 6 habitats in Guizhou.[Results]The characters and microscopic characteristics of H.moellendorffii were described in detail.The TLC thin-layer identification spots were clear,with strong specificity.The limit of the medicinal material extract was determined to be not less than 22.0%.[Conclusions]This study can provide data support for the quality evaluation and standard improvement of the medicinal material H.moellendorffii.

Research Progress of the Wild Medicinal Plant,Pinellia ternata

Based on literature reviews and analysis of research reports on Pinellia ternata found locally and abroad in recent years,this article summarizes and arranges them.The research on Pinellia ternata mainly focuses on its cultivation,tissue culture,and so on.There are only a few research on its active components and its regulation mechanism.The wild resources of Pinellia ternata are gradually decreasing,hence it is urgent to take effective measures to protect these wild resources as well as to establish germplasm resources bank and nursery.In order to meet the needs of the domestic market,it is necessary to investigate the distribution of wild Pinellia ternata resources,explore the best growing environment and conditions,artificially cultivate Pinellia ternata,as well as implement resource industrialization,sustainable development,and utilization.