Autoquant x

For quantification of signal number and intensity, the ImageJ Software was used¹. Puncta were counted along dendrites and puncta density was calculated as puncta per dendrite length. Puncta intensity was measured likewise and shown as relative puncta density normalized to control values.
For the analysis of dendritic spines we deconvolved the RFP signals (F-actin visualized by LifeAct) using the AutoQuant X software (MediaCybernetics). The reconstructed model of dendrites and spines was designed with the Filament Tracer software (Imaris, Bitplane) using default settings. Spines were reconstructed and their length was analyzed with the software. Spine classification was subsequently determined as follows: spines were classified into four categories by the following settings: “Mushroom” (spine length < 2 μm; spine mean width > 0.5 μm; spine neck length > 0.2 μm), “Thin” (spine length < 2 μm; spine mean width < 0.5 μm), “Stubby” (spine length < 2 μm; spine mean width > 0.5 μm; spine neck length < 0.2 μm), and “Filopodia” (spine length > 2 μm; spine mean width < 0.5 μm).

To quantify the 3D position of FISH-labeled genomic loci within the nucleus, cell images were first taken as z-stacks on a widefield microscope (Nikon TE-2000 microscope equipped with a 1.45 NA 100 × objective, Sedat quad filter set, PIFOC Z-axis focus drive from Physik Instruments, and CoolSnapHQ High Speed Monochrome CCD camera from Photometrics, all run by MetaMorph image acquisition software) with a 0.2 μm spacing in z. Images of Alexa 488-labeled loci and DAPI-labeled nuclei were deconvolved using AutoQuant X (Media Cybernetics, United Kingdom).
The z stacks containing the alleles in focus were manually thresholded, and the distance of the locus to the nearest point on the nuclear periphery labeled with antibodies to Lamin A/C (rabbit polyclonal 3262) (Schirmer et al., 2001 (link)) was scored using ImageJ. Only nuclei with three or more labeled alleles were used for analysis in order to be able to account for allelic exclusion type of phenotypes, if any. Statistics were performed using Mann–Whitney test (2 groups) or Kruskal–Wallis ANOVA (> 2 groups) for non-parametric data. Data are presented as scatters overlaid with the median and interquartile range and taken as statistically significant at p < 0.05.