Machine learning architecture to predict drug response based on cancer cell FLIM images

With different states of two intrinsic fluorophores, nicotinamide adenine dinucleotide (phosphate) (NAD(P)H) and flavin adenine dinucleotide (FAD), we developed a single-layer autoencoder (AE) for feature extraction, which outputs condensed features representing the full metabolic FLIM information with lower dimensionality. We also described distributions of AE features and fluorescence lifetime redox ratio (FLIRR) from single cells by Gaussian mixture models (GMM), and predicted the values of FLIRR based on feature data from each time point for the HeLa cell lines and Caucasian-American (LNCaP) prostate cancer cell lines by the polynomial regression model and the random forest regression model.

[1]  Ata Mahjoubfar,et al.  Deep Learning in Label-free Cell Classification , 2016, Scientific Reports.

[2]  Amedeo Napoli,et al.  Feature Selection Methods for Early Predictive Biomarker Discovery Using Untargeted Metabolomic Data , 2016, Front. Mol. Biosci..

[3]  Cornelis W. Oosterlee,et al.  Generalization in fully-connected neural networks for time series forecasting , 2019, J. Comput. Sci..

[4]  Amy Loutfi,et al.  A review of unsupervised feature learning and deep learning for time-series modeling , 2014, Pattern Recognit. Lett..

[5]  Geoffrey E. Hinton,et al.  Reducing the Dimensionality of Data with Neural Networks , 2006, Science.

[6]  Daisuke Komura,et al.  Machine Learning Methods for Histopathological Image Analysis , 2017, Computational and structural biotechnology journal.

[7]  DeLiang Wang,et al.  Unsupervised Learning: Foundations of Neural Computation , 2001, AI Mag..

[8]  Hongxun Yao,et al.  Auto-encoder based dimensionality reduction , 2016, Neurocomputing.

[9]  Horst Wallrabe,et al.  Segmented cell analyses to measure redox states of autofluorescent NAD(P)H, FAD & Trp in cancer cells by FLIM , 2018, Scientific Reports.

[10]  Robert M. Clegg,et al.  Flim Microscopy in Biology and Medicine , 2009 .

[11]  F. Massey.  The Kolmogorov-Smirnov Test for Goodness of Fit , 1951 .

[12]  B. Hess,et al.  Metabolic control mechanisms. 5. A solution for the equations representing interaction between glycolysis and respiration in ascites tumor cells. , 1960, The Journal of biological chemistry.

[13]  Carla E. Brodley,et al.  Feature Selection for Unsupervised Learning , 2004, J. Mach. Learn. Res..

[14]  Melissa C Skala,et al.  Fluorescence lifetime imaging microscopy: fundamentals and advances in instrumentation, analysis, and applications , 2020, Journal of biomedical optics.

[15]  Julio Saez-Rodriguez,et al.  Machine Learning Prediction of Cancer Cell Sensitivity to Drugs Based on Genomic and Chemical Properties , 2012, PloS one.

[16]  Geoffrey E. Hinton,et al.  Rectified Linear Units Improve Restricted Boltzmann Machines , 2010, ICML.

[17]  H. Bourlard,et al.  Auto-association by multilayer perceptrons and singular value decomposition , 1988, Biological Cybernetics.

[18]  Miroslava Cuperlovic-Culf,et al.  Machine Learning Methods for Analysis of Metabolic Data and Metabolic Pathway Modeling , 2018, Metabolites.

[19]  Binlin Wu,et al.  Identifying metastatic ability of prostate cancer cell lines using native fluorescence spectroscopy and machine learning methods , 2021, Scientific Reports.

[20]  Horst Wallrabe,et al.  Investigation of Mitochondrial Metabolic Response to Doxorubicin in Prostate Cancer Cells: An NADH, FAD and Tryptophan FLIM Assay , 2017, Scientific Reports.