Optimization of classification and regression analysis of four monoclonal antibodies from Raman spectra using collaborative machine learning approach

TALANTA

卷:184

Article

Le, Laetitia Minh Mai^1,2  Kegl, Balazs³  Gramfort, Alexandre^3,5,6  Marini, Camille^3,4  Nguyen, David¹  Cherti, Mehdi^3,7  Tfaili, Sana²  Tfayli, Ali²  Baillet-Guffroy, Arlette²  Prognon, Patrice^1,2  Chaminade, Pierre²  Caudron, Eric^1,2 

[1] Hop Europeen Georges Pompidou, AP HP, Pharm Dept, F-75015 Paris, France

[2] Univ Paris Saclay, Univ Paris Sud, Grp Chim Analyt Paris Sud, FKA EA4041,Chim Analyt Pharmaceut,EA7357,Lip Sys2, F-92290 Chatenay Malabry, France

[3] Univ Paris Saclay, Paris Saclay Ctr Data Sci, F-91440 Orsay, France

[4] Ecole Polytech, CMAP, F-91128 Palaiseau, France

[5] INRIA, Parietal Team, F-91120 Palaiseau, France

[6] Telecom ParisTech, LTCI, F-75013 Paris, France

[7] CNRS, LAI, F-91440 Orsay, France

关键词: Machine learning;    Chemometrics;    Raman spectroscopy;    Monoclonal antibody;    Classification analysis;    Regression analysis;   

DOI  :  10.1016/j.talanta.2018.02.109

来源: Elsevier

【 摘 要 】

The use of monoclonal antibodies (mAbs) constitutes one of the most important strategies to treat patient suffering from cancers such as hematological malignancies and solid tumors. These antibodies are prescribed b; the physician and prepared by hospital pharmacists. An analytical control enables the quality of the preparation to be ensured. The aim of this study was to explore the development of a rapid analytical method for quality control. The method used four mAbs (Infliximab, Bevacizumab, Rituximab and Ramucirumab) at various con centrations and was based on recording Raman data and coupling them to a traditional chemometric and ma chine learning approach for data analysis. Compared to conventional linear approach, prediction errors an reduced with a data-driven approach using statistical machine learning methods. In the latter, preprocessing am predictive models are jointly optimized. An additional original aspect of the work involved on submitting th. problem to a collaborative data challenge platform called Rapid Analytics and Model Prototyping (RAMP). Thi allowed using solutions from about 300 data scientists in collaborative work. Using machine learning, the prediction of the four mAbs samples was considerably improved. The best predictive model showed a combines error of 2.4% versus 14.6% using linear approach. The concentration and classification errors were 5.8% an( 0.7%, only three spectra were misclassified over the 429 spectra of the test set. This large improvement obtainer with machine learning techniques was uniform for all molecules but maximal for Bevacizumab with an 88.34 reduction on combined errors (2.1% versus 17.9%).

【 授权许可】

Free