This repository has been archived by the owner on Oct 18, 2021. It is now read-only.
-
Notifications
You must be signed in to change notification settings - Fork 1
gpseq_anim run
Gabriele Girelli edited this page Aug 23, 2018
·
2 revisions
The most basic way of running gpseq_anim
, which uses the default values of all optional parameters, requires only the path to an input directory (see the input page for details on the proper input directory structure) and of an output directory (created if not present). For example:
gpseq_anim /home/ggirelli/Data/input /home/ggirelli/Data/output
Run gpseq_anim
with no parameters for a brief recap of all its options. Run it with -h
for a detailed help page.
usage: gpseq_anim [-h]
[--skip [{inst,seg,an,box,plot,report} [{inst,seg,an,box,plot,report} ...]]]
[-l log] [-a Z Y X] [-U unit] [-d dna_name [dna_name ...]]
[-s sig_name [sig_name ...]] [-z min_z] [-R min_radius]
[--seg-type {sum_proj,max_proj,3d}] [-m folder] [-M prefix]
[-2 MAN2DDIR] [--an-type {sum_proj,max_proj,3d,mid}]
[--mid-type {central,largest,maxIsum}]
[--dist-type {center_max,center_percentile,diffusion}]
[--nuclear-sel [{size,surf,shape,sumI,meanI,flat_size,flat_sumI} [{size,surf,shape,sumI,meanI,flat_size,flat_sumI} ...]]]
[--sigma-smooth sigmaValue] [--sigma-density sigmaValue]
[--description [DESCRIPTION [DESCRIPTION ...]]]
[-t nthreads] [--note NOTE] [--regexp REGEXP]
[--nbins NBINS] [--no-hole-filling] [--labeled]
[--compressed] [-r] [-n] [-u] [-y] [--version]
inDir outDir
-
-a
to specify the voxel sides in Z/Y/X order (space-separated). This parameter is critical for proper isotropic distance calculations in 3D. -
-d
to specify the DNA stain channel. Multiple channels can be specified. For each series, the specified channels are matched agains the series channels and the first match is used. -
-s
to specify the GPSeq signal channel. Multiple channels can be specified. For each series, the specified channels are matched agains the series channels and the first match is used. -
--seg-type
to specify how to segment the nuclei based on the DNA stain channel. Available segmentation types are:-
sum_proj
for 2D segmentation on the sum Z-projection. -
max_proj
for 2D segmentation on the max Z-projection. -
3d
for 3D segmentation.
-
-
--an-type
to specify how to analyze each nucleus. Available analysis types are:-
sum_proj
to analyze the gradients in the nucleus sum Z-projection (after masking). -
max_proj
to analyze the gradients in the nucleus max Z-projection (after masking). -
3d
to analyze the gradients in the 3D nuclear volume. When this option is selected, a partial volume analysis is also performed, when only the top half of the volume is considered. This analysis mode is available only if the segmentation was performed with3d
mode. -
mid
to analyze the gradients in the mid-section of the 3D nuclear volume. This analysis mode is available only if the segmentation was performed with3d
mode. The mid-section can be defined in differnet ways, check the--mid-type
parameter for more details.
-
-
--mid-type
to change how the mid-section is identified when using--seg-type 3d --an-type mid
. Available mid-section definitions are:-
central
to select the slice in the middle of the nucleus Z-stack. -
largest
to select the nuclear slice with the largest area. -
maxIsum
to select the nuclear slice with the largest intensity integral.
-
-
--dist-type
to specify how to calculate the lamina distance.center_percentile
(default) to define the nuclear center as the 1% voxels with the highest distance from the nuclear lamina (mask border).center_max
regards only the top voxel(s) as center.diffusion
simulates enzyme diffusion (assuming and homogeneously dense nucleus) with an iterative anisotropic Gaussian blurring and uses the time at which the enzyme reaches a voxel as proxy for the "distance" from lamina.
-
--nuclear-sel
to specify the features used to sub-select nuclei. For each feature, the major peak in the distribution is identified and only nuclei in its FWHM are retained. If no selection features are specified, the nuclei are not filterd. Up to three (3) selection features are currently allowed. Ultimately, this allows to extract the most homogeneous nuclear population from the current sample. Available features for nuclear selection are:-
size
for the volume in voxels (if--seg-type
was not3d
, this corresponds to theflat_size
feature). -
surf
for the nuclear surface (available only with--seg-type 3D
). -
shape
for the nuclear circularity (if not--seg-type 3d
) or sphearicity (if--seg-type 3d
). -
sumI
for the DNA stain intensity integral over the nuclear volume/area. -
meanI
for the DNA stain intensity average (per voxel) over the nuclear volume/area. -
flat_size
for the nuclear area in the Z-projection. -
flat_sumI
for the DNA stain intensity integral over the nuclear area in the sum Z-projection.
-
-
--no-hole-filling
to skip the hole filling step during binarization (segmentation). -
--center-as-percentile
to define the nuclear center as the set of voxels with the top percentile lamin distance. By default, the voxel(s) with the maximum lamin distance are used as center instead. -
-r
to re-scale images if deconvolved with the Huygens suite. -
-n
to normalize the distances for inter-nuclear comparison.
-
-m
to provide the path to a folder with the previously generated masks. The masks are generated and saved to the provided folder if not found. -
-M
for the prefix of the masks. -
-2
for a folder containing (manual) 2D segmented images, that are enforced on the automatic 3D segmentation. Use--labeled
if the 2D segmented images are labeled. -
--labeled
whether to generate/expect binary or labeled masks. -
--compressed
whether to generate masks in compressed tiff format. The compressed format is not comaptible with FIJI, usetiffcu
to un/compress them if needed.
-
-l
to specify a different log file path. Defaults tooutputDirectory/log
. -
-U
for the unit of measurement of the voxel aspect data (nm
by default). -
-z
for the minimum number of slices occupied by an object to be considered a nucleus. If the provided value is >1, then it is interpreted as the number of slices. If the value is instead in the [0; 1] interval, it is then interpreted as the fraction of the Z-stack (in number of slices) that need to be occupied. -
-R
for the minimum radius (in vx) of an object. Objects with a volume lower than a sphere of the specified radius are discarded during segmentation. Default to 10 vx. -
--sigma-density
for the smoothing parameter applied when building the density profile of nuclear feature during the filtering step (see--nuclear-sel
). -
--sigma-smooth
for the smoothing parameter applied to the profiles. -
--description
to provide a readable lable for each condition. It expects a space separated list ofcondition:description
strings, wherecondition
is the name of a condition folder anddescription
is a descriptive labels used in plots instead of the default folder names. Use -- after the last one if not followed by another flag. -
--note
to provide a dataset/analysis description that is reported in the final report. Use double quotes. -
--nbins
for the number of bins (sampling) used when building the profiles. -
--version
to check the installed version of the script.
-
--skip
can be used to skip any of the pipeline steps, if previously run.-
inst
does not scan the input directory, and tries to use the previous scan results. This step is required to identify available conditions and images. -
seg
to skip the segmentation part. -
an
to skip the nuclear analysis part, where the single-voxel data are extracted to produce single-nucleus profiles. -
box
to skip the generation of multi-condition nuclear feature boxplots. -
plot
to skip the final plot step (i.e., gradients and profiles). -
report
to skip the final pdf report generation.
-
-
-t
to specify how many threads to use for parallelization. It is limited to the maximum number of threads detected by themultiprocessing
package. -
--regexp
to provide a regular expression used to identify the tiff images. -
-u
to trigger debugging mode, which saves a number of intermediate results.
pygpseq v3.3.4
is published under the MIT License - Copyright (c) 2017-18 Gabriele Girelli
- Introduction
- Installation
- GPSeq image analysis
- GPSeq FISH analysis
- TIFF tools
- F.A.Q.
- Contributing