Improved resolution of chromatographic peak analysis using multi-snapshot imaging

Abstract

Snapshot imaging has a number of advantages in automated gel electrophoresis compared with the finish-line method in capillary electrophoresis, although at the expense of resolution. This paper presents a novel signal processing algorithm enabling a multi-capture imaging modality which improves resolution. The approach takes multiple snapshots as macromolecules are electrophoresed. Peaks from latter snapshots have higher resolution but poor signal-to-noise ratio (SNR), while peaks from earlier snapshots have lower resolution but better SNR. Signals at different capture-times are related by a scale-in-separation, amplitude scaling, and an arbitrary shift. The multiple captures are realigned and fused together using least-squares estimation and a physically inspired signal model. Since partial waveforms are observed as the chromatic peaks exit the sensor's field-of-view, this is accounted for in the realignment algorithm. The proposed technique yields improved resolution, improved fragment concentration and size estimates, and allows the removal of static background noise.