-
Notifications
You must be signed in to change notification settings - Fork 0
/
Copy pathkoopa.cfg
141 lines (132 loc) · 4.95 KB
/
koopa.cfg
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
[General]
# path to raw input image (nd/stk or czi) files. paths should be given absolutely to
# improve safety. make sure to use the server based path style (/tungstenfs/...).
input_path = ./
# path where analysis results are saved (created if not given).
output_path = ./
# file extension of raw files (nd or czi).
file_ext = nd
# do 3d analysis?
do_3d = False
# do timeseries analysis? ignored if do_3d is set to true.
do_timeseries = False
[PreprocessingAlignment]
# align images?
alignment_enabled = False
# path to bead alignment files.
alignment_path = ./
# strategy for image registration [options: pystackreg, deepblink].
alignment_method = deepblink
# path to deepblink model used for alignment.
alignment_model = ./model.h5
# channel index for reference channel to stay static.
channel_reference = 0
# channel index for alignment channel to be transformed.
channel_transform = 1
[PreprocessingNormalization]
# registration method. ignored if do_3d or do_timeseries is set to true. [options:
# maximum, mean, sharpest]
registration_method = maximum
# frame to start analysis. only if do_3d or do_timeseries is true.
frame_start = 0
# frame to end analysis. only if do_3d or do_timeseries is true.
frame_end = 0
# position to start crop x and y.
crop_start = 0
# position to end crop x and y.
crop_end = 0
# mapping of axes to bin-scale.
bin_axes = []
[SpotsDetection]
# list of channel indices to detect spots.
detect_channels = []
# list of models to use for spot detection. will use the same order as the channels
# provided above. must be passed using quotes (["", ...]). there should be one model
# for every channel listed in `detect_channels`.
detect_models = []
# radius around spots for characterization.
refinement_radius = 3
[SpotsTracking]
# maximum number of frames to skip in tracks. set to 0 if do_3d is set to true!
gap_frames = 3
# minimum track length. if do_3d is set to true, shorter tracks will be removed!
min_length = 5
# pixel search range between spots in tracks/stacks. calculated by euclidean distance.
search_range = 5
# if ensemble drift xy(t) should be subtracted. only available if do_timeseries is set
# to true.
subtract_drift = False
[SpotsColocalization]
# do colocalization analysis?
coloc_enabled = False
# list of channel index-pairs for colocalization. in format of [[reference, transform],
# ...]. must contain values from channels in spotsdetection.
coloc_channels = [()]
# relative z-distance given x/y-distances are set to 1. only if do_3d is set to true.
z_distance = 1
# maximum distance for colocalization.
distance_cutoff = 5
# minimum number of frames for colocalization.
min_frames = 3
[SegmentationCells]
# which option for cellular segmentation should be done. [options: nuclei, cyto, both]
selection = both
# method for nuclear segmentation. [options: cellpose, otsu]
method_nuclei = otsu
# method for cytoplasmic segmentation. will only be used for the cytoplasmic part of
# selection `both`. [options: otsu, li, triangle]
method_cyto = triangle
# channel index (0-indexed).
channel_nuclei = 0
channel_cyto = 0
# paths to custom cellpose models.
cellpose_models = []
# expected cellular diameter. only if method is cellpose.
cellpose_diameter = 150
# if segmap should be resampled (slower, more accurate). only if method is cellpose.
cellpose_resample = True
# sigma for gaussian filter before thresholding. only if method nuclei is otsu.
gaussian = 3
# upper clip limit before thresholding to remove effect of outliers.
upper_clip = 0.95
# minimum object size - to filter out possible debris. only if method is otsu.
min_size_nuclei = 5000
# minimum object size - to filter out possible debris. don't set too large if medium-
# sized blobs should be kept. only if method is otsu.
min_size_cyto = 5000
# minimum radial distance to separate potentially merged nuclei (between centers). only
# if method is otsu.
min_distance = 50
# should segmentation maps touching the border be removed? currently only implemented
# if selection is both!
remove_border = True
# add a column where the distance of each spot to the perifery is measured.
border_distance = False
[SegmentationOther]
# enable other channel segmentation?
sego_enabled = False
# list of channel indices.
sego_channels = []
# list of methods. must be passed using quotes (["", ...]). [options: deep, otsu, li,
# multiotsu].
sego_methods = []
# list of models. must match the order of the methods above. set to none if using a
# mixture. only if method is deep.
sego_models = []
# list of segmentation_model backbones. only if method is deep.
sego_backbones = []
[FlyBrainCells]
# drosophila fly brain segmentation.
brains_enabled = False
# index for nucleus channel.
brains_channel = 0
# cellpose model batch size (default 8 too large).
batch_size = 4
# minimum mean pixel intensity of nuclei to not get filtered.
min_intensity = 0
# minimum area in pixels of nuclei (anything below filtered).
min_area = 200
# maximum area in pixels of nuclei (anything above filtered).
max_area = 8000
# dilation radius for cell segmentation.
dilation = 3