Barekatain, M. | bioRxiv Science

bioRxiv2021

iDEP-assisted isolation of insulin secretory vesicles

Organelle heterogeneity and inter-organelle associations within a single cell contribute to the limited sensitivity of current organelle separation techniques, thus hindering organelle subpopulation characterization. Here we use direct current insulator-based dielectrophoresis (DC-iDEP) as an unbiased separation method and demonstrate its capability by identifying distinct distribution patterns of insulin vesicles from pancreatic {beta}-cells. A multiple voltage DC-iDEP strategy with increased range and sensitivity has been applied, and a differentiation factor (ratio of electrokinetic to dielectrophoretic mobility) has been used to characterize features of insulin vesicle distribution patterns. We observed a significant difference in the distribution pattern of insulin vesicles isolated from glucose-stimulated cells relative to unstimulated cells, in accordance with functional maturation of vesicles upon glucose stimulation, and interpret this to be indicative of high-resolution separation of vesicle subpopulation. DC-iDEP provides a path for future characterization of subtle biochemical differences of organelle subpopulations within any biological system.

biophysics↗

bioRxiv2017

Modeling positional effects of regulatory sequences with spline transformations increases prediction accuracy of deep neural networks

MotivationRegulatory sequences are not solely defined by their nucleic acid sequence but also by their relative distances to genomic landmarks such as transcription start site, exon boundaries, or polyadenylation site. Deep learning has become the approach of choice for modeling regulatory sequences because of its strength to learn complex sequence features. However, modeling relative distances to genomic landmarks in deep neural networks has not been addressed.\n\nResultsHere we developed spline transformation, a neural network module based on splines to flexibly and robustly model distances. Modeling distances to various genomic landmarks with spline transformations significantly increased state-of-the-art prediction accuracy of in vivo RNA-binding protein binding sites for 114 out of 123 proteins. We also developed a deep neural network for human splice branchpoint based on spline transformations that outperformed the current best, already distance-based, machine learning model. Compared to piecewise linear transformation, as obtained by composition of rectified linear units, spline transformation yields higher prediction accuracy as well as faster and more robust training. As spline transformation can be applied to further quantities beyond distances, such as methylation or conservation, we foresee it as a versatile component in the genomics deep learning toolbox.\n\nAvailabilitySpline transformation is implemented as a Keras layer in the CONCISE python package: https://github.com/gagneurlab/concise. Analysis code is available at goo.gl/3yMY5w.\n\nContactavsec@in.tum.de; gagneur@in.tum.de

bioinformatics↗