Blind demixing methods for recovering dense neuronal morphology from barcode imaging data

• DOI: 10.1371/journal.pcbi.1009991
• Publication Date: 2022-04-08T17:38:18Z
• Authors: Shuonan Chen; Jackson Loper; Pengcheng Zhou; Liam Paninski
• Keywords: Barcode

Cellular barcoding methods offer the exciting possibility of ‘infinite-pseudocolor’ anatomical reconstruction—i.e., assigning each neuron its own random unique barcoded ‘pseudocolor,’ and then using these pseudocolors to trace microanatomy neuron. Here we use simulations, based on densely-reconstructed electron microscopy microanatomy, with signal structure matched real data, quantify feasibility this procedure. We develop a new blind demixing approach recover barcodes that label neuron, validate method data known barcodes. also neural network which uses recovered reconstruct neuronal morphology from observed fluorescence imaging ‘connecting dots’ between discontiguous barcode amplicon signals. find accurate recovery should be feasible, provided density is sufficiently high. This study suggests mapping projection pattern many individual neurons simultaneously, at high resolution large scale, via conventional light microscopy.