Fig. 5: DeepCellMap applied to SARS-CoV-2 (COVID19) fetal human brain confocal data. | Nature Communications

Fig. 5: DeepCellMap applied to SARS-CoV-2 (COVID19) fetal human brain confocal data.

From: Unraveling microglial spatial organization in the developing human brain with DeepCellMap, a deep learning approach coupled with spatial statistics

Fig. 5

A 3 channels : Laminin (blood vessels), IBA1 (microglia/macrophages), and CD68 (Lysosomal marker). White scale bar = 10 μm. B Cells are classified according to the channels from which they are segmented. We segmented four types of objects: blood vessels from channel Laminin (green), microglia from channel IBA1 (white), lysosomal cells from CD68 channel (red) and phagocytic microglia from the superposition of CD68 and IBA1 channels (n = 3) controls and n = 4 SARS-CoV-2 exposed sample. Error bars correspond to standard errors (all values are available in Source Data). White scale bar = 10 μm. C Coupling frequency and mean distances. D Fraction of isolated vs clustered cells computed from the DBSCAN algorithm. E Fraction of (A) cells having (B) cells as first neighbors. Differences between control and SARS-CoV-2 are represented in the matrix. Source data are provided as a Source Data file.

Back to article page