Flow cytometric analysis using digital signal processing.

Abstract

Current commercial flow cytometers employ analog circuits to produce the feature values of the pulse waveforms that result from particle analysis. The use of analog pulse processing limits the features that can be measured to pulse integral, pulse height, and pulse width, and a large amount of potentially relevant information about the shape of the pulse waveform is lost. Direct digitizing of the waveform provides a means for the extraction of additional features, for example, pulse skewness and kurtosis, as well as the Fourier properties of the pulse. Here we describe a digital pulse waveform processing system that is compatible both with a commercial flow cytometer, and with a readily available computational platform. The performance of the digital and analog systems were compared through analysis of synthetic waveforms, and the waveforms produced by standard fluorescence microspheres and biological particles. The digital waveform processing system was found to be accurate and flexible, and the value of several of its unique attributes was demonstrated using biological cells. A protocol was designed in which digital pulse processing provided a means for the quantitative monitoring of the optical alignment of the flow cytometer. It was shown that digital pulse processing could be used to discriminate between particle classes which produce feature values indistinguishable through analog pulse processing, and to discriminate accurately single cells from doublets and larger aggregates.