Filopodia density and length in circular invasion assays correlate with reported degree of cancer cell malignancy. (A) MCF10A, MCF10AT, and DCIS.COM cells were left to migrate into GFR Matrigel for 14 d, fixed, stained, and imaged using a confocal microscope. Bars: (main) 50 µm; (inset) 20 µm. (B–E) MCF10A and DCIS.COM cells were plated in circular invasion assays and left to invade through GFR Matrigel, fibrillar collagen I, or media (no overlay) for 3 d. Cells were then fixed, stained for actin and DAPI, and imaged using an SDC microscope (100× objective, CMOS camera). For each condition, representative maximal z projection and magnified area (squares) are displayed. Bars: (main) 20 µm; (inset) 5 µm. Filopodia density was compared between cells (C; ***, P < 1.54 × 10⁻⁴) and in the same cells in different overlay conditions (D; **, P = 0.002; ***, P < 9.99 × 10⁻⁰⁵). Mean filopodia length was also calculated in each cell line (E; ***, P < 3.46 × 10⁻¹⁰). Results from three independent experiments are displayed as Tukey box plots (condition, fields of view analyzed; MCF10A no overlay, 43; MCF10A fibrillar collagen I, 31; MCF10A GFR Matrigel, 37; DCIS.COM no overlay, 30; DCIS.COM fibrillar collagen I, 73; DCIS.COM GFR Matrigel, 37). The Tukey box plots represent the median and the 25th and 75th percentiles (interquartile range); points are displayed as outliers (represented by dots) if 1.5 times above or below the interquartile range (represented by whiskers). Statistical analysis: Student’s t test (unpaired, two tailed, unequal variance). The filopodia detection images generated by FiloQuant are displayed in Fig. S2.