dbscan function of matlab r2020b  (MathWorks Inc)

Journal: iScience

Article Title: Flat clathrin lattices are linked to metastatic potential in colorectal cancer

doi: 10.1016/j.isci.2023.107327

Figure Lengend Snippet: Cluster identification and classification workflow illustrated in a metastatic KM12L4a cell Individual localizations (ii) that make up the PALM-TIRF image (i) were clustered using the DBSCAN algorithm with minimum 17 neighbors (MinPts) and search radius (ε) 0.55. Retaining clusters with ≥50 localizations resulted in identification of individual clathrin clusters (iii). The designed cluster classification model (iv and v) classifies these as either a classical CCS (blue, indicated as CCP) or an alternative FCL (pink) based on 5 cluster-specific parameters (cluster area, number of localizations NrPts, eccentricity, perimeter, and the distance to the nearest neighboring cluster MinDistance) and a global threshold. Scale bars 5 μm, except scale bars of enlargements 1 μm.

Article Snippet: Then, the MATLAB DBSCAN function was executed for a minimum of 17 neighbors (MinPts) within a search radius (ε) of 0.55.

Techniques:

Journal: iScience

Article Title: Flat clathrin lattices are linked to metastatic potential in colorectal cancer

doi: 10.1016/j.isci.2023.107327

Figure Lengend Snippet:

Article Snippet: Then, the MATLAB DBSCAN function was executed for a minimum of 17 neighbors (MinPts) within a search radius (ε) of 0.55.

Techniques: Recombinant, Software, Imaging, Microscopy, Fluorescence