Extended Data Fig. 2: Spatial segregation analysis across platforms.
From: Optimizing multiplexed imaging experimental design through tissue spatial segregation estimation
(a) Distribution of the number of individual cells in regions of the size of Visium capture spots in a lymph node (left) and a breast cancer sample (right). (b) Relationship between the threshold T parameter and the τ value for a FoV’s width of 400 µm in the IMC lymph node sample (1). The dashed line corresponds to an ordinary least square regression. (c) Relationship between the threshold T parameter and the α value for the IMC lymph node sample (1). (d) Plot of α values from healthy (n = 7) and tumor (n = 15) samples. The p-value was computed using a two-sided Mann-Whitney rank test. Large bars correspond to the median and small bars to the IQR. (e) Sampling analysis of the first CosMX lung sample replicate. Each dot corresponds to the mean number of clusters recovered vs number of sampled regions for FoV widths ranging from 200 to 400 µm and vertical bars correspond to the standard error. (f) Relationship between τ and w for the CosMX lung dataset. The dashed line corresponds to the linear regression after log10 transform. (g) Approach used to estimate α from a set of small IMC FoVs. (h) Relationship of cell phenotype spatial segregation with α. Each panel represents a different Visium sample where each point corresponds to a capture spot and is colored based on its associated cell phenotype.
