Cell Line Classification Using Electric Cell-substrate Impedance Sensing (ECIS)
We consider cell line classification using multivariate time series data obtained from Electric Cell-substrate Impedance Sensing (ECIS) technology. The ECIS device, which monitors the attachment and spreading of mammalian cells in real time through the collection of electrical impedance data, has historically been used to study one cell line at a time. We greatly broaden the scope of ECIS by considering simultaneous tracking of multiple cell lines in a search for new methods that can help mitigate the current reproducibility crisis in the biological literature. Our approach capitalizes on the multi-frequency data ECIS provides, which have been underutilized in previous studies. We consider classification of fifteen different mammalian cell lines, construct a dictionary of 29 different characteristic features, and obtain 95 out-of-sample. Our preliminary findings provide a baseline for future large-scale studies in this field.
READ FULL TEXT