Figure 2

Improvement of measurement by reducing the neighbor effect. (A) Position of crowded (row = 11-21 × column = 18-31) and uncrowded (row = 1, 32 × column = 1, 48 without corner) colonies. (B) Growth of colonies in two positions was monitored to assess the neighbor effect. The average (line) and SD (gray area) of wild-type colony growth in the two positions are shown. Colony growth was quantified by integrating the mass of the whole (mass) or the center (mass*) region of each colony. (C) Growth values were determined by the conventional method (CONV; white bar) and Colony-live (MGR; gray bars), and the MGR values were calculated based on the mass (MGR-mass) and mass* (MGR-mass*) values. The coefficient of variation (CV) of growth values for 1536 wild-type colonies within a plate was determined using 5 independently prepared plates. (D) Intra-class correlation (ICC) of multiple growth measurements for the Keio mutant collection was evaluated in two experimental designs: the colony array format of the Keio collection was the same (Same array, 7 independent experiments for entire set of Keio collection) or randomly shuffled (Shuffle array, randomly 4 plates) for each measurement. Error bars represent 95% confidence intervals.