bioRxiv Science⌕ Search

Biology subjects

Goldberg, H.

Publications and source records attributed to Goldberg, H..

2 recordsLinked to original sources

A unified photosensitizer platform for in situ DNA, RNA, and protein directed proximity labeling

Cells depend on the spatial organization of proteins, RNA, and DNA into discrete subcellular compartments. Previous methods have largely centered on measuring spatial organization based on only one of these biomolecular classes at a time. Here, we demonstrate that POCA photocatalytic proximity labeling can serve as a unified photosensitizer-based platform for profiling the proximal proteomes of protein, RNA, and DNA targets within a single experimental framework. We show that POCA can harness standard immunofluorescence or in situ hybridization workflows to specifically target organic fluorophore photosensitizers to intracellular targets for proximity labeling in fixed cells. POCA-targeted proximity labeling requires minimal cellular input and does not require genetic engineering. Additionally, POCA photosensitizers are selected to also be fluorescent, enabling direct confirmation of on-target localization by imaging prior to proteomic analysis. To demonstrate broad utility, we apply POCA across multiple molecular targets spanning protein, RNA, and genomic DNA, including components of the nuclear pore complex, nucleolus, nuclear speckles, telomeres, and pericentromeric heterochromatin. By anchoring proximity labeling to both a protein and an RNA within the same nuclear compartment, we resolve shared and distinct proximal proteomes from orthogonal molecular perspectives.

cell biology↗

Neuronal components of evaluating the human 1 origin of abstract shapes

Communication through visual symbols is a key aspect of human culture. However, to what extent can people distinguish between human-origin and artificial symbols, and the neuronal mechanisms underlying this process are not clear. Using fMRI we contrasted brain activity during presentation of human-created abstract shapes and random-algorithm created shapes, both sharing similar low level features.\n\nWe found that participants correctly identified most shapes as human or random. The lateral occipital complex (LOC) was the main brain region showing preference to human-made shapes, independently of task. Furthermore, LOC activity was parametrically correlated to beauty and familiarity scores of the shapes (rated following the scan). Finally, a model classifier based only on LOC activity showed human level accuracy at discriminating between human-made and randomly-made shapes.\n\nOur results highlight the sensitivity of the human brain to social and cultural cues, and point to high-order object areas as central nodes underlying this capacity.

neuroscience↗